key: cord-0005560-rki4yrus authors: nan title: Physicians Poster Sessions: Poster Session / Day 1 date: 2016-03-21 journal: Bone Marrow Transplant DOI: 10.1038/bmt.2016.48 sha: 53e3ada6e125f35ee678f054a59a0806235a5d9b doc_id: 5560 cord_uid: rki4yrus nan The laboratory data collected in our institution showed that processing of marrow cells in a Cobe Spectra separator ends in obtaining a mononuclear population enriched in hematopoietic, mesenchymal/stromal and endothelial stem cells. All these stem cells may regenerate damaged tissues according to their biologic potential. The marrow cells but not those from mobilized blood seem to be appropriate to use in regeneration of damaged tissues assuming that the mobilization process reduces the ligand-receptor interaction important for trafficking of stem cellsrelease and homing. Material (or patients) and methods: The program of using marrow-derived mononuclear cells for intra-muscular implantation in regeneration of tissue damaged due to critical limb ischemia (CLI) started in the year 2003. From that time until 2005 twelve patients with CLI (an experimental cohort to investigate the tolerability and the presence of wanted and unwanted symptoms after implantation) and in the period 2009-2012 another 12 patients entered the study. All patients suffered from ischemia (Fontaine IV) due to atherosclerosis and all appropriate surgical procedures were exaggerated prior to the cellular therapy. The procedure of cellular therapy was as follows: (i) bone marrow harvest from posterior iliac crest in a volume of 500 mL, (ii) mononuclear cells enrichment using a Cobe Spectra separator which ended with a volume of 100 mL, (iii) 70 mL of the cell suspension were injected in 0.5 mL portions into calf muscles of the affected limb. The analysis of the implanted cells revealed the presence of (mean +/-SEM): (i) hematopoietic progenitors: CD34+, CD45+: 1.29% +/-0.19, (ii) CD45-CD34-cells were CD90+: 0.1% +/-0.02, CD105+: 2.4% +/-0.33, CD73+: 0.08% +/-0.01, and 24.4 +/ -(iii) CFU-F 4.5 /10-6 WBC. Results: Seven patients had the posterior tibia muscle biopsied for immunohistochemistry and genetic study. Of note, immunochemistry revealed the presence of HIF-1 and Hsp 90, which was accompanied by higher expression of genes encoding these proteins. Out of 25 legs being implanted (24 patients were enrolled in the study) 7 had the leg amputated due to the lack of responsiveness to the therapy or relapse. Four-year observation was conducted in 17 and 10-year or longer observation in 5 instances of cellular therapy interventions. At the 1, 3 and 12 months checkups, pain reduction was documented in the first (and second) cohort in 83% (88%), 67% (57%) and 28% (38%) respectively. The same proportions for wound healing were as follows: 25% (38%), 42% (43%), 50% (33%). Distance to claudication increased from 50 to 200 m (median) 12 months after the procedure, and the median of the distance to claudication was higher than that prior to the treatment at 4, 6 and 10 years after cellular therapy implementation. Progression-free survival (median 26 months) was valid for 42% of patients receiving cellular therapy. Conclusion: In conclusion, the intra-muscular implantation of a marrow cell population enriched with a mononuclear cell fraction in patients with CLI may result in a years long decrease in the ischemic symptoms. Disclosure of Interest: Supported by INNOMED/I/1/NCBR/ 2014 CellsTherapy NCBiR grant. None declared. Inhibition of Akt-signaling promotes the generation of superior tumor-reactive T cells for immunotherapy post allogeneic stem cell transplantation Introduction: Allogeneic stem cell transplantation is a potential curative treatment for patients suffering from a haematological malignancy. Nevertheless, relapse is often observed, indicating the need for additional therapies. T cell therapy targeting only tumor cells would specifically enhance the graft-versus-tumor effect without increasing the risk for graft-versus-host disease. Moreover, effective T cell therapy against cancer is dependent on long-lived stem cell-like T cells with the ability to self-renew and differentiate into potent effector cells. However, current ex vivo-generation protocols result in terminally differentiated effector T cells. Here, we investigated whether early memory tumorreactive T cells could be generated ex vivo, by interfering with the Akt-signaling pathway, involved in T cell differentiation. Material (or patients) and methods: To expand Minor Histocompatibility Antigen (MiHA)-specific T cells from the naive repertoire, MiHA-negative CD8 + T cells were stimulated with autologous monocyte-derived dendritic cells loaded with MiHA-peptide in the presence or absence of Aktinhibitor VIII. MiHA-specific T cells were analyzed in vitro and in vivo. Following this proof of concept study, Akt-inhibitors which are currently tested in clinical studies, were evaluated for their potential to inhibit T cell differentiation. Results: We found that by inhibiting Akt-signaling, MiHAspecific CD8 + T cells displayed an early memory-like phenotype (CCR7 + CD62L + CD45RO + CD27 + CD28 + CD95 + ) and transcriptome signature (TCF7 hi FOXO1 hi EOMES low KLRG1 low ). This resulted in a MiHA-specific CD8 + T cell population containing a high proportion of stem cell-like T cells, compared to terminal differentiated effector T cells in control cultures. Importantly, these Akt-inhibited MiHA-specific CD8 + T cells showed a superior expansion capacity both in vitro as well as upon infusion in immune deficient NSG mice. Moreover, adoptive immunotherapy of Akt-inhibited MiHA-specific CD8 + T cells resulted in a superior anti-tumor effect in intrafemural human multiple myeloma-bearing mice (Figure) . Finally, in mixed lymphocyte reactions Akt-inhibitors AZD5363, GSK2141795 Introduction: The risk of malignant transformation of ex-vivo expanded human mesenchymal stromal cells (huMSCs) has been debated in the last years (1, 2) ; however, the biosafety of these cells after exposure to supramaximal physical and chemical stress has never been systematically investigated. We established an experimental in vitro model to induce supramaximal physical (ionizing radiation, IR) and chemical (starvation) stress on ex-vivo expanded bone marrow (BM)derived huMSCs and investigated their propensity to undergo malignant transformation. To this aim, we examined MSC morphology, proliferative capacity (in terms of population doublings), immunephenotype, differentiation potential, immunomodulatory properties (PHA-induced T-cell proliferation assay) and genetic profile (by conventional karyotype and array-CGH) before and after exposure to stressors. Material (or patients) and methods: MSCs were isolated from 20 healthy donors (HDs; median age: 16 years; range: and expanded in culture medium supplemented with 5% platelet lysate (PL) up to passage 2. Thereafter MSCs were exposed both to escalating doses of IR (30, 100, 200 Gy) and to starvation culture conditions (1% PL instead of 5%). Results: When exposed to escalating doses of radiations, MSCs lose their typical spindle-shaped morphology, look progressively more exhausted and their growth rate markedly slows down and eventually stops (at P4-P6) by reaching early senescence (see figure 1 ). Nonetheless, in the presence of 1% PL, the effects on morphology are less evident, thus suggesting that the lack of PL-derived growth factors may slow down the senescent process induced by ionizing radiations. Despite these morphology changes which typically reflect a process of early senescence, the outgrowth of clones with uncontrolled expansion rate was never observed in the cultures, even after long-term observation of the cells following stress induction (up to 8 weeks). Irradiated and starved MSCs maintain the typical immunophenotype (positivity for CD105, CD90, CD13; negativity for CD34, CD45, CD14), capacity to differentiate into adipocytes and osteoblasts (although this latter was reduced) and to inhibit PHAinduced T-cell proliferation. The study of the genetic profile (conventional karyotype and array-CGH) did not show any chromosomal alteration in stressed MSCs. Conclusion: Our data indicate that irradiated and starved MSCs, although presenting altered morphology and growth rate, maintain the typical characteristics of HD-MSCs and do not display an increased propensity for malignant transformation. These data are in support of the biosafety profile of ex vivo PL-expanded huMSCs for clinical application. Comparison of Lovo and Sepax 2 for post-thaw DMSO depletion from cryopreserved stem cells B. Calmels 1,* , G. Bouchet 1 , J. Couquiaud 1 , L. Regimbaud 1 , C. Lemarie 1 , P. Houze 2 , M. Mercier 3 , C. Chabannon 1 1 centre de therapie cellulaire, institut paoli-calmettes, marseille, 2 laboratoire de biochimie, hopital saint-louis, paris, 3 Introduction: Cryopreserved stem cell grafts are still widely infused to many patients, both in the autologous or allogeneic settings. Cryopreserved grafts can be thawed at the bedside or thawed and washed at the cell therapy laboratory. We recently published that post-thaw washing did not impair hematopoietic engraftment, in a cohort of 2,930 autologous transplanted patients receiving either unwashed or washed grafts (Calmels B et al, Bone Marrow Transplant. 2014) . Postthaw washing can be implemented using various methods such as manual centrifugation, automated centrifuge-based (Sepax 2, Biosafe) or spinning-membrane devices such as the Lovo (Fresenius Kabi). Material (or patients) and methods: We here report a comparative study of Lovo versus Sepax 2 for washing thawed stem cell grafts. We took advantage of 11 apheresis products intended for destruction and cryopreserved in 2 identical bags; after dry-thawing (PlasmaTherm, Barkey), bags were connected to the Sepax 2 or to the Lovo, diluted volume to volume with +4-8°C 6% hydroxyethylstarch 130/0.4 (Voluven, Fresenius Kabi) and processed using the SmartWash program (Sepax 2) or a 2-cycles standard wash protocol on Lovo (a cycle referring to one pass through the spinning membrane). The Lovo settings were customized for this application: reduction retentate pump rate 75 ml/min, desired inlet PCV 10%, and automated volume to volume dilution. Results: After processing, CD34 and CD45 absolute counts and viability were evaluated (Stem-Kit, Beckman Coulter) and DMSO was quantified by capillary zone electrophoresis (P/ACE, Beckman Coulter). Post-wash data show comparable CD34 recovery, viability and effective DMSO depletion. Conclusion: We conclude that Lovo enables high efficiency DMSO depletion while preserving optimal CD34 viability and recovery. Comparison with the Sepax 2, a widely used automated centrifuge-based device, reveal comparable efficiency. Moreover, the short length of the procedure when using the Lovo (less than 15 min) does not significantly delay the process as compared to bedside thawing. Post-thaw washing using automated cell processing systems have thus to be preferred over bedside thawing since they provide multiple benefits such as DMSO and cell debris depletion, precise determination of infused CD34 cell dose, increased cellular stability as well as adjustable washing efficiency and final volume. Disclosure of Interest: None declared. and CD4+ T cells (HXTCs) could be expanded from patients on antiretrovirals (ARVs) as well as HIV negative individuals to effectively target HIV infection using a non-HLA restricted approach for patients receiving an autologous or allogeneic HSCT for HIV-associated hematologic diseases. Material (or patients) and methods: PBMCs from healthy donors or HIV+ patients were stimulated with gag, pol, and nef peptide libraries (pepmixes) in the presence of co-stimulatory and growth factors. HXTCs expanded to clinically relevant numbers (Mean = 1.62e8 cells, Range (3.72e7, 2. 87e8 cells), n = 16) even in the presence of ARVs. Results: 5 of 7 patient HXTCs showed specific activity to all 3 HIV antigens in IFNg ELISPOT assays. HXTCs were broadly specific to gag (median = 99.33 SFC/10e5 cells), pol (median = 131.11 SFC/10e5 cells) and nef (median = 337.26 SFC/10e5 cells). HXTCs were also expanded from 9 healthy (HIV negative) donors. HXTCs released IFNγ in response to gag (median = 163.79 SFC/1e5 cells, n = 9) and nef (median = 291.25 SFC/1e5 cells, n = 6) but not an irrelevant antigen (median = 7.0 SFC/1e5 cells). Importantly, HXTCs expanded from both HIV+ and HIVneg individuals were cytotoxic, as HXTCs lysed HIV antigen loaded autologous PHA blasts (mean = 67.55% specific lysis at 10:1 effector:target ratio) but not PHA blasts alone (mean = 0.46% specific lysis at 10:1 effector target ratio). HXTCs from HIV+ patients also showed a greater ability to suppress HIV outgrowth in vitro compared to unexpanded CD8+ T cells when co-cultured with autologous, reactivated resting CD4+ T cells, the authentic latent reservoirs ( p40.006 by Mann Whitney). Similarly, HXTCs derived from HIVneg donors were able to suppress HIV replication more than non-specific CD8+ T-cells when co-cultured with autologous CD4+ T cells infected with HIVSF162 (HXTCs blunted viral replication more effectively and demonstrated a mean 78.62% viral suppression compared to 34.19% suppression by unexpanded CD8+ T-cells and CMV-and EBV-specific T-cells derived from the same healthy donors. Finally, we have infused autologous HXTCs to 2 patients on ARVs and evaluated the total functional antiviral capacity of participant T cell responses against HIV prior to and following infusion, as measured by a novel viral inhibition assay using autologous reservoir virus and a latency clearance assay. Conclusion: In summary, we can ex vivo expand HIV-specific T cells (HXTCs) from HIVpos and HIVneg donors for clinical use after autologous or allogeneic HSCT. Infusion of 2 patients with autologous HXTCs we demonstrated safety and a marked and sustained increase in the antiviral capacity of patient CD8 + T cells against cells infected with the virus derived from the patient's own latent reservoir. Given the low frequency of circulating HXTCs expected weeks after infusion, this encouraging data suggests that HXTCs expand in vivo and may have substantial specific effect on rare viruses emerging from latency. Disclosure of Interest: None declared. Comparison of Optia MNC and CMNC systems for mononuclear cell collection in patients undergoing extracorporeal photopheresis: a prospective cross-over equivalence study (NCT 02490163) C. Del Fante 1,* , L. Scudeller 2 , A. Martinasso 1 , G. Viarengo 1 , C. Perotti 1 1 Fondazione IRCCS Policlinico S.Matteo, Immunohaematology and Transfusion Service-Centre for transplant immunology, 2 Fondazione IRCCS Policlinico S.Matteo, Scientific Direction, Pavia, Italy Introduction: Extracorporeal photopheresis (ECP) is an effective cell therapy employed in several diseases, including graft versus host disease (GvHD) and organ rejection. Off-line technique implies: 1) MNCs collection by leukapheresis; 2) MNCs transfer in a UV-A-permeable bag dilution with a saline solution, addition of 8-methosxypsoralen; and 3) finally, UV-A irradiation of the product (2 J/cm2) and immediate reinfusion to the patient. We report the results of the first prospective cross-over study aimed at comparing yield and quality of MNCs collected from patients undergoing ECP with two different automated systems: MNC and CMNC (working with intermittent and continuous flow, respectively), both released by Terumo BCT. The primary objective of the study was to assess the equivalence of CMNC and MNC systems in term of concentration measured as number of MNCs/ml contained in the collection bag. Secondary objectives were to assess equivalence of CMNC and MNC systems in terms of MNCs collection efficiency and purity. Furthermore platelet and red blood cell bag contamination, bag volume and procedure time were evaluated. Material (or patients) and methods: Fifty one patients (15 males and 36 females) affected with GvHD or Chronic Lung Allograft Dysfunction (CLAD) were consecutively enrolled and randomly assigned to MNCs collection alternatively by CMNC or MNC system between May and August 2015. The same patient was randomized to both devices within each ECP cycle in two consecutive cycles. Each ECP cycle consisted of 2 procedures performed within 24-48-72 hours (in any case, always within a week). Cycles were performed 1-2-4 weeks apart, according to patients' clinical condition. ECP procedures were performed using the off-line technique, according to our internal protocol, processing 1.5 blood volumes. Any related side effect was registered. Results: Results about 204 procedures are detailed in Table 2 . No equivalence was found between CMNC and MNC systems in terms of concentration (MNCs/ml) in collection bag: MNC system showed a higher concentration than CMNC ( p40.001). On the other hand collection efficiency was higher using CMNC system ( p40.001). HCT was higher for MNC system ( p40.001) and procedure time was equivalent for both systems. No side effects were recorded with both systems. 2008]. While reproducibly safe, the efficacy of MSC in this setting remains controversial and may depend on several factors, including time to intervention. Here, we describe a single center's experience on early MSC treatment in patients with SR-aGVHD. Material (or patients) and methods: We report 33 patients consecutively treated with MSC for grade II-IV SRaGVHD(22 men, 11 women; median age 46, range 18-61; 22 cutaneous, 25 gastrointestinal, 8hepatic; 17 grade II, 9 grade III, 7 grade IV) following allogeneic HCT (14 matched related, 10unrelated, 9 cord blood) for various indications (14 AML/MDS, 3 lymphomas, 3 myelomas, 2 CML,11 others). Third-party allogeneic donor MSC were isolated and expanded from bone marrow as previously described [Gonzalo-Daganzo R, et al Cytotherapy 2009 ], using platelet lysate containing medium since 2011, and administered in early (1) (2) or late (3) (4) passage in 55% and 45% of patients, respectively. A median of 4 doses of MSC (range 1-16; median dose 1.06x106/kg, range 0.66 to 1.76) were administered twice-weekly starting at a median of 10 days from the diagnosis of aGVHD (IQR: . Results: Response to MSC was evaluable in 32 patients. Sixteen patients (50%) showed a complete response (CR) 90 days after the first cell dose, and 8 patients each (25%) had a partial response (PR) or did not respond to MSC (NR). The median time to response after the first dose of MSC was 14 days (IQR: [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] . The rates of CR were 65%, 50% and 14% in patients with grades II, III and IV SR-aGVHD, respectively (P = 0.08). Patients with SR-aGVHD achieving a CR after MSC treatment had an improved median overall survival of 43.2 months (80% at 1 year), compared with 3.2 months in cases with PR/NR ( p40.001) ( Figure 1 ). Gastrointestinal SR-aGVHD involvement associated with a lower rate of CR (40% vs 86%; P = 0.03) and a trend towards a poorer median overall survival (2.9 vs 30.9 months; P = 0.089). Only 2 patients treated with MSC in this series (6%) had disease relapse. There were no adverse events related to the MSC infusion. Conclusion: Allogeneic third-party MSC are a safe therapeutic option for patients with SRaGVHD. Our single center experience shows that in addition to their safety, early intervention withMSC may lead to an overall response in 75% of patients, with a CR rate of 50%, which associates with an improved overall survival. Gastrointestinal involvement may associate with a poorer response to MSC and outcome. Future prospective studies to analyze the role of MSC in SRaGVHD are warranted. Disclosure of Interest: None declared. Introduction: Mesenchymal stromal cells (MSCs) are adult, fibroblast-like multipotent cells that can be expanded from a variety of human tissues, including bone marrow (BM). MSCs have been employed in clinical trials in the context of hematopoietic stem cell transplantation (HSCT) to facilitate engraftment of hematopoietic stem cells (HSCs) and to prevent graft failure (1, 2) . Hematopoietic Stem Cells (HSC)-based gene therapy (GT) has provided therapeutic benefit in primary immunodeficiencies, thalassemia and leukodystrophies. Technologies for targeted editing of genes through artificial endonucleases are being developed; however, limited numbers of gene-edited HSCs become available by employing current tecniques (3) . We evaluated the possibility to isolate and expand ex vivo MSCs from the CD34 negative fraction (CD34-MSCs) of BM, as alternative autologous source of MSCs for application in GT strategies. Material (or patients) and methods: Mononuclear cells (MNCs) were isolated from 10 healthy donor (age range 20-35 years) BM samples (4-5 ml) by density gradient centrifugation on Ficoll. CD34+ cells were purified from MNCs by immunomagnetic depletion with anti-CD34 microbeads. The remaining CD34fraction was plated in the presence of DMEM Low-Glucose with 5% platelet lysate to isolate and expand MSCs according to previously reported methods (4) . Morphology, clonogenic potential (CFU-F), proliferative capacity (Population Doubling, PD), immunophenotype, differentiation capability (by immunohistochemical staining), senescence and immunomodulatory properties (by inhibition of PHA-induced PBMC proliferation) of CD34-MSCs were evaluated. Results were compared with those of MSCs obtained by plating BM-derived total mononuclear cells (defined here as standard MSCs). Results: CD34-MSCs were successfully isolated from 10/10 BM samples. CD34-MSCs were comparable with standard MSCs in terms of morphology (spindle-shaped) and immunophenotype (positivity for CD105, CD90, CD13; negativity for CD34, CD45, CD14). While CD34-MSCs displayed a slightly reduced clonogenic capacity at 7 days of culture as compared with standard MSCs (CD34-MSCs: 0,24 ± 0,24; standard MSCs: 0,45 ± 0,30; mean ± standard deviation), they showed similar proliferative capacity in terms of population doublings. CD34and standard MSCs were similarly able to differentiate into adipocytes and osteoblasts and showed a superimposible dose-dependent capacity to inhibit lymphocyte proliferation after PHA stimulation at MSC:PBMC ratio 1:2 and 1:10. Experiments on MSC long-term culture and senescence are in progress. Conclusion: Our data indicate that CD34-fraction, obtained after immunomagnetic depletion, may be considered a rich source of MSCs. This method could allow to obtain sufficient numbers of autologous CD34+ cells and MSCs for clinical application in GT strategies. In particular, it would permit the co-infusion of MSCs with the aim to facilitate engraftment and favor chimerism for transduced cells in those GT protocols in which low numbers of gene-modified cells are infused (i.e. gene-editing Introduction: Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder affecting upper and lower motor neurons that causes paralysis of almost all skeletal muscles. Despite advances in medical sciences, no cure for ALS has been found. Riluzol is the only drug with proven efficacy, however with modest benefit on survival. First promising results of mesenchymal stem cell (MSC) therapy in ALS have already been published worldwide. The mechanisms behind MSC action are not entirely known, they probably enhance of neuroprotection by producing soluble factors that modulate biological functions of local glial cells, change response to inflammation and up-regulate regulatory T-cell activity. Polish Stem Cell Bank (PBKM) has provided Wharton's jelly-derived MSC (WJ-MSC) for clinical application in a medical therapeutic experiment for patients with ALS. Material (or patients) and methods: We have administered WJ-MSC from third party donors after Bioethical Committee approval to 17 patients (pts) with ALS aged from 25 years to 67 years (median age of 59 years). Nine pts have received infusions intrathecally (i.t.), 1 pt intravenously (i.v.) and 7 pts via both routes (first i.v., next i.t.). The cells were previously collected from healthy newborns, processed, screened for bacterial contamination as well as endotoxin content, and frozen in liquid nitrogen. MSC immunophenotype was confirmed using flow cytometry analysis assay. The patients have received from 1 to 7 infusions (median: 2.0) with intervals of 4-8 weeks. Intrathecal dose was 30x10 6 cells per infusion, while median i.v. cell dose equaled 1.17x10 6 /kg body weight per infusion. Each patient has been always examined by the same neurologist at the day of each infusion and result of examination has been described in a follow-up form. Results: Seven patients had positive changes in neurological examination (i.a. improved muscle tension, greater muscle strength). Clinical condition of 3 pts deteriorated which could be attributed either to disease progression or stem cell therapy. One patient withdrew from the treatment and one died due to disease progression. Six patients remained stable, no changes in their neurological status have been noticed. Minor adverse events, including neck stiffness, headache, subfebrile state or nausea have been observed. The majority of those side effects were observed shortly after the infusion. One patient was lost to follow-up. Introduction: Infusion of freshly isolated or ex vivo expanded donor regulatory T cells (Tregs) can prevent Graft Versus Host Disease (GVHD) after hemopoietic stem cell transplantation (HSCT). Before using this approach in clinical protocol, optimal strategies and conditions for Treg GMP expansion must be validated. Here, we aimed to establish an efficient ex vivo clinical grade GMP protocol for isolation and expansion of Tregs, producing cells with stable phenotype and function after 3 weeks expansion and cryopreservation/thawing. Material (or patients) and methods: Isolation: Tregs were purified from apheresis of normal donors (N = 2) by an immunomagnetic method (CD8/CD19 depletion and CD25 enrichment) using the CliniMACS system (Miltenyi Biotech). Expansion: Tregs were purified from standard buffy coats with immunomagnetic isolation (CD8 depletion and CD25 enrichment) using the MACS system (Miltenyi Biotech). After isolation, cells were cultured in wells (N = 5) and in bags (N = 2) with anti-CD3/CD28 coated-beads in the presence of Rapamycin and IL-2 (Miltenyi Biotech). Cells were restimulated every 7 days for 21 days. Phenotypical and functional characterization of expanded Tregs was performed at specific time points and after thawing. Results: Isolation: We started from 2 apheresis, containing 363.78x10 6 and 401.01x10 6 CD4 + CD25 + cells, respectively. The absolute number of freshly isolated CD4 + CD25 + T cells was 101.88x10 6 and 93.52x10 6 , with a recovery of 28% and 23.32% respectively. The mean percentages of CD4 + CD25 + and CD4 + CD25 + FoxP3 + cells were 80% and 66%. Expansion: We started from 60 mL of buffy coat. After isolation, the mean absolute number of freshly isolated CD4 + CD25 + T cells was 8.8x10 6 ± 6.7 and the mean percentage of CD4 + CD25 + FoxP3 + cells was 65 ± 10%. Cell contaminants included 1% CD8 + T cells, 5% Th17 cells, 20% CD19 + cells, 5% CD3 + / CD56 + cells and 5% CD14 + cells. Cell expansion in wells peaked between the second and third re-stimulation and after 21 days the median fold increase was 342 (range 180-826). After 21 days of ex vivo expansion, the cells were almost totally CD4 + CD25 + FoxP3 + (97%). Consistently, the percentages of Th17 cells, CD8 + T cells, CD14 + cells, B cells and T-NK cells progressively decreased, reaching values below 1% on day 21. The fold increases in bags were 82 and 26 on day 21. However, the phenotype was comparable to that in wells with 480% CD4 + CD25 + FoxP3 + cells and contaminant cells below 5% at the end of expansion. Moreover, xpanded Tregs, either in bags or in wells, efficiently reduced CD3/CD28 induced T effector cells (Teff) proliferation as compared with freshly isolated Tregs (80 vs 40% inhibition at a Tregs-Teff ratio 1:2). After thawing, the mean percentage of viable expanded Tregs, either in wells or in bags, was 75 ± 5% with 495% CD4 + CD25 + FoxP3 + cells. Thawed expanded Tregs also efficiently reduced Teff proliferation. Conclusion: We successfully set up an efficient and safe (low content of B and Th17 cells) ex vivo GMP-compliant isolation and expansion protocol. Moreover, we expanded cells up to 21 days and we obtained a high number of stable and functional Tregs. Interestingly, ex vivo expanded Tregs, either in bags or in wells, maintain highly suppressive capacity after cryopreservation/thawing. This finding would allow more flexibility in terms of timing and number of the infusions in Treg immunotherapy trials. Disclosure of Interest: None declared. Introduction: Cord blood transplantation (CBT) from a related family member is an effective therapy for patients with Sickle Cell Disease (SCD) resulting in encouraging outcomes with similar or superior survival to adult donor transplant. Efforts to implement family-directed umbilical cord blood (UCB) banking have been developed in the past two decades for siblings requiring stem cell transplantation (SCT). Public umbilical cord blood banks are faced with the challenge regarding the units to be stored or to be discarded or used for other endeavors such as research. To this end, we report our 20-year experience of family-directed CB banking for SCD and review the characteristics of the CB units collected between 1995 and 2014. Material (or patients) and methods: Families were eligible if they had a child with SCD, and expecting the birth of a sibling. Participation was voluntary and free of charge. All mothers underwent a panel of serologic blood donor screening assays. CB units were collected in remote sites, cryopreserved and stored in a single bank. CB testing included viral serology, bacterial cultures and cell counts. HLA typing on the CB were not routinely performed unless requested by the transplant physician. Results: 189 families from 27 different centers were enrolled. 17 (10%) families had 41 CB unit stored, due to multiple pregnancies. Potential recipients had a median age of 6 years (range 11months-12 years) at time of CB harvest. 14 families had 41 affected child. A total of 210 CBs were collected from 189 mothers. All CBs were negative for HIV and 64 CBs (30%) had positive anti-HBs and/or anti-HBc with negative HBsAg. Median CB volume at cryopreservation was 92.5 mL (range .Median total nucleated cell (TNC) count at cryopreservation was 9.2 x10 8 (range 1-75.3) . Median collected CD34+ and CFU-GM cell counts were 4.5 x10 6 (range 0.18-61) and 4.5 x10 5 (range 0.14-67) respectively. The hemoglobin status of CBs was assessed through the neonatal screening. Data were available for 179 CBs (85%) including 64 (30%) hemoglobin AA, 85 (40%) hemoglobin AS, 25 (14%) hemoglobin SS, 3 ( o1%) hemoglobin AC, 1 S-beta +-thal and 1 beta-thal. Eight (4%) out of the 210 banked CB units were released for CBT with a median TNC count of 7.0x10 8 (3.0x10 8 -21.8x10 8 ) . Five patients were transplanted using a single CB and 3 patients with the sibling's bone marrow and CB. Posttransplant data were available for 6 patients: all of them had stable engraftment of donor cells and are alive, free of SCD. Conclusion: Our data showed that family-directed CB banking is feasible and yields good quality cord blood units for sibling transplantation. However, the number of CBT performed is disappointing despite the good results of sibling transplantation in SCD. Therefore, we must think about the costeffectiveness of this approach when an HLA identical sibling donor is available. Disclosure of Interest: None declared. Material (or patients) and methods: In this single randomized research 22 patients at the median age of 13 years (range, 5-24) were included. The 10 patients received HSCT and cotransplantation of mesenchymal stem cells (ALL -8, AML -2) -"MSC+" group. The number of infused MSCs was 1,56 ± 0,4 × 10 6 /kg. MSCs were derived from bone marrow of donors. In the "MSC-" group were included 12 patients (ALL -7 patients, AML -4, MDS -1) with transplanted HSC without contransplantation of MSCs. Patients of both groups were identical in nosological forms of the disease, age, sex, type of donors, conditioning regiment, GVHD prophylaxis and graft composition. The evaluation of standard immunological parameters in peripheral blood of the patients were carried out on +30, +60, +100 +180 и +365 days after allogeneic HSCT by flow cytometry. Results: The patients from the "MSC+" group had tendency to recover leukocytes (P = 0,09), neutrophils (P = 0,09) and platelets (P = 0,05) faster than patients from the "MSC-" group. Median days of reconstitutions of leukocytes, neutrophils, platelets were 16 (12-30), 19 (12-20) , 18 , respectively, in the "МSC+" and 22 (14-45), 24 (16-45), 23 (16-144), respectively, in the "MSC-" group. The incidence of acute GVHD grades II -IV was 10% in the "МSC+" group (1 patient with grade II GVHD (skin only) of 10 patients) when in the "MSC-" group it was 25% (3 patients with grade III -IV GVHD of 12 children), (P = 0,36). The incidence of extensive form of chronic GVHD in the "MSC+" group was 10% (1 recipients of 10) when in the "MSC-" group it was diagnosed in 33% (4 of 12 patients), (P = 0,37). The "МSC+" and the "МSC-" cohorts showed similar 1-year cumulative incidence of infections. It was documented 8 episodes of virus diseases in the "МSC+" vs 7 episodes in the "МSC-"; 2 episodes of bacterial infections in the "МSC+" vs 5 episodes in the "МSC-"; 1 episode of fungal disease was diagnosed in the "МSC-" group. After a median follow up time of 38 months (range, 5,7-59,4 months) one patient of the "МSC+" group had relapse (at +150 day), whereas nobody relapsed in the "MSC-" group. Overall survival in the "MSC+" group was 90 ± 9,5% vs 83 ± 10,7% in the "МSC-". It was not revealed significant differences in immunological recovery of patients from the "MSC+" and "MSC-" groups. It was noted only a higher level of NK and B-cells at +100 day. Thus, absolute count of NK in the "MSC+" group was 0,1 (0,02-0,23) х10 9 /l, whereas in the "MSC-" group -0,01 (0, 06) х10 9 / l (р = 0,07). Absolute B-cells count at +100 day in the "MSC+" group was 0,18 (0, 43) х10 9 /l, whereas in the "MSC-" group -0,07 (0, 19) х10 9 /l (р = 0,05). Conclusion: In conclusion, co-transplantation HSC and MSC improves leukocytes, neutrophils and platelets engraftment with reducing risk of severe acute GVHD and extensive form of chronic GVHD without increasing rate of relapse and infection complications. Disclosure of Interest: None declared. Combining Natural Killer cell adoptive immunotherapy with DNA methyltransferase inhibition: a double attack against acute myeloid leukemia J. Cany 1,* , M. Roeven 2 , R. FrancoFernandez 1 , F. Maas 1 , G. Huls 2 , N. Schaap 2 , H. Dolstra 1 1 Laboratory Medicine, 2 Hematology, Radboudumc, Nijmegen, Netherlands Introduction: Adoptive transfer of NK cell is an emerging treatment approach against cancer including AML. We reported previously a GMP-compliant culture system for the generation of NK cells from CD34 + hematopoietic progenitor cells (HPC) with high cell numbers, purity and functionality, and the first phase I dose-escalation study with allogeneic HPC-NK cells has been completed recently. A next step in maximizing NK cell adoptive immunotherapy is the combination with anti-cancer drugs that exert a direct antileukemic effect as well as potentiate NK cell reactivity towards AML. The DNA methyltransferase inhibitors (DNMTi) Azacitidine and Decitabine demonstrated promising clinical activity in AML patients. The present study consists of a head-to-head comparison of these two DNMTi, to evaluate their impact on HPC-NK cells as well as their additive and/or synergistic anti-leukemic effect in combination with HPC-NK cell transfer. Material (or patients) and methods: The effect of DNMTi on AML and HPC-NK cells was examined in vitro using clinically-relevant low drug concentrations. Phenotypical analysis were performed using 9-color flow cytometry, and gene expression profiling using quantitative Taqman assay. To test the effect of DNMTi in vivo, HPC-NK cells were adoptively transferred into NSG mice in combination with IL-15 support. Anti-leukemic studies were conducted using luciferase-expressing AML cells implanted in the femur of the mice, and tumor load was monitored by bioluminescence imaging. Results: We first examined the effect of DNMTi on HPC-NK cells in vitro. Although a small but significant decrease in proliferation was noticed using higher dosage of the drugs, HPC-NK cells maintained high expression of activating receptors (NKG2D, NKp44, NKp46, DNAM-1), as well as TRAIL and CXCR3. Phenotypical differences following treatment with DNMTi were only observed in the percentage of KIRexpressing NK cells. However, functional assays revealed impaired cytolytic functions of NK cells following treatment with Azacitidine, while Decitabine did not affect their killing nor IFNγ production capacity. Next, using cell lines as well as primary blasts, we found that DNMTi and HPC-NK cells both exerted potent anti-leukemic activity in vitro. Although their effects were mostly additive, a synergistic effect was observed for some AML cells tested. This occurred with concomitant upregulation of NKG2D ligands on AML cells, indicating that DNMTi can sensitize AML cells to NK cell-mediated killing. Then, we investigated the effect of DNMTi on AML and HPC-NK cells in vivo. While both agents exerted a significant and dose-dependent effect on AML cell progression, the persistence of adoptively transferred NK cells was not affected upon treatment, with sustained expression of activating receptors, and remarkable KIR acquisition on NK cells compared to untreated mice. HPC-NK cells isolated from DNMTi-treated mice also showed higher NKp44 expression level. Most importantly, treatment of AML-bearing mice with HPC-NK cells in combination with Decitabine significantly reduced tumor progression compared to DNMTi alone. Conclusion: All together, these data indicate that HPC-NK cells and Decitabine can potently cooperate to combat AML, and provide a strong rationale to explore this combined treatment approach in patients. Disclosure of Interest: None declared. CD34 negative fractions of leukophereses collections permit generation of highly functional donor-derived CMV-specific cytotoxic T lymphocytes J. Montoro 1,* , M. Guerreiro 2 , R. J. O'Reilly 3 , G. Koehne 4 1 Hospital Universitario La Fe, Valencia, Spain, 2 São João Hospital, Porto, Portugal, 3 Memorial Sloan Ketteriing Cancer Center, 4 Memorial Sloan Kettering Cancer Center, New York, United States Introduction: Reactivation of CMV remains a major risk post HSCT, but can be effectively controlled by adoptive transfer of donor-derived CMV-specific T cells (CMV CTLs). CMV CTLs are usually generated from PBMCs from donors collected before mobilization with granulocyte colonystimulating factor (G-CSF). Although previous studies showed impairment of T cell function after mobilization with G-CSF, recent reports suggested that PBMCs exposed to G-CSF retain antiviral function and are suitable for generation of antigen-specific CTLs. In this study, we evaluated the feasibility of generating donor-derived CMV CTLs from the CD34 negative fraction of the leukapheresis obtained from G-CSF mobilized allogeneic donors of CD34 selected HSCT. Material (or patients) and methods: For the generation of CMV CTLs, PBMCs were isolated after Ficoll-Hypaque centrifugation with 10 ml taken from 8 separate unrelated donor CD34apheresis collections. 1x10 6 cells/mL of unmodified and cryopreserved PBMCs, were stimulated with 0.5x10 5 /mL 6000 CGy irradiated donor-derived dendritic cells (DCs) or 0.5x10 5 /mL irradiated donor-derived Epstein-Barr virus-transformed B lymphocyte cell lines (BLCLs), both pulsed with the pool of overlapping pentadecapeptides of CMVpp65. Cultures were weekly restimulated and IL-2 was added to the culture at day 8 and 2-3 times weekly thereafter. After 28 days, T cells were assessed by multi-parameter flow cytometry, MHC tetramer analyses and cytotoxicity assays. Results: We were able to generate and expand CMV CTLs from all 8 donor-derived CD34 neg specimens. In side-by-side comparison using CMVpp65 loaded BLCLs and DCs, CMVspecific T cell were expanded from 8/8 donors for the BLCL group, but only 5/8 in the DC group which may reflect the previously described impairment of DC function after G-CSF mobilization. While cultures from the BLCL sensitized T cells were predominantly CD8+ (95%) and were specific for CMV (64%) as well as EBV (24%) antigens, the DC sensitized cultures consisted of CD8+ (68%) and CD4+ (32%) with CMV specificity of 42% and 18%, respectively. CMV and EBV specific CD8+ T cells were multifunctional expressing high levels of CD107a, TNFa and IFN-ƴ. CMV-specific CD4+ T cells also produced IL-2 and up-regulated CD154, suggesting their potential to sustain T and B cell expansion. Degranulation observed by flow cytometry correlated with high levels of cytotoxicity, assessed by the standard 4 h 51 chromium assay, against antigen loaded DCs. No alloreactivity, NK cell expansion or CD4+CD25+CD127lowFOXP3+ Tregs were present at the end of any cultures. In the donors expressing HLA-A*0201 and B*0702, for which tetramers to NLV and TPR sequences of pp65-protein were available, we confirmed that high proportions of responses were directed to these epitopes. Introduction: Graft-versus-leukemia (GvL) relies on donor T-cells killing host leukemia cells post-transplant. T-cells with preferential recognition of leukemia cells have been well documented but their clinical relevance remains challenged because thymic selection prevents a high-affinity interaction between T-cells and antigen presenting cells (APC) expressing regular antigens in self-HLA. ATIR101, a personalized T-cell immunotherapeutic selectively depleted of HLA-haplotype mismatched T-cells, provides a unique platform to study leukemia-reactive T-cells as highaffinity interactions with antigens expressed in the mismatched haplotype may occur, whereas T-cells responding to the antigenpresenting foreign HLA-molecule have been eliminated. Material (or patients) and methods: Two out of the first 10 ATIR101 batches manufactured in clinical phase 2 study CR-AIR-007 (NCT01794299) met the requirement of having a mismatched haplotype known the be able to express a known leukemia-associated antigen; these batches were used to screen for the presence of leukemia-associated antigen reactive T-cells. Because the frequency of leukemia-associated antigen reactive T-cells is expected to be very low, we used peptide-MHC monomers of the mismatched HLA-haplotype presenting leukemia-associated antigens and established a stimulation platform with artificial APCs (aAPC). Those aAPC consisted of streptavidin coated microspheres loaded with a biotinylated anti-CD28 antibody and the respective biotinylated peptide HLA monomer. Results: In one of the two batches, leukemia-associated antigens specific T-cells were detected: ATIR101 cells were stimulated with Myb 628 /HLA-B44 aAPCs and Myb 628 specific T-cell expansion was assessed after one or two rounds of stimulation; an irrelevant HLA-B44 multimer was used as negative control. Clearly, we were able to detect CD8+ T-cells with specific reactivity against one HLA-B44-restricted leukemia-associated HLA ligand derived from the MYB gene (figure 1). Conclusion: These data show that T-cells recognizing leukemia-associated antigens expressed in the mismatched HLA-haplotype are retained in ATIR101 from which the T-cells responding to the antigen-presenting foreign HLA-molecule have been eliminated. Conceivably, these cells may contribute to the Graft-versus-leukemia (GvL) effect of ATIR101. Disclosure of Interest: None declared. Safety and efficacy of G-CSF mobilized granulocyte transfusions in the setting of allogeneic hematopoietic stem cell transplantation: a single centre experience K. Scholl 1,* , S. Introduction: Uncontrolled infections due to prolonged neutropenia, especially fungal infections, significantly increase non-relapse mortality (NRM) following allogeneic hematopoietic stem cell transplantation (HSCT) and are an absolute contraindication for allogeneic HSCT at our centre. G-CSF mobilized granulocyte transfusions (G-GTX) might provide sufficient neutrophils to control infection and thus provide a bridge toward HSCT and are potentially able to reduce early NRM. Objective: To evaluate the safety and efficacy of peritransplant G-GTX. Material (or patients) and methods: We performed a retrospective analysis of all adult patients (pts) who received G-GTX between 01/2005 and 07/2015 in the Clinic of Internal Medicine III at the Ulm University Hospital. G-GTX were prepared from unrelated AB0-compatible and anti-CMV negative donors. The indication for G-GTX was a diagnosed uncontrolled infection during prolonged neutropenia due to either delayed neutrophil recovery after treatment or refractory disease. The endpoints evaluated were efficacy of G-GTX as a bridge to HSCT, control of infection, NRM at day+30 and at day+100 as well as overall survival and cause of death. Results: During the evaluated period 30 pts received G-GTX. 25 pts were evaluable for the efficacy and safety of G-GTX. The cohort consisted of 19 males and 6 females, median age was 50 years (range 22 -60). The most common underlying disease was acute leukaemia (72%), 56% of pts had not reached a complete remission. The median EBMT score was 5. The median duration of grade 4 neutropenia prior to G-GTX was 31 days (range 0-92). The sites of infection were respiratory tract (68%), soft tissue (28%) and colon (4%). 56% of pts received G-GTX for fungal infections. 19 pts (76%) were treated with G-GTX pre-and post-transplant, 2 pts only pre-transplant and the remaining 4 pts post-transplant. Patients received a median of 11 G-GTX-products (range 2-34) usually every other day with no less than 1x10E10 granulocytes per m2 bsa and transfusion. The G-GTX were well tolerated, the only relevant adverse events recorded were temporary chills and febrile episodes in 28% of pts. The median time to neutrophil recovery after the first G-GTX was 24 days (range . Control of the infection was achieved in 24/25 pts (96%). Bridging to transplant was successful in 18/ 19 pts (94,7%). A complete remission of the underlying disease was achieved in 92% of pts, and we were able to discharge 92% of pts. NRM at day+30 was 0% and 8% at day+100, thus equivalent to transplant-patients not requiring G-GTX. Still the overall survival of this patient cohort has been poor. Only 6 pts are currently alive (with a median follow up of nearly 3 years, range 186-1682 days from G-GTX), 19 pts have died (with a median time after G-GTX of 6 month, range 47-698 days). The cause of death being either relapse (32%) or NRM (68%). All NRM deaths were due to infection with only 3 cases of recurrence of initial fungal infection. Conclusion: G-GTX are a safe and effective method to control otherwise uncontrollable infections. They are an effective bridge to transplant and maintained early NRM within the expected range for hematologic patients at high risk of disease associated early mortality. Disclosure of Interest: None declared. [P020] Introduction: Allogeneic stem cell transplantation (allo-SCT) is the only curative option for several hematological malignancies. Unfortunately, a sustained remission after allo-SCT is not often seen. Donor lymphocyte infusions (DLI) are frequently applied, but only effective in a minority of patients. Furthermore, DLI are non-specific and can induce severe graft-versus-host disease (GVHD). This indicates that novel strategies are needed to prevent or treat relapse after allo-SCT. We therefore aim at improving the graft-versus-tumor effect, without increasing the risk of GVHD, by targeting hematopoietic lineage-restricted and tumor-associated minor histocompatibility antigens (mHags) using peptide-loaded dendritic cell vaccinations. Material (or patients) and methods: In the present phase I/II study, we investigate the feasibility, safety and efficacy of this concept using nine different hematopoietic mHags (HA-1, HA-2, UTA2-1, LRH-1, ACC-1, ACC-2, PANE-1, HB-1, CD19 L ). We include patients with a donor mismatched for one or more hematopoiesis-restricted mHags. In case of persistent disease after DLI, patients are treated with a second, equivalent dose of DLI, combined with a therapeutic vaccine consisting of monocyte-derived, mHag peptide and KLH pulsed dendritic cells (DCs) of donor origin. Results: So far, nine multiple myeloma patients received vaccinations loaded with HA-1, UTA2-1 and/or LRH-1 peptide. Vaccinations were well tolerated, with only transient local symptoms at the injection site and mild fever. Six patients showed progressive disease after a median of 4.0 months (range [2] [3] [4] [5] [6] [7] [8] [9] [10] . Two patients show stable disease after a median follow up of 12 and 24 weeks respectively. One patient also received radiotherapy for a bone lesion during follow up and achieved molecular remission (based on plasma cell chimerism) and still remains in remission for 13 months now. Immune monitoring, thus far completed in 7 patients, revealed the induction of mHag-specific T-cells in the peripheral blood of 5/7 patients, detected by staining with HLA-mHag-peptide tetramers. In all 7 patients, we observed the induction of KLHspecific IFN-y production, proliferation and activation of, predominantly, CD4+ T-cells in peripheral blood. Introduction: High dose chemotherapy followed by autologous stem cell transplantation (autoSCT) is considered a standard of care for many hematological malignancies. The management of these patients requires a central venous catheter (CVC) for administration of high dose chemotherapy, parental nutrition, transfusions and antibiotics. However, the use of CVC carries risks that may be related to the positioning (eg. pneumothorax) and others (eg. infection and thrombosis) occurring later. The use of peripherally inserted catheters (PICCs) has increased in recent years, but there are few studies in transplanted patients. Objective: To evaluate the safety, ease of use and benefits of catheter peripherally inserted in autologous transplants in our center. Material (or patients) and methods: 52 autoSCT were performed since June 2013 to November 2015. In 37 of them a PICC was inserted by an specialized group of nurses in our hematolgy department. Most of the PICCs inserted in multiple myeloma were positioned just before the autoSCT, otherwise in patients with lymphoproliferative diseaset PICCs were inserted for salvage chemotherapy before transplant. No major complications were observed at the time of catheter insertion. As early complications during insertion of the catheter one local hematoma was observed due to accidental arterial puncture in one patient (2.7%), and two PICCs were repositioned because of primary malposition (5.4%). As late complications one accidental removal of the catheter was observed in one patient (2.7%), symptomatic pericatheter thrombosis in three patients (8.1%) , and catheter-related infection in five cases (13.5%) germ isolation was performed in four patients (10.8%), St. aureus (2 patients), St. epidermidis (2 patients) . Cellulite at insertion point was seen in one patient (2.7%). The main reason for removal of the catheter was conclusion of treatment in thirty-one patients (83.7%). Catheter related infection in four patients (10.8%), one death (2.7%) shortly after transplant due to progression of his hematologic disease, and one accidental (2.7%). None PICC was removed for thrombosis. Conclusion: Our data demonstrates that PICCs are safe and useful in the management of patients receiving autoSCT. PICCs are a good alternative to conventional central venous catheters in these patients. Introduction: The Society to Support Children Suffering from Cancer, also known as MAHAK, was set up in 1991 as a nongovernmental and non-profit organization. In the past two decades, the organization has attracted a vast public support and fulfilled a great part of its mission which is to support children with cancer, reduce the child mortality rate and create an appropriate environment that empowers families who have children with cancer. Pediatric Stem Cell Transplant also is used to treat many types of conditions affecting children and adolescent, including cancer and certain hematologic, immunologic and genetic disorders. The pediatric stem cell transplantation ward was inaugurated in Mahak hospital (Iran-Tehran) on April, 2012. Pediatric stem cell transplant ward practice that performs about 34 transplants per year. All patients are kept in high efficiency particulate air (HEPA) -filtered, positive-air-pressure sealed rooms, and strict hand hygiene is practiced. Material (or patients) and methods: We analyzed the outcome of 118 patients from a single institution who underwent allogeneic & autologous stem cell transplantation from between 2012 -2015. 106 of patients had peripheral blood stem cell as the stem cell source, eleven of patients' bone marrow and in 1 patient cord blood used. The majority of patients are: ALL = 30, Neuroblastoma = 27, AML = 12, Hodgkin's dis = 25, Retinoblastoma = 4, Ewing's sarcoma = 2, Rhabdomyosarcom = 2, Wilm's tumor = 4, Hepatoblastoma = 1, Aplastic Anemia = 3, Hemoglobinopathy = 2, Germ cell tumor = 4, Epedymoma = 1, Osteopetrosis = 1.The conditioning regimens used were mainly myeloablative in allogeneic transplantation. Age of patients 7 month to 26 years with median age 9.5 years, M/F = 67/51. 47 patients transplanted Allo HSCT and 71 patients transplanted Auto HSCT. GVHD prophylaxis regimen was cyclosporine+Mtx in Allo HSCT. All patients engrafted. The type of donor in allogeneic SCT includes 39 related sibling and 8 unrelated allogeneic. Results: In allogeneic PBSCT patients' median time to reach absolute neutrophil count (ANC) 40.5 × 10 9 /L was 12 days, and the median time to platelet count420 × 10 9 /L was 14 days vs 18 and 23 days in Allo BM patients. In autologous PBSCT median time to reach absolute neutrophil count40.5 × 10 9 /L was 13 days, and the median time to platelet count420 × 10 9 was 15 days (71 pt's). Acute graftversus-host disease of grade II to IV was observed in 64% of patients and chronic graft-versus-host disease in 55% of patients. At present 104 pt's are alive(88%) and 14 pt's died due to ARDS, VOD, hemorrhagic stroke, sepsis and relapse. With a median follow-up of 29 months (2-42 months) after transplant the three years overall survival were 84.1% and event-free survival was 82.1%.In Hodgkin's disease patient's overall survival and event free survival after autologous SCT better than neuroblastoma. Conclusion: Autologous SCT can lead to durable remissions in children and adolescents with Hodgkin's disease & solid tumor. These results indicate that despite our ward is new status both allogeneic and autologous HSCT are feasible with outcomes similar to developed countries. These preliminary data suggest that HSCTs have been used as one of the standard treatments for hematological diseases and malignancies in Iran. Disclosure of Interest: None declared. Intra-Bone Stem Cell Transplantation (IB-HSCT), based on the direct inoculation of umbilical cord blood CD34+ hematopoietic stem cells (UCB-CD34+), raised a great interest toward the mechanism regulating the BM-niche. In order to better understand the biological mechanisms occurring in the BMniche, we set up a co-culture system of UCB-CD34+ and MSCs and we studied their reciprocal relationship evaluating the proliferation and differentiation patterns and their gene expression profiles (GEP). Material (or patients) and methods: UCB-CD34+ cells were cultured with third party MSCs. After co-culture, proliferation rates were determined by flowcitometry count. Clonogenic assays were performed by culturing in methylcellulose and counting CFU-GEMM, CFU-GM and BFU-E. The GEP were performed on UCB-CD34+ and MSCs both in single culture and in co-culture, in order to establish which molecular signatures were activated in UCB-CD34+ and in MSCs. Microarrays were analysed with Parteks software. Gene set enrichment and pathway analysis were performed by using DAVID, GSEA, IPAs. Results: UCB-CD34+ in single culture were compared with UCB-CD34+ co-cultured with MSCs and a fold increase of 14.68 (P = 0.019) was found. By colony-forming unit (CFU) assay, we found a significant increase in the BFU-E ( p40.01), but we observed no differences in the CFU-GEMM and CFU-GM. GEP analysis of UCB-CD34+ in single culture appeared clearly distinct from UCB-CD34+ co-cultured-derived populations. By gene set enrichment analysis we found a significant enrichment in genes involved in heme metabolism (i.e. BMP2K), mitotic spindle formation and proliferation (i.e. SOX1), and TGFb signaling (i.e. ID1). The pathway analysis revealed a significant involvement of cell to cell signaling and interaction (i.e. ARHGEF2), hematological system development and function (i.e. HOXA9, HOXA10) and cell cycle, growth and proliferation (i.e. CCNA1, PCGF2). Some of these genes appeared to be key regulators of different processes affecting hematopoiesis. Microarrays analysis was also performed on MSCs in single culture and in co-culture with UCB-CD34+. A significant enrichment in genes involved in angiogenesis and VEGF signaling (i.e. VEGFA), proliferation and Aurora A kinase activity (i.e. AURKA, AURKB) and NF-kB pathway (i.e. TNFAIP3) was found. Furthermore, we observed a significant participation in cell to cell signaling and interaction, cell cycle and cardiovascular system development and function (i.e. ARF1). Conclusion: These results are consistent with a bidirectional relationship between UCB-CD34+ and MSCs in the BM-niche. MSCs play a role in the proliferation of UCB-CD34+ and in promoting the erythroid differentiation. UCB-CD34+ appear to induce a shift toward a vascular commitment profile in MSCs. These observations suggest that the interplay between UCB-CD34+ and MSCs modulates the BM-niche by inducing stromal and functional conditions typical of vascular BMniche and by favoring the proliferation and the differentiation of UCB-CD34+. This project is supported by BCC "Pompiano e Franciacorta" and Lions "Bassa Bresciana". Disclosure of Interest: None declared. Introduction: HSCs reside in the ex-vivo expansion by their self-renewal capacity and also they differentiate into a myriad of cell types, as well as the regeneration of stem cells. Whereas the DNA methyltransferase1 is thought to be principally involved in the maintenance of pre-existing methylation, the role of Dnmt1 in primitive stem cells pool preservation also is not known. In this study, mRNA expression of Dnmt1 was evaluated during CB-HSC ex vivo expansion with and without mesenchymal stem cells (MSCs). Material (or patients) and methods: Ex vivo cultures of CB-HSCs were performed in three culture conditions for 14 days: cytokines with MSCs feeder layer, cytokines without MSCs feeder layer and co-culture with MSCs without cytokine. After expansion measuring total number of cells (TNC), CD34 + cells and colony-forming unit (CFU) assay were performed.Dnmt1 mRNA expression was evaluated by real time -PCR. Results: Maximum CB-CD34 + cells expansion was observed in day 7 of expansion in all three cultures. Ex vivo expansion of CB-CD34 + cells after 7 days resulted in, significant decreased expression of Dnmt1 in cytokine culture, whereas in two coculture conditions expression of Dnmt1 was increased. Also a significant difference between expression of Dnmt1 in CD34 + cells and CD34cells of co-culture system with cytokine was observed. Conclusion: Expression of Dnmt1 that are an important regulator in normal hematopoietic regeneration was increased in CD34 + expanded cells on mesenchymal stem cells. Disclosure of Interest: None declared. Introduction: GLD is an autosomal recessive disorder due to deficiency of galactocerebrosidase (GALC) and characterized by myelination failure with progressive neurologic deterioration and death. We report the clinical outcome and long-term follow-up after HSCT of 7 patients (pts) affected by GLD who had been referred from 2000 to 2012 to our Bone Marrow (BM) Transplantation Unit. Material (or patients) and methods: 6 pts presented late onset and 1 infantile GLD. All pts received the graft from unrelated ≥ 9/10 HLA matched bone marrow (BM) donors except one who received a 6/6 unrelated matched cord blood unit (CB). Conditioning included busulfan plus cyclophosphamide. Graft-versus-host disease (GvHD) prophylaxis consisted of T-cell depletion by means of CD34+ positive selection and cyclosporine-A (CyA), methylprednisolone (MPred) and anti-thymocyte globulin (ATG) for the first 3 pts. (as previously reported*). The pt who received CB graft received CyA, MPred and ATG and the other pts CyA, methotrexate and ATG. Median follow up range: 1 -13 years. Results: Neutrophils cell engraftment occurred at day +19 (median, range: +12 -+27). Acute GvHD was limited in grade, i.e. ≤ II in 3 pts, and resolved following steroid therapy; none of the pts developed chronic GvHD. Post-HSCT course was with common early transplant-related complications (mucositis, hemorrhagic cystitis, infections) which solved with proper treatment and progressive immunological reconstitution; one pt presented posterior reversible encephalopathy syndrome. On the long-term, 5 pts show full donor chimerism and 2 have a mixed stable (85%) chimerism. All pts are alive, free of any immunosuppressive treatment and present normal sustained peripheral GALC activity. In four pts HSCT halted Central Nervous System (CNS) disease progression: three of them are successfully attending school and one graduated in chemistry few months ago. One patient presented myoclonus one year after HSCT and years later refractory epilepsy with unchanged MRI from pre-HSCT. One pt who presented seizures and mild cognitive impairment before HSCT worsened in the following years. In the early onset patients HSCT prolonged life expectancy, but the disease rapidly worsened soon after transplant. Introduction: Significant progress has been made towards standardisation of the ISHAGE protocol for enumeration of CD34+ cells. However commercial multi-level quality control reagents available for CD34 enumeration coupled with external quality assurance proficiency (QAP) programs do not control or monitor viability assessment of CD34+ cells using vitality dyes such as 7-AAD. Despite these limitations, many laboratories have replaced haemopoietic colony assays such as CFU-GM with the enumeration of viable CD34+ cells by flow cytometry using a single platform ISHAGE protocol including 7-AAD for the assessment of viability in thawed pilot vials prior to autologous or allogeneic haemopoietic progenitor cell (HPC) transplantation. After anecdotal reports that ammonium chloride (NH 4 Cl) lysis should not be used for thawed HPC, many centres either eliminated the lysis step or used alternative diluents such as saline or 5% dextran 40 / 2.5% albumin (DAS buffer) as a replacement. Material (or patients) and methods: At our centre like many others in Australia, the enumeration of viable CD34+ cells in thawed HPC was performed using a single platform ISHAGE method with DAS diluent followed by immediate data acquisition (in house method). The in house method was replaced in June 2014 by the Becton Dickinson Stem Cell Enumeration Kit (SCEK) for labelling and clinical software on the FACSCanto II for acquisition and analysis. The SCEK kit includes a 10 min NH 4 Cl lysis followed by placement on ice for o1 h prior to acquisition. Pre-implementation validation of the SCEK showed a high degree of correlation between the in house and SCEK methodologies (r 2 = 0.94). Results: Following implementation of the SCEK, a small group of patients were identified with very low recovery of viable CD34+ cells during pilot vial analysis. Further validation studies were performed that included (i) a comparison of the in house methodology and the SCEK for the enumeration of viable CD34+ cells with CFU-GM colony assays for viability assessment (n = 23, r 2 = 0.89 /0.85) (ii) mixing of "dead" HPC, Apheresis (HPC(A)) generated by incubation at 57 o C for 25 min with "alive" HPC(A) at the following ratios: 0:100, 5:95, 10:90, 20:80; 40:60, 60:40; 80:20; 90:10, 95:5 and 100:0 ( Fig. 1 ) (iii) individual confirmation of HPC with low viability by CFU-GM colony assays and (iv) demonstration of viable CD34+ cell stability in the CD34 enumeration assay using the SCEK for up to 2 h post NH 4 Cl lysis. Conclusion: These studies endorse the use of the SCEK with an NH 4 Cl lysis step prior to data acquisition during the enumeration of CD34+ cells in thawed HPC and highlight that caution should be applied to the use of DAS buffer as a diluent. Viable CD34+ cell recoveries may be overestimated due to interference with the ISHAGE Boolean gating strategy. Careful manual gating by an experienced operator may reduce the apparent overestimation of viable CD34+ cells. Retrospective studies will be performed to determine if any delays in engraftment kinetics following infusion of cryopreserved HPC prior to June 2014 could be attributed to an overestimation of viable CD34+ cells by the in house methodology. Contributing factors to low viability observed using SCEK in a small number of cryopreserved HPC is also under assessment. 2 General Hospital "Alexandra", Athens, Greece Introduction: Unrelated donor cord blood transplantation has become a widely accepted treatment for hematologic diseases. Today, the role of total nucleated cell (TNC) dose and HLA match is affecting the time of engraftment, as well as the transplant-related events and survival. Data suggest that the maternal-fetal experience may convey tolerance to the maternal HLA that was not inherited by the fetus. These noninherited maternal antigens (NIMA) may define permissible HLA mismatches and could be used to extend the genotypes that are suitable matches for particular donors or umbilical cord blood units (CBUs) . CBU transplantations to patients mismatched for only 1 HLA antigen, identical to the CBU's NIMA, are designated as having a 6/6 "virtual" NIMA-matched phenotype (VP) and have a prognosis similar to 6/6 inherited HLA-matched CBUs. The aim of the study was to investigate the role of NIMA in the increase of the number of available compatible CBUs for a specific patient-recipient, in order to be transplanted or used as "back up" CBUs. Furthermore, was to study the role of NIMA in acceleration of the deployment process of the high quality CBUs of the HCBB, as the long term storage has a negative effects on their viability. Material (or patients) and methods: The initial step was to create VP for each one of the banked CBUs of the HCBB: by replacing the compatible HLA-antigens with non-inherited ("acceptable mismatches") maternal antigens of the CBU, in order to increase the available CBUs-for a specific recipient. Such VP of CBUs can be created by replacing the inherited alleles with 1 or more NIMAs. The second step was to search among the HCBB's banked CBUs and find every CBU with HLA antigens identical to a VP of each CBU that was analysed. Subsequently, the VP-matched CBUs were classified, depending on their cellular content. Results: At the study of the first 300 banked CBUs, included in the category of CBUs with 3 mismatches with the HLA-A, B and DRB1 antigens of the mother of the neonatal donor, we resulted in the creation of 26 VP for each one of the 90 CBUs. The final number of VP designed for 90 CBUs, having 3 HLA mismatches with the mother of the specific donor was 90x26 = 2340. Within the HCBB's banked CBUs, 151 CBUs with HLA phenotypes similar to the VP generated on 80 of the 90 CBUs analyzed previously, were found and recorded. In addition, the number of nucleated cells and CD34 + cells of each stored CBU was recorded, in order to classify the fully compatible -to any VP-CBUs. For example, a randomly-picked CBU, had the following HLAantigens: A 11, 24 / B 35, 51 / DRB1 11, 14 and the CBU's mother had the following HLA-antigens: A 2, 24 / B 35, 49 / DRB1 13, 14 Conclusion: Apart from a specific CBU banked, a CBB can count on more CBUs, VP-matched to the initial CBU-requested by a transplant center. The final number of CBUs could be released, additionally, for a specific recipient, at the same time and cost, or could play the role of "back-up" CBUs for a transplant center, in case that something goes wrong with the initially requested CBU. Disclosure of Interest: None declared. Shift in progenitor subset composition in bone marrow of stem cell transplant recipients in comparison to stem cell graft composition L. Kordelas 1,* , A. Görgens 2 , P. Horn 2 , D. Beelen 1 , B. Giebel 2 1 Department of Bone Marrow Transplantation, 2 Institute for Transfusion Medicine, University Hospital Essen, Essen, Germany Introduction: By combining CD34, CD133 and CD45RA staining hematopoietic stem cells (HSC) can be discriminated into the following progenitor subsets according to their immunophenotype via flow cytometry (FC): multipotent progenitors (MPP: CD133+ CD34+ CD45RA-), lymphomyeloid progenitors (LMP: CD133+ CD34+ CD45RA+) and erythro-myeloid progenitors (EMP: CD133low CD34+ CD45RA-). In this study we investigated the relationship of the frequencies of these progenitor subsets in GCSFstimulated peripheral blood stem cell (PBSC) grafts and in the bone marrow (BM) of the PBSC recipients three to four weeks after allogeneic stem cell transplantation (AlloSCT). Material (or patients) and methods: FC analyses were performed in samples of 20 different PBSC grafts and in the BM of the patients having received these PBSC grafts using antibodies for CD34, CD133 and CD45RA. After lysis of Material (or patients) and methods: Bone marrow cells obtained from 8 individuals (7 healthy donors and one pts with limb ischemia, age (yrs) : range 39-71 years, 4 female/ 4male) were cultured in a closed and automated medical device (Quantum Cell Expansion System). Twenty five mls of S127 fresh bone marrow cells were cultured in Minimal Essential Medium alpha and human platelet lysate (5%). A final product was described functionally, phenotypically and subtracted to the quality control parameters measurements. Results: The cultures were terminated and passaged when the level of lactates reached 4,0 mmol/l (on 9-16 day of culture) After that the cells were usually passaged to next run of culture in the same culture conditions. Harvested cells (after the first passage) characteristics was as follow. Total count (median) 7,6E+07 cells (6,91E+06 to 1,0E+08 cells), CD45-CD34-cells constituted (mean) 96,88% ± 2,0%, in the latter population we found 98,83 ± 0,78%, 98,58 ± 1,16% and 96,43 ± 2,68% of CD105+, CD90+ and CD73+ cells, respectively. Microbiologic surveillance proved that the products were free from mycoplasma, chlamydia as well bacteria and fungi, in addition the cultures were LPS free. The donors were HBV DNA, HCV RNA and HIV RNA negative. Introduction: Mesenchymal Stromal/Stem Cells (MSC) isolated from adult human tissues are ideal candidates for tissue regeneration due to their immunosupresive properties and regenerative potential. In this study we analyzed phenotype and differentiation potential of MSC isolated from human bone marrow and skeletal muscle to characterize their biological diversity. Material (or patients) and methods: Bone marrow (BM) and skeletal muscle were collected from 16 deceased donors, from 12,5 h to 48 h after dead, with approval by the local Bioethics Committee. Mononuclear BM cells were isolated on ficoll gradient. Skeletal muscle samples were placed in PBS supplemented with antibiotics and muscle-origin stromal/ progenitor cells were isolated with collagenase or trypsin digestion. After tissue enzymatic digestion, long-term cultures in standard DMEM medium (skeletal muscle) or in alpha-MEM medium (bone marrow) supplemented with 10% FBS and antibiotics, at 37 o C and 5% CO 2 were performed. Cells were observed up to 12 passages (P) and examined for presence of: CD34, CD45, CD56, CD73, CD90, CD105, CD146, PDGFR-α, desmin and dystrophin using immunocytochemistry and flow cytometry. Osteogenic, adipogenic, neurogenic and chondrogenic processes were carried out from 14 to 30 days with dedicated differentiation medium. Results: Adherent cells isolated from bone marrow and those of myogenic origin express phenotype characteristic for naïve MSC CD73, CD90, CD105 as confirmed by flow cytometry and immunofluorescence staining, however their expression downregulated during follow-up period (after P7). A fraction of MSC isolated from skeletal muscle express myogenic markers: CD56, desmin (up to P7) and dystrophin and their expression increased in late passages (after P7). Co-experssion of CD73/CD146, CD90/CD146 and CD105/CD146 on the proportion of adherent cells was detected in MSCs of bone marrow and skeletal muscle origin. A fraction of cells expressing CD146 strongly co-expressed PDGFR-α (up to P7). Cells isolated from both tissues (bone marrow and skeletal muscle) were capable to differentiate into chondrocytes, osteoblasts and neurons, however, only MSC isolated from skeletal muscle were not capable to form adipocytes. Conclusion: Adherent cells propagated from bone marrow and from skeletal muscle express phenotype characteristic for naïve MSC. Co-expression of CD146 and markers specific for naïve MSC suggest that those cells are mesenchymal progenitors capable to differentiate into pericytes. Coexpression of CD146/PDGFR-α and increased expression of dystrophin characterize distinct population of MSC with myogenic potential. Multipotent clonogenic capacities of MSC of bone marrow origin were confirmed, whereas mesenchymal progenitor cells from skeletal muscle have limited differentiation potential. Disclosure of Interest: Studies supported by NCN grant NN407121940. None declared. Topical treatment of chronic GVHD-related severe dry eye with Autologous Serum A. De Rosa 1,* , P. Iudicone 1 , R. Leone 1 , C. Lavorino 1 , S. Costantino 1 , D. Fioravanti 1 , R. Bonfili 2 , L. Pierelli 3 1 UOC SIMT, 2 UOC Oculistica, AZIENDA OSPEDALIERA S.CAMILLO-FORLANINI ROMA ITALIA, ROMA, 3 Experimental Medicine, Sapienza University, Rome, Italy Introduction: Ocular Graft vs Host Disease (GVHD) occurs in patients who have undergone allogeneic hematological stem cell transplant and it is more common in patients with chronic GVHD (cGVHD). Ocular complication secondary to GVHD can include keratoconjunctivitis sicca, keratitis, uveitis and corneal ulceration but the sicca syndrome (SSy) is the most frequent. The decrease in tear production is mainly due to the infiltration of mononuclear cells involved in the inflammatory process at lacrimal gland level. Although systemic immune suppression for cGVHD can improve SSy, supportive local therapy are usually required aimed to lubricate the surface of the eye and to reduce inflammation. Currently, specific treatment for SSy is not available and topical cyclosporine and commercially artificial tears often fail to improve the ocular symptoms. Material (or patients) and methods: Eleven patients(5 females 6 males, 30-64 years) with cGVHD-related severe dry eye have been treated with autologous serum (AS) eyedrops. All patients were refractory to conventional treatments. Before enrollment, the patients were evaluated for blood parameters, Schirmer Test (ST) and Tear Film Break Up Time (TFBUT). Moreover, before starting the treatment and after 2, 6, and 12 months, the patients underwent to a self-assessment questionnaire to evaluate "sensitivity to light", "gritty feeling", "pain" and "blurred vision "on a scale of 1 at 4 with high score representing greater symptoms. Individual preparation of AS was produced by collecting 100 ml of whole blood in bag without anti-coagulant. The bag was left at room temperature (90') waiting for clotting, then centrifuged to separate the serum, diluted to a final concentration of 80%, by adding isotonic saline, aliquoted in 1 ml sterile vials and stored at -20°C until delivery to the patients. Sterility tests for aerobic, anaerobic and fungi were carried out for every AS preparation to exclude the risk of contamination, which could result in severe ocular infections. The concentration of platelet growth factors FGF-basic TGF-ß, PDGF-AB, VEGF-C, EGF were evaluated in AS samples by ELISA. Patients received AS aliquots with the instructions to keep them at +4°C for 1 month and to apply the drops several times (6-8) a day according to individual symptoms. Results: All the AS preparations were found sterile. FGF-basic was not detectable in AS samples, whereas the mean values of the other growth factors were: TGF-ß 8895 pg/ml, PDGF-AB 4390 pg/ml, VEGF-C 3283 pg/ml, EGF 276 pg/ml. The follow-up at 2, 6 and 12 months indicated that the patients had no side effects and they experienced, already after 2 months, a considerable improvement of ocular symptoms, based on selfassessment questionnaire score. No relevant change were found in ST and TFUBT. Conclusion: AS tears provide elements able to restore ocular surface environment offering to the patients beneficial clinical effects which, however, were stable as long as they were receiving the AS treatment. Disclosure of Interest: None declared. Mesenchymal stromal stem cell (MSC) immunotherapy for experimental septic shock: systematic review and metaanalysis with trial sequential analysis of mortality A. Patel 1,2,* , S. Jhanji 3 , J. Pavlu 4 , M. Laffan 2 , M. Ethell 5 , A. Bradshaw 6 , E. Olavarria 7 , K. Harrington 1 , J. Apperley 2 , S. Brett 8 1 Targeted Therapy, Institute of Cancer Research, 2 Centre for Haematology, Imperial College London, 3 Critical Care Unit, The Royal Marsden NHS Foundation Trust, 4 Imperial College Healthcare NHS Trust, 5 Haemato-oncology Unit, The Royal Marsden NHS Foundation Trust, 6 John Goldman Centre for Cellular Therapy, 7 Centre for Haematology, 8 Centre for Perioperative Medicine and Critical Care Research, Imperial College Healthcare NHS Trust, London, United Kingdom Introduction: Septic shock is a life-threatening form of an inappropriate host response to severe infection, with impaired survival in patients with cancer. Mortality is 60-75% for patients with haematological or solid cancer -1/3 higher than patients without cancer. Cellular immunotherapy with allogeneic MSCs is a promising multi-targeted inflammation and bacteria responsive personalised treatment, which may address this unmet clinical need. We hypothesised that MSC immunotherapy might improve survival in experimental septic shock models. Material (or patients) and methods: We performed a systemic review and meta-analysis as described previously 1 with the research question: what is the safety and efficacy of MSC immunotherapy for experimental septic shock. We chose mortality at the longest follow-up as the primary outcome measure in wild-type animals. Unmodified MSC groups from any source were compared to non-MSC groups. Results: All 21 included studies were assessed as either unclear or high risk of bias, using the Cochrane Risk of Bias Assessment Tool. This was mainly because of poor reporting: possible selection bias (due to poorly reported random sequence generation and allocation concealment) and performance bias (due to the lack of blinding). The overall pooled effect size was in favour of MSC immunotherapy, using a random effects model: RR 0.59, 95% CI 0.48-0.73; P = 0.0001; I 2 = 82%. The large effect size was robust to sensitivity and meta-regression analyses, including adjustments for baseline mortality (60%) and co-treatments (fluids and antibiotics). Egger's regression did not detect publication bias (P = 0.20). To ensure our meta-analysis was sufficiently large and adequately powered, and to test the robustness of the finding of MSC benefit, we subjected it to trial sequential analysis. 1 The cumulative z score crossed the conventional boundary of benefit (alpha of 0.05) and the constructed trial sequential boundary (Figure 1 ), correcting for repetitive testing of the same hypothesis. Furthermore, the z score also crossed the 95% power boundary demonstrating that our meta-analysis was sufficiently large to be confident of the result. Figure 1 . Trial sequential analysis. The cumulative z score of 21 studies, ordered by year, are indicated in blue. The conventional alpha (0.05) significance boundaries are indicated by parallel horizontal red lines. The curved red lines indicate the trial sequential corrected boundaries of benefit (upper half) and harm (lower half). The inner wedge to the right of the graph indicates the boundary of futility. The red vertical boundary indicates the information size for 95% power. The cumulative z score crosses the conventional boundary of benefit, the trial sequential boundary of benefit, and the 95% power boundary indicating after accounting for bias there remains robust evidence of the efficacy of MSC immunotherapy for experimental septic shock. Conclusion: We robustly demonstrate a 41% relative reduction in death with MSC immunotherapy, representing a number needed to treat of just 4. This compelling experimental evidence requires clinical translation in patients with haematology cancer suffering from septic shock. References: 1. Patel A, Laffan MA, Waheed U, Brett SJ. BMJ 2014;349:g4561. Disclosure of Interest: None declared. The effect of time and temperature during transportation of umbilical cord tissue on viability, functionality and differentiation potential of mesenchymal stem cells following cryopreservation and thawing A. Smith 1,* , A. Sanchez 2 , M. Fabra 2 , N. Aparicio 2 , E. Ainse 2 , K. Hussain 1 , F. Delgado-Rosas 2 1 Smart Cells International, London, United Kingdom, 2 Ivida, Cord Blood Bank of the IVI Group, Madrid, Spain Introduction: There is considerable interest in the potential of mesenchymal stem cells (MSCs) derived from umbilical cord blood (UCB) and more recently cord tissue (CT) for use in regenerative medicine and banking of CT along with UCB. There are recognised procedures to extract and culture viable, functional MSCs pre/post cryopreservation but the effects of transit conditions from collection to laboratory have not been delineated. Transit conditions are integral to quality assurance to ensure an efficacious end product. This study aimed to define such conditions. Material (or patients) and methods: CT samples (n = 15) were transported at approximately 10 o C from clinic to laboratory with continual temperature monitoring. On receipt, dissected S129 segments were either cultured fresh or subjected to temperatures of 2 o C, 10 o C or 28 o C for 2, 3, 4 or 5 days prior to cryopreservation, thawing and subsequent assessment. In all cases, segments were cultured in DMEM-F12 with antimicrobial agents and 10% foetal bovine serum at 37°C with 5% CO 2 for 14 days with regular medium changes. Following second passage, viable cells which typically formed an 80-90% confluent monolayer were trypsinised. Viability was assessed with trypan blue. Flow cytometric analysis of markers specific for human MSCs (CD90, CD105and CD73) and also negative (CD11b, CD19, CD34-, CD45 and HLA-DR) was performed. The in vitro differentiation potential of the harvested MSCs along adipogenic, osteogenic and chondrogenic lineages was undertaken using commercially available induction media. Results: Viability of MSC precursors in fresh CT over all transit times at 28 o C, was significantly impaired (p ≤ 0.05) compared to no significant effect at 2 o C and 10°(P = 0.38 and P = 0.069 respectively). At 28 o C, viability levels were 80%, 40% and 20% after 3, 4 and 5 days respectively. Following thawing and culture, explants that had been stored at 2 o C for 4 and 5 days exhibited viability of 100% and 80% respectively. At 10 o C, 100% viability was apparent for all storage time points. Storage at 28 o C resulted in viability levels of 80%, 40% and 0% for 3, 4 and 5 day time points. Without exception, all viable cells recovered after thawing displayed phenotypic and differentiation characteristics typical of MSC cells. Conclusion: This study demonstrates no significant adverse effect on generation of viable and functional MSCs from frozen/thawed CT samples following up to 5 days in storage/ transit while fresh at 2 o C or 10 o C with 10 o C being optimal. Furthermore, cells following fresh storage at 28 o C for 3 days recovered after freeze/thawing with 80% viability and were phenotypically typical of mesenchymal cells exhibiting adipocytic, osteocytic and chondrocytic differentiation potential. CT derived MSCs are being increasingly considered for indications in the growing field of regenerative medicine and also for their ability to modulate the immune response. This study has delineated permissible time lapse from collection to storage of CT and the range of transit temperatures over which samples are able to yield viable MSCs that retain their biological and regenerative properties. Disclosure of Interest: None declared. Development of an internet intervention targeting reproductive and sexual health following cancer J. Winterling 1,2,* , M. Wiklander 1 , C. Lampic 1 , C. Micaux Obol 1 , L. E. Eriksson 3, 4, 5 Introduction: Approximately 50% of adolescents and young adult patients with cancer (15-39 years old), including those with leukaemia and lymphoma, report concerns about their reproductive and sexual health following cancer treatment. It is also known that these concerns seldom are addressed in clinical care. The internet has shown to be effective for delivery of information, support and psychological treatment, especially when the target group i accustomed to the technologies and are at a geographical distance. The aim of the study was to develop and test an internet intervention ro reduce fertility distress and sexual problems in young persons diagnosed with cancer. Material (or patients) and methods: The development process is described according to the holistic framework presented by van Gemert-Pijnen. Twelve research partners, former patients and significant others, were recruited. The internet intervention was built together with the patient research partners in a long-term collaboration and will be tested in a randomized controlled trial (RCT). A national cohort of adolescents and young adults with selected cancer types will be approached one year post-diagnosis. Those rating high levels of fertility distress and/or sexual dysfunction on standardized self-reported measures will be invited to participate in the RCT with two arms, intervention versus control group. Results: A 12 weeks interactive internet psycho-educational program (Fertility and sexuality following cancer, Fex-Can) was developed. The program consists of two parts, one focusing on reducing fertility distress and the other part on handling sexual problems. The program includes medical and healthrelated information e.g. treatment effects on fertility and sexuality, fertility preservation options, post-treatment pregnancy, genetic risk, alternative ways to build a family, bodily changes, and therapeutic aids to increase sexual function. The program aims at increasing participants' knowledge and skills to handle worry and stress reactions, and teaches techniques for both acceptance and behavioral change. The program also includes short videos of patients sharing their experiences and interactive components like posting questions to experts and a discussion forum. The part of the program focusing on handling sexual problems includes two individual telephone-counseling sessions, to set up and evaluate individual goals. Conclusion: An interactive internet psycho-educational program to alleviate fertility distress and sexual problems was created in collaboration with patient research partners and the effectiveness of the program will be tested in the main trial (RCT) starting fall 2016. Disclosure of Interest: None declared. The use of filgrastim alone compared with filgrastim plus etoposide, with or without plerixafor, for stem cell mobilization -a single centre experience A. Khoder 1,2,* , M. Sever 3 , Z. Allwood 2 , R. Haynes 2 , E. Bray 2 , A. Chaidos 2 , D. MacDonald 2 , T. Karadimitris 1 , H. Auner 2 , A. Rahemtulla 1 , J. Apperley 1 , E. Kanfer 2 , E. Olavarria 2 1 Haematology, Imperial College London, 2 Haematology, Hammersmith Hospital, London, United Kingdom, 3 Haematology, University Med. Center, jubljana, Slovenia Introduction: The great majority of autologous stem-cellsupported procedures (auto-SCT) are performed with mobilized blood stem cells (HSC). GCSF alone or combined with chemotherapy is most commonly used. Here we compared the efficacy of GCSF alone (G-only) or in combination with etoposide (E+GCSF) with plerixafor in failing cases. Material (or patients) and methods: This prospective study included all patients who underwent stem-cell mobilization (SCM) at Hammersmith Hospital between 01/02/15 to 01/10/ 2015. A target of 2x10 6 CD34 + cell/kg was used. All myeloma patients received G-only, germ-cell-tumour patients received E +GCSF, and lymphoma either GCSF or E+GCSF at the clinician's discretion. Etoposide (total 1.6 g/m 2 ) was given on D1&2 and biosimilar filgrastim (5 μg/kg) SC OD followed from D3 to D13. Apheresis took place on D13 if PB CD34 + count420/μl. In the G-only regime SC biosimilar filgrastim was given at 10 μg/kg OD for 5 days and the harvest was scheduled on 5 th day if PB CD34 + count410/μl. Results: A total of 39 patients underwent mobilization with G-only and 18 E+GCSF; MM (n = 34), NHL (n = 10), HD (n = 9) and GCT (n = 4). Compared with E+GCSF, G-only yielded significantly lower PB CD34 + counts and percentages; median 32.4/μl (3.4-192.4) vs. 94.98/μl (2.7-1900) , P = 0.013 with median 0.09% (0.1-0.57) vs. 0.18% (0.02-4.2) , P40.0001 respectively. In G-only, 14/39 (36%) required 1 day of harvest, 22/39 (56%) 2 days and 3/39 (7%) 3 days; a mean of 1.7 days. In E+GCSF group 14/18 (78%) required 1 day, 5/18 (22%) 2 days; a mean of 1.3 days. A median CD34 + cell dose of S130 4.1x10 6 /kg (1.5-11.2) vs. 6.9x10 6 /kg (0.37-69) (P = 0.002) was achieved in G-only compared with etoposide group. Eleven patients (28%) failed to mobilise with G-only; of these 9/11 (82%) needed one dose of plerixafor and 2/11 (18%) two doses. In E+GCSF group, 2 patients required plerixafor 2/18 (11%) (P = 0.18,)Fisher exact test). Characteristics of patients who failed to mobilize were: median age 59 (30-69), 54% males, median platelets 210x10 9 /l (89-448), median neutrophil count 3.5 x10 9 /l (1.1-7.4 ), 1/13 advanced disease (known previous harvest failure), 1/13 BM involvement, 2/13 had received 3 lines of chemotherapy, 1/13 involved-field-radiotherapy, 5/13 low BMI and 3/15 had a Karnofsky score of 70-80. Nine patients (50%) with E+GCSF required hospital admissions (median 5 days) for neutropenic sepsis compared to one admission (2%) with G-only for venous access care. PB CD34 + count on D5 of G-only correlated with harvested CD34 + cell dose; Spearman r 0.75 P40.0001. Receiver operating characteristic curve for prediction of CD34 + /kg harvest was plotted for G-only cohort. PB CD34 + count for those who failed to harvest on D1 was compared to those who harvested450% of target on D1. A new cut off value of 17x10 6 /L is suggested to decide the start of apheresis; sensitivity 96% & specificity 81%. Although 28% of G-only harvested o 2x10 6 /kg target, all patients underwent SCT and engrafted. Median time to engraft was 13 days (9-25) vs 15 days (12-28); P = 0.05, for patient meeting target vs those who partially met respectively. Conclusion: In conclusion, biosimilar GCSF approach can be recommended as a cost-effective and safe SC mobilization method. However, rescue alternative e.g. plerixafor should be available for potential failure. Disclosure of Interest: None declared. Efficacy analysis of plerixafor "on demand" combined with chemotherapy and granulocyte colony-stimulating factor for stem cell mobilization in patients with lymphomas Introduction: High-dose chemotherapy with autologous haematopoietic stem cell transplantation (ASCT) is an effective treatment for patients with lymphomas. However, a significant proportion of patients with non-Hodgkin lymphomas (NHL) or Hodgkin lymphoma (HL) fails to mobilize or to collect an adequate number of CD34+ hematopoietic stem cells (HSC) for a safe procedure. The aim of this retrospective study was to evaluate the efficiency of HSC mobilization with the use of chemotherapy in combination with granulocyte colonystimulating factor (G-CSF) or G-CSF alone and plerixafor administered "on demand" as a rescue strategy adopted in our centre. Material (or patients) and methods: The records of patients with NHL/HL who underwent HSC mobilization with chemotherapy plus G-CSF (4 5 μg/kg/day) or G-CSF alone (10 μg/ kg/day) and received plerixafor (240 μg/kg/day) "on demand" in the absence of increase in the number of CD34+ cells above 10/μl on the day of the schedulded apheresis procedure (within 20 days following the chemotherapy and after at least 4 days of G-CSF) between January 2011 and November 2015. Patients records and peripheral blood stem cell collection data were reviewed to obtain patients' and disease characteristics at the time of mobilization procedure, the number of apheresis needed to collect at least CD34 + 2x10 6 cells and the number of CD34+ cells (mln/kg bw) obtained after each performed apheresis. Results: The study group consisted of 20 patients, 9 women and 11 men with median age of 36 years (range, 21-64 years) with NHL (n = 12) or HL (n = 8) who underwent HSC mobilization with chemotherapy plus G-CSF (n = 17) or G-CSF alone (n = 3) and received plerixafor "on demand". The patients received a median of 3 (range, 2-7) chemotherapy lines prior to HSC mobilization with plerixafor. Eight patients (40%) were previously exposed to radiotherapy. Eighteen patients (90%) failed at least one previous HSC mobilization attempt (CD34+ cells yield less than 2,0 x 10 6 /kg). After plerixafor administration the peripheral blood (PB) CD34+count increased to at least 10/μL in 17 patients (85%). Those patients collected a median of 2.96 × 10 6 (0.28 × 10 6 -6.67 × 10 6 ) CD34+ cells/kg after a median of 3 (range, 1-4) apheresis days. The median CD34+ cell dose collected in patients with HL and with NHL was 3.57 × 10 6 /kg and 0.42 × 10 6 /kg, respectively (P = 0.02). The addition of plerixafor enabled 9 patients (45%) to reach the 2.0 × 10 6 CD34+ cells/kg minimum required for ASCT. With regard to diagnosis, 6 of 8 patients (75%) with HL and 3 of 12 patients (25%) with NHL collected the minimum stem cell yields (P = 0.06). Conclusion: Our results confirm, that plerixafor administered "on demand" in HL patients with poor HSC mobilization with chemotherapy and G-CSF is an effective rescue strategy. However, based on our experience, the efficacy of this strategy appears to be limited in heavily pretreated patients with NHL who are proven "poor mobilizers". References: Milone G, Tripepi G. Algorithms for early identification of poor mobilization and for on-demand use of plerixafor in patients mobilized by chemotherapy and granulocyte-colony stimulating factor. Leuk Lymphoma. 2014 Mar;55 (3) Introduction: Neutropenia is associated with high-dose conditionings and autografting. Biosimilar filgrastim (Nivestim TM , Hospira Inc.) (BioG-CSF) is a granulocyte-colony stimulating factor approved for clinical use in Europe since 2010 to favor hematological recovery. This single-institution study was designed to evaluate safety and efficacy of the use of BioG-CSF in the setting of "real-life" medical practice. The study is of particular interest given the propensity for biosimilar growth factors to reduce overall costs. Material (or patients) and methods: We designed a "mixed retrospective-prospective study" to evaluate the impact of BioG-CSF on CD34+ cell collections and on engraftment kinetics after high-dose chemotherapy and stem cell support. Moreover, data were compared with a historical patient cohort treated with "originator G-CSF" (G-CSF). Data in the BioG-CSF cohort have partly been collected prospectively. The study period includes the years 2007 up to 2015 during which standard policies for autografting have not changed. Primary endpoints were engraftment kinetics: median first day of absolute neutrophil count ≥500/ul for 3 consecutive days and duration of neutropenia defined as the number of days with neutrophil count o500/ul; and number of red blood cell and platelet transfusions / patient. Secondary objectives, categorized by disease type, included non-relapse mortality at day 100 post-transplant, overall and event free survivals at 1 year. Results: 189 patients have so far been enrolled. Diseases include plasma cell disorders (no. 148), non-Hodgkin and Hodgkin lymphomas (no.31 and no.5), chronic lymphocytic leukemia (no.1) and acute leukemias (no.4). Most patients had adequate CD34+ cell mobilization and harvests for at least 2 autografts. Less than 3% were poor mobilizers in both cohorts. No allergies were documented. All patients were hospitalised and prepared for the autograft with melphalan at 200 mg/m 2 for plasma cell disorders, with BEAM or FEAM conditionings for lymphoproliferative disorders and busulfan-cyclophosphamide for acute leukemias. After the autograft, the majority of patients were prescribed 34 million units (MU) of G-CSF or 30 MU dose of BioG-CSF administered IV starting on day +1/+3. Overall, 118 received G-CSF and 71 BioG-CSF. Antibiotic prophylaxis consisted of ceftazidime or levofloxacin. Median first day of absolute neutrophil count ≥500/ul for 3 consecutive days was +11.6 (r 7-35) in the G-CSF group vs. +11. 3 (r 8-18) in the BioG-CSF group whereas duration of neutropenia was 7.2 (r 3-31) days and 6.45 (r 4-12) days. Median numbers of red blood cell and platelet transfusions/patient were 1.17 (r 0-10) and 1.96 (r 0-12), and 0.84 (r 0-6) and 1.74 (r 0-13) in the G-CSF group and the Bio-G-CSF group respectively; hospital stays lasted a median of 22.78 (r 16-62) days and 22.65 (r 8-49) days respectively. All patients were discharged except one who died on day +10 for septic shock. No significant differences in documented infections were reported. In regard with the secondary endpoints of the study, OS, EFS, and NRM at 1 year did not differ. Conclusion: BioG-CSF was as effective and well-tolerated as G-CSF in mobilising CD34+ cells and treating post-transplant neutropenia in patients with hematological malignancies. Moreover, its extensive use led to a significant cost reduction. Updated study results will be presented. Introduction: Plerixafor is used prior to an ASCT as a salvage mobilization therapy in patients who mobilize poorly (CD34+ o10cells/ml in the peripheral blood) after 4 days with G-CSF. Plerixafor-mobilized grafts have been described to contain fewer CD34+ cells than good mobilizers' grafts 1 . The aim of this retrospective single center study is to compare the number of CD34+ cells contained in the grafts of poor mobilizers who required plerixafor (plerixafor group) with good mobilizers (control group), and evaluate its consequences on engraftment and transfusion requirements. Material (or patients) and methods: Between December 2012 to October 2015, 17 patients (11 adults and 6 children) median age 28 (range 2-63) with non-Hodgkin's lymphoma (n = 5) or Hodgkin's disease (n = 7) and solid tumors (n = 5) were mobilized with plerixafor 0.24 mg/kg subcutaneously. In one patient plerixafor was immediately administrated as salvage after G-CSF, while in the rest of patients it was used after remobilization with G-CSF. We compared this group to a control cohort of 19 patients, median age 37 (range 1-68) who had similar distribution of hematologic malignancies and solid tumors but who were good mobilizers with G-CSF alone (n = 9) or G-CSF+chemotherapy (n = 10). Results: Median infused CD34+ cells was 2.2x10 6 /kg (1.33-4. 8x10 6 /kg) in the plerixafor group and 3.60x10 6 /kg (2.12-7. 1x10 6 /kg) in the control group ( p40.005). Median time to neutrophil engraftment (40.5x10 9 /L) was comparable in both groups, (13 days in plerixafor group vs 12 days in control group, P = 0.4). Median time to platelet engraftment (420x10 9 /L) was slightly faster in plerixafor group (18 days vs 20 days, P = 0.09), although partial hematologic reconstitution (450x10 9 /L platelets) within 30 days was achieved in 41% of patients in plerixafor group and 61% of patients of control group. The number of red blood cells units transfused was similar in both groups (3 in plerixafor group vs 4 in control group) as well as platelet pools (4 vs 5) . Conclusion: Blood grafts in the plerixafor group contained significantly lower CD34+ cell dose. However, we do not observe delay in neutrophil engraftment and there is slightly faster platelet engraftment in the plerixafor group. Moreover, transfusion support is comparable between both groups. Higher proportion of primitive hematopoietic precursors (CD34+CD38-) has been previously described in grafts after plerixafor mobilization 2 , which could explain the high efficiency of the grafts and short term hematologic recovery observed in our patients. Our findings need to be confirmed in larger prospective studies Introduction: High dose chemotherapy followed by autologous stem cell transplantation is the standard of care for patients affected by multiple myeloma. Quality and quantity of mobilisation of peripheral blood stem cells determine the safety and efficacy of tandem autologous transplantation in myeloma. In this study, we have compared the ability of granulocyte colony stimulating factor (GCSF) and it's pegylated form (pegGCSF) to mobilise peripheral blood stem cells for collection. Aim of this study is to compare efficiency of GCSF and pegGCSF in mobilisation of peripheral blood stem cells and to compare time from mobilisation to leukapheresis as well as number of leukapheresis needed to obtain required number of the stem cells. Material (or patients) and methods: From August 2004 to October 2015, hundred and twenty-two patients (122 pts) with MM (59 male, 63 female) with median age 54 (range 30-71) underwent ASCT. Patients were divided in two groups. S132 First group consisted of MM patients treated between January 2004 and August 2011 and all of them received VAD induction therapy for 4-6 cycles. Second group were patients treated between September 2011 and October 2015, all of whom were treated with bortezomib based therapy (either VD or VCD) for 3-4 cycles. All patients included have achieved at least very good partial remission. Also, all patients were treated with 3 g/m2 of cyclophosphamide as peripheral blood stem cell mobilisation therapy. First group received GCSF daily, starting on day 5 after cyclophosphamide while second group received pegfilgrastim 24 hours after cyclophosphamide was applied. Results: Group of patients receiving pegGCSF had a statistically significant shorter mean time to leukapheresis, specifically, 9.77 days for pegGCSF compared to 11.05 days for GCSF, P40.05. More than one leukapheresis was necessary in 9 (14,28%) patients receiving pegGCSF compared to 37 (62, 7%) patients receiving GCSF (P40,05). In group treated with pegGCSF average number of CD34+ stem cells collected was 12.49x10 6 cells per kilogram of body weight while in group treated with GCSF average of 8.83x10 6 cells per kilogram of body weight was collected. Conclusion: Our study has shown that patients receiving pegGSCF had lower number of leukapheresis, shorter time to leukapheresis and also higher number of CD34+ stem cells collected in this group. Application of single dose pegGCSF is easier for both patient and medical staff. Disclosure of Interest: None declared. Impact of CD3+ cell dose in the graft on cumulative incidence of relapse after allogeneic stem cell transplantation in children and adolescents with acute leukemia D. Prudnikau 1,* , Y. Isaykina 1 , J. Barouskaya 1 1 Belarussian Research Center for Pediatric Oncology, Hematology and Immunology, Minsk, Belarus Introduction: Relapse of the original malignancy following allogeneic hematopoietic stem cell transplantation (HSCT) remains the most frequent cause of treatment failure and mortality. More then 25% of patients with acute leukemia who received of HSC from HLA-identical sibling or unrelated donor had relapse. The purpose of this study was to evaluate the influence such parameter of graft as number of CD3+ cells on risk of posttransplantation relapse in children with acute lymphoblastic leukemia (ALL) and acute myeloblastic leukemia (AML). Material (or patients) and methods: We analyzed 102 patients at the age of 13,5 years (range, 1,2-29,7) with ALL and AML (66 patients with ALL and 36 with AML) who were underwent allogeneic HSCT between 1998 and 2015. All patients were in hematological remission, at the time of transplantation. 42% of recipients received graft from HLAidentical siblings and 58% from unrelated donors. Sixty patients (59%) were male, forty two of patients (41%)female. The Fine and Gray model and Gray's test were used for evaluation of cumulative incidences. Po .05 was regarded as significant. Results: Eighteen (18%) out of the 102 patients relapsed. The median duration from the time of transplantation to relapse was 232 days (range, 62-1826). The cumulative incidence of relapse (CIR) for the entire group of patients was 19,0 ± 4,1%. Relapse was not observed in patients received CD3+ cells ≥ 2,0x10 8 /kg in the graft, whereas CIR was 27,0 ± 6,0% when the number of CD3+ cells was o 2,0 x 10 8 /kg (P = 0,0123). In patients with ALL the level of CD3+ cells in the graft had a statistically significant impact on development the posttransplant relapse. The relapse was not observed in patients received CD3+ cells ≥ 2,0x10 8 /kg in the graft, whereas CIR was 28,2 ± 7,5% in the group of patients received dose of CD3+ cells o2,0x10 8 /kg (P = 0,0478). The level of CD3+ cells ≥ 2,0x10 8 /kg did not show statistical significance in probability of CIR in AML patients. This is perhaps due to the small number of cases. The probability of CIR in patients groups with ALL and AML was not statistically different (21,2 ± 5,3% and 15,0 ± 4,6%, respectively, P = ns). Conclusion: The level of СD3+ lymphocytes ≥ 2.0х10 8 /kg in the graft have a positive impact on reducing the probability of developing the post-transplant relapse in patients with ALL. Disclosure of Interest: None declared. On-demand plerixafor added to chemotherapy and granulocyte-colony stimulating factor for pbsc mobilization in multiple myeloma: final results and cost effectiveness analysis G. Milone 1,* , A. L. Di Marco 1 , A. Spadaro 1 , G. Avola 1 , M. G. Camuglia 1 , A. Cupri 1 , V. Zammit 1 , E. Martino 1 , C. Pirosa 1 , S. Leotta 1 , M. Morelli 2 , M. Martino 3 , A. Olivieri 4 1 Hematology, BMT UNIT, Catania, 2 Hematology and BMT, HS Raffaele, Milano, 3 Hematology, Azienda Ospedali Riuniti, Reggio Calabria, 4 Hematology, Università delle Marche, Ancona, Italy Introduction: We performed a multicenter phase II prospective study to evaluate the use of on-demand plerixafor (PLX) for patients multiple myeloma (MM) who received chemotherapy and granulocyte colony-stimulating factor (G-CSF) for mobilization of peripheral blood stem cells (PBSC). 111 MM patients have been analysed. Results were compared with those obtained in a retrospective control group. Material (or patients) and methods: The primary aim of the on demand prospective study study was to determine, in an unselected and consecutive population of patients diagnosed with MM the usefulness of PLX administered "on demand" in addition to mobilizing chemotherapy and G-CSF. Inclusion criteria for the prospective study were: diagnosis of symptomatic MM; age 18 to 70 years; first mobilization attempt. All MM patients in the prospective study were mobilized with cyclophosphamide (CTX) at a 4 g/m 2 , followed by G-CSF.Plerixafor was used according to a pre-established algorithm. Results were compared to a control group comprising MM patients mobilized using the same schedule but without the use of Plerixafor. The control group (n 183) was selected retrospectively. The mobilization schedule for the control group was CTX 4 g/m 2 and G-CSF 5 to 10 mcg/kg. In both groups, the costs of mobilizing therapy were estimated on the basis of costs reported in literature for a mobilization using chemotherapy and growth factor. The incremental cost effectiveness ratios (ICERs) was obtained by comparing results and costs in the 2 groups (prospective study and retrospective study) using the following formula: (cost in study 2cost in study 1) / (results in study 2results in study 1). Introduction: The number of CD34 + cells in the grafts infused has an effect on the engraftment and may also be associated with outcome of patients receiving autologous stem cell transplantation. In addition, the various lymphocyte subsets in the grafts may also be of considerable interest in regard to the post-transplant phase. In previous studies the number of the CD34 + cells collected has been reported to decrease after cryopreservation. However, the loss has been reported to vary a lotfrom about 10% up to almost 50%. To our knowledge there is no data on the effect of the cryopreservation on the various lymphocyte subsets in the autologous blood grafts. Material (or patients) and methods: Twenty-one patients were included in this single-center analysis. Twelve of the patients had non-Hodgkin lymphoma (NHL) and nine patients had multiple myeloma (MM). There were seven females and 14 males, the median age was 63 (38-73) years. After each apheresis session (Spectra Optia s , Terumo BCT), the flow cytometry analysis (FACSCanto, Beckton Dickinson) of the CD34 + cells and the T lymphocyte subsets of the fresh apheresis product was performed. To protect the cells from the stress/death caused by the cryopreservation, dimethyl sulfoxide (DMSO) was added into the apheresis product in a final concentration of 10%. Thereafter a 0.5 ml sample was taken from each apheresis product and frozen in liquid nitrogen freezer using a controlled rate freezing program at the same time as the apheresis product. Later, the samples were thawed at +37°C water bath and analyzed using the same methods as for the fresh samples. One apheresis product per patient was analyzed. Results: The median duration of cryopreservation was 82 days (14-202) and did not differ significantly according to disease (81 d in NHL vs. 54 d in MM, P = 0.454). The median number of CD34 + cells before cryopreservation was 2.3 x 10 6 /kg and after cryopreservation 1.9 x 10 6 /kg (P = 0.695). The median loss of CD34 + cells was 8.7% . In the entire patient population the median loss of CD3 + cells was 15.2%, CD3 + CD4 + cells 20.2%, CD3 + CD8 + cells 16.7% and NK cells 21.9%. The loss of CD3 + and CD3 + CD4 + lymphocytes as well as NK cells was greater in patients with MM (Table 1 ). There was no correlation with the used mobilization method or length of the cryopreservation and the number of CD34 + cells or T lymphocyte subsets measured after cryopreservation. The use of plerixafor (8 patients) was associated with agreater loss of NK cells (median 0 vs. 35%, P = 0.014). Conclusion: There seems to be important and highly variable loss of CD34 + cells during the process of cryopreservation. Also various T lymphocyte subsets seem to be affected by the cryopreservation in the same manner as the CD34 + cells, the median loss being even somewhat greater. In patients with MM the loss of T lymphocytes and CD3 + CD4 + lymphocytes was greater than in NHL. The loss of cells in the grafts during processing and freezing should be taken into consideration in performing aphereses and deciding the number of cells to infused. Disclosure of Interest: None declared. Immune characteristics of bone marrow harvests and peripheral blood products primed by recombinant human granulocyte colony-stimulating factor S. Gao 1 , L. Su 1,* , Y. Tan 2 , X. Liu 1 , Y. Zhao 1 , Y. Liu 1 , P. Yu 1 , N. Chen 1 , W. Han 1 1 Cancer center, the First Hospital of Jilin University, 2 Cancer center, Jilin University, Changchun, China Introduction: Allogeneic Hematopoietic stem cell transplantation (allo-HSCT) is the only curative method available for certain malignant hematologic diseases. HSCT most consists of bone marrow (BM) stem cells or peripheral blood (PB) stem cells that are recovered by leukapheresis after granulocyte colony-stimulating factor (G-CSF) mobilization. Previous studies reported some difference of the immune characteristics between BM and PB harvests primed by G-CSF. In this study, we will explore the difference of immune characteristics of BM and PB products in healthy related and unrelated donors. Material (or patients) and methods: Seventy-five healthy donors were retrospectively analyzed in this study. Recombinant human granulocyte colony-stimulating factor (rhG-CSF) was used to mobilize stem cells at a dose of 5~7.5 ug/kg/d. Blood routine and percentage of stem cells in PB were determined every day during mobilization. BM stem cells were collected on the 4 th day of treatment and PB stem cells were obtained on the 5 th and /or 6 th day by leukapheresis. Lymphocyte subsets were determined by flow cytometry, including in total lymphocytes (CD45 + SSC low ), total T lymphocytes (CD3 + ), helper T cells (CD3 + CD4 + ), inhibitory/ cytotoxic T cells (CD3 + CD8 + ), CD4/CD8 ratio, nature killer (NK) cells (CD3 -CD56 + ), NKT cells (CD3 + CD56 + ), B lymphocyte (CD3 + CD56 + ), active T cells (CD3 + HLA-DR + ), static T cells (CD3 + HLA-DR -), and regulatory T cells (CD4 + CD25 + CD127 -). Results: Sixty-five BM harvests and 104 peripheral blood leukapheresis products were collected in this study. The percentage of lymphocyte in PB harvests was 18.87% (7.36%, 34.04%), which was higher than BM products [1.50% (1.00%, 2.00%)] (u = 6.877, P40.001). T lymphocyte percentage in BM harvests was significantly lower than that in peripheral blood products (68.48 ± 8.42% versus 71.61 ± 7.23%; t = 2.484, P = 0.014). T helper cell (71.61 ± 7.23% versus 68.48 ± 8.42%; t = 2.484, P = 0.014) and B lymphocyte (15.76 ± 6.98% versus 12.94 ± 6.19%; t = 2.661, P = 0.009) were obviously higher in PB products than BM harvests. However, inhibitory/cytotoxic T cells, CD4/CD8 ratio, NK cells, NKT cells, active T cells, static T cells, and regulatory T cells were not different between BM harvests and PB products (P40.05). Conclusion: The different immune characteristics of BM and PM products primed by rhG-CSF may offer beneficial information for donor selection, conditioning regimen, and GVHD prophylaxis. Introduction: Autologous stem cell transplantation (ASCT) with cryopreserved peripheral blood stem cells (PBSC) is a therapeutic method mainly indicated in the treatment of patients with lymphoma, myeloma and solid tumors. Material (or patients) and methods: A total of 263 patients with hematological malignancies that underwent autologous transplantation during 2000 to 2015 at University Clinic of Hematology, Skopje, were included in the study. Cryopreservation of autologous grafts was provided using standard protocols with 5% and 10% final DMSO concentration. Freezing was computer controlled on Nikool Plus PC, Air Liquid (Cryogenie). Results: During 15 years of experience with cryopreservation of stem cells a total of 482 procedures were performed. The longest storage of cryopreserved PBSC was 7 years. We had two contaminated grafts with malignant cells and 6 unadministered grafts that are still stored under -198 o C. In 223 patients (85%) PBSC were used as stem cell source. Median graft volume was 149.62 ml (range 60-200), median number of cryobags 4.4 (range 2-8) , median days in liquid nitrogen 33,89 (range 2-330), median number of CD34+ cells was 2,34 x10(6)/ kg (range 0.1-5.7), median number of apheresis procedures was 2,15 (1-6), median amount od DMSO infused 20 ml . Time to engraftment was median 11 days (9-22). Statistical comparison between cryopreserved PBSC grafts and BM showed benefit for PBSC in the terms ( p40.01) of faster engraftment, less infective complications, less transfusion support and less hospital stay. In 226 patients (86%) DMSO related events were not registered during graft administration. In 36 patients (14%) mild to moderate DMSO related events were registered, as nausea in 29 patients (83.3%), vomitus in 18 patients (50%), tachycardia in 7 (20.8%), hematria in 5 patients (16%) and 2 patient (4.16%) with bradycardia, hypotension, fever and high temperature during graft infusion. Conclusion: At the end we can conclude that cryopreserved grafts as stem cell source for transplant procedures remain the "gold standard" as source for ASCT that helps the storage of more graft quantities for tandem transplants and at the same time provide planning of the transplant procedure and administration of more days' prettransplant conditioning chemotherapy. In conclusion, future clinical and translational studies will be needed to define and standardize the optimal cryopreservation techniques with optimal clinical outcomes and minimal clinical, environmental and financial adverse effects. References: · Fleming KK, Hubel A. Cryopreservation of hematopoietic and non-hematopoietic stem cellsTransfusApher Sci. 2006; 34 (3) Introduction: In low-weight children with cancer and healthy donor children, peripheral blood progenitor cells (PBPCs) have largely replaced bone marrow as source of autologous and allogeneic stem cells in part because of their relatively easy collection. However, there is a concern regarding medical, psychosocial and technical difficulties in small children. Material (or patients) and methods: We retrospectively analyzed peripheral blood stem cell apheresis in 33 collections. Twenty eight patients were with cancer (17 patients = Neuroblastoma, 4 patients = Retinoblastoma, 4 patients = Germ cell tumor, 1patient = Hepatoblstoma, 2 patient = Wilm's tumor) and 5 healthy children donors. The study was conducted between 2012 -2015. Peripheral stem cell apheresis was performed in the Mahak cancer children's hospital in a nice room for children where the patients stayed with their families. Patients were not routinely sedated. PBPC were collected by a COBE Spectra cell separator (COBE, Denver, CO, USA). Harvesting was performed after 5days mobilization. Results: Mean body weight was 11 kg (range 8 kg-15 kg) for a median age of 3.8 years (range 1.3-5 years). Mean duration of harvesting was 220 min (range 160 -274 min).Mean volume of stem cell collection was 140 ml (rang 110 ml -250 ml). The mean number of total nucleated cells collected was 4.1 x10 8 / kg (range 2.5-9.1 x 10 8 /kg recipients). No side effects occurred. Children didn't require an additional haematopoietic progenitor mobilization or additional apheresis in other day. PBSC collection was without transfusion in healthy donor children. Conclusion: PBSC collection may be difficult in small children owing to the large volume apheresis compared to the child's weight. Various problems, such as metabolic or haemodynamic disorders may be were seen. Peripheral Stem cell harvest can be performed in low-weight children under safe and effective conditions even when systematic priming by blood is avoided. Processing with increase of blood volume may to increase in the yield by recruiting progenitor cells. Disclosure of Interest: None declared. A new method to assess the viability of collected CD34+ cells before reinfusion: a prospective study in 50 Introduction: In the setting of peripheral blood stem cell transplantation (PBSCT),the strongest predictive factor for successful engraftment is the dose of reinfused CD34+ cells (per kg of body weight, not corrected for viability). Currently the amount of harvested CD34+ cells is assessed after completing the aphereses, before cryopreservation. However, such measurement does not account for the variable loss of viable CD34+ cells which occurs during freezing or thawing processes. Here we propose a novel method to measure the CD34+ cell viable content after cryopreservation, before reinfusion, through the use of a small bag (mini-bag, mB) which replicates the conditions to which CD34+ cells are subjected in the mother-bag (MB) to be reinfused. Material (or patients) and methods: We analyzed 93 samples from 63 aphereses collected from 50 patients undergoing PBSCT for hematological malignancies (22 multiple myeloma, 24 lymphoma, 4 acute leukemia). Total and viable CD34+ cells were quantified by flow cytometry; analyses were performed according to ISHAGE method with 7-AAD exclusion, before controlled freezing (ICE-CUBE14M system) from an aliquot of the final apheresis volume (FAV); after 1 week of storage for the mB; just prior to reinfusion from an aliquot of the MB. The mB contained 5 ml from FAV, were structurally similar to the MB and were cryopreserved together. Results were statistically compared by Student's t test and Pearson correlation. Univariate and multivariate linear regression analyses were used to identify variables influencing the viability of CD34+ cells. Results: Mean viability before cryopreservation was 99.7% (range 94-100%); median cell concentration in the FAV was 220x10 6 /ml for leukocytes (range 10-376)and 3x10 6 /ml for CD34+ cells (range 0.3-27) .After diluting and splitting the FAV, the median amount of CD34+ in each MB was 2.2x10 6 cells (range 0.2-25) and the median concentration of neutrophils was 19x10 6 /ml (range1-37). After thawing the mean viability of CD34+ cells was 76% (range 23-97%) for mB and 71% (range 28-99%) for MB (P = NS). The two viability values had good linear correlation with high statistical significance (Pearson's rho 0.59, p40.0001). We investigated factors affecting the viability of CD34+ cells according to the two methods: at univariate analyses, FAV and the bag leukocyte content were inversely correlated with viability both for mB and MB. In a multivariate model including all covariates with significance P40.2, FAV remained as the only significant factor for viability (both for mB and MB). FAV was highly correlated with both the leukocyte and neutrophil content of the apheresis, MB and mB. Conclusion: We showed that viability check of CD34+ cells after cryopreservation can be done with comparable results in thawed MB or mB. However, results of mB are available before reinfusion, allowing for a correction of the planned CD34+ dose to be reinfused, thus making it preferable as quality control of the freezing/thawing process. Leukocyte/neutrophil contamination of the collection (reflected by the size of FAV) significantly impacts CD34+ cells viability and should be minimized. Disclosure of Interest: None declared. Plerixafor use in prevention of mobilization failure of autologous stem cell: A single-center experience A. Birekul 1,* , A. Unal 1 , C. Unal 1 , L. Kaynar 1 , M. A. Baktir 2 , E. E. Turak 1 , S. Sivgin 1 , B. Eser 1 , M. Cetin 1 1 Haematology, 2 Physiology, Erciyes University School of Medicine, Kayseri, Turkey Introduction: Stem cells harvested from peripheral blood are used in diseases with severe risk for mortality such as multiple myeloma, Hodgkin lymphoma and non-Hodgkin lymphoma (NHL). In this study, we aimed to investigate mobilization success and factors associated in patient in whom Plerixafor, a CXCR-4 inhibitor, was added to mobilization regimens in case of mobilization failure. Material (or patients) and methods: The study included 36 patients in whom stem cell collection was failed or sufficient number of stem cells couldn't be reached despite CT plus G-CSF use as mobilization regimen at Apheresis Unit of Hematology Department of Erciyes University, Medicine School between 01.01.2012 and 09...2015. G-CSF plus Plerixafor (0.24 mg/kg) was given to these patients in whom insufficient number of stem cell was collected or CD34+ count was o20/mL. CBC and CD34+ count were repeated on the next day and stem cells were harvested. Results: Of the donors, there was AML in 1 (2.7%), ovarian carcinoma in 1 (2.7%), testicular tumor in 6 (16.62%), NHL in 8 (22.2%), Hodgkin lymphoma in 9 (24.93%) and multiple myeloma in 11 (30.4%). Median age was 44 years (min-max: 16-71 years) and median weight was 72 kg (min-max: 49-120 kg) in donors. Median pretreatment CD34+ cell count was 18.24/μL (min-max: 2.5-152/μL). Results of pretreatment CBC and biochemistry test are presented in the tables. Median post-treatment CD34+ cell volume was 219 mL (minmax: 116-436 mL). A significant positive correlation was detected between product volume and pretreatment WBC, MNC and LDH values (WBC rho:0.33; p:0.49, MNC rho:0.33; p:0.49, LDH rho:0.37; p:0.037) . It was seen that duration of procedure and BMI (rho:0.473; p:0.04) were positively correlated with product volume. A negative correlation was detected between total pretreatment totalWBC (rho:0.380; p:0.022) level and product volume. A significant positive correlation was detected between percent CD34+ count and pretreatment values of Glucose (rho:0.338; p:0.047) and totalWBC (rho:0.380; p:0.022)levels.When association between product volume and devices used to collect stem cell, it was seen that amount of product collected was smaller in Amicus Apheresis device when compared to Spectra Optia Apheresis device. It was shown that sufficient CD34+ cell count was achieved and sufficient amount of product was harvested by using Plerixafor in 36 of 36 patients in whom sufficient number of CD34+ cell couldn't be reached or sufficient number of CD34+ cell was achieved but sufficient amount of product couldn't be collected despite Chemotherapy plus G-CSF. Conclusion: In our study, it was seen that adding Plerixafor to G-CSF improves mobilization success when G-CSF plus chemotherapy was failed to achieve sufficient number of stem cell in patients scheduled to autologous stem cell transplantation. It was also concluded that BMI, WBC and MNC levels and duration of procedure had positive effect on volume of stem cell harvested. It was found that gender had no effect on stem cell collection but type of device was important in stem cell collection. Disclosure of Interest: None declared. Comparison of the viability, presence of adverse effects and reconstitution of hematopoiesis in hematopoietic stem cell transplantation of peripheral blood using dimethyl sulfoxide 10% vs. 5% B. L. Acosta Maldonado 1,* , L. Rivera 1 , G. Barranco-Lampon 1 , M. Alfaro 1 , L. Valero-Saldaña 1 1 Hematology, INSTITUTO NACIONAL DE CANCEROLOGIA, Mexico city, Mexico Introduction: Dimethyl sulfoxide (DMSO) is essential for the preservation of liquid nitrogen-frozen stem cells, the majority of centers continue to use 10% DMSO but is associated with toxicity in the transplant recipient. the reduction of the concentration of DMSO has diminished the presence of adverse effects but but the effect in hematology recovery (HR) and viability of Haemathopoietic stem cell post thawing (vHSC) still are unknown. Material (or patients) and methods: Retrospective study, we described 108 patients in Instituto Nacional de Cancerología, Mexico City between 2012 to 2015 who received autologous (Auto) and allogenic (Allo) of peripheral blood progenitor cells trasnplant (PBPCs), cryopreserved both the DMSO 10% and 5% and then be compared.Patients who had complete data for analysis were selected. Adverse effects were evaluated according to the criteria for adverse events version 4.0.; engrafment and vHSC also they assessed comparatively. Results: Were evaluated 108 patients, 65.75% Leukemia 25% (n = 27), LH 13% (n = 14), NHL 23% (n = 25) % MM 25% (n = 27), LGC 6.5% (n = 7), SMD 2.8% (n = 3), germ cell tumors 2.8% (n = 3), aplastic anemia 1% (n = 1) and blast dendritic cell neoplasia 1% (n = 1) of which 65.7% (71) received auto PBPCs transplant and 34.3% (n = 37) alloPBPCs transplant. The 64.8% (n = 70) received PBPCs cryopreserved DMSO 10% vs 35.2% (n = 38) DMSO 5%. viability of Haemathopoietic stem cell post thawing (vHSC) DMSO 5% was 94.9% vs 93.7% DMSO10% (P = 0.222) was observed. Comparing the HR parameters it was observed that leukocytes ≥ 1000 were reached at day 12 vs 10 (P = 0.005), neutrophils4500 11 vs. 10 per day (P = 0.029) and plate-lets450,000 at day 14 vs 13 (P = 0.031) DMSO10% vs. DMSO5% respectively. Average infused volume was 408.44 ml and 274.5 ml to DMSO10% and DMSO5% respectively (P40.001). Presence of adverse events 34% vs 37% with DMSO10 and DMSO5% respectively (P = 0.790). The recovery characteristics and hematology in autologous allogeneic relative to the DMSO concentration shown in Table 1 . Introduction: The prognosis for steroid-refractory cGVHD remains poor. ECP has been used with varying degrees of success to treat steroid-refractory cGVHD, and it is viewed as a well-tolerated intervention with steroid-sparing effect [1] [2] . The aim of this study was to evaluate the effectiveness of ECP and to determine which subset of patients with cGVHD would benefit most from treatment. Material (or patients) and methods: 34 patients who underwent HSCT and developed moderate (n = 7) or severe (n = 27) steroid-refractory cGVHD were included. Patients were treated for cGVHD with ECP in the period september 1998 to November 2011. Chronic GVHD was retrospectively categorized according to the NIH consensus criteria [3] . ECP procedures were performed on two consecutive days every week until a clinical response was achieved, and then tapered by slowly extending the treatment intervals. Treatment response was categorized as being a complete response (CR) if full resolution of cGVHD was observed. If improvement in cGVHD was observed with a decrease of ≥ 1 point on the organspecific NIH score, this was defined as a partial response (PR). Stable disease (SD) was defined as being when no change in cGVHD activity was observed and progressive disease (PD) was defined as being when cGVHD activity progressed during ECP treatment. Results: Overall survival for all patients was 82% at one year and 58% at 5 years after the start of ECP. ECP was initiated in median 161 (range 10-1421) days after the onset of cGVHD. Eighteen patients (53%) had PR, 5 patients (15%) had CR, 8 (23%) had SD and three patients (9%) developed PD.Patients with CR/PR were predominantly those with cGVHD of the skin and oral mucosa with an organ-specific score (OSS) of 1-3 [3] . Patients with GI, liver and pulmonary cGVHD were most frequent in the SD/PD group. In the patients with PD, the most common organ involvement was the lungs (OSS 2-3). In the CR/PR group, the corticosteroid dose was significantly lower 8 weeks after ECP than at the start ( p40.001) and at 6 months after ECP a further decrease was seen (P = 0.02). Higher levels of Albumin and Platelets were observed in patients with CR/PR as compared to SD/PD (mean difference 10 grams/L and 100x10 9 , P40.01). Conclusion: In conclusion, we found that patients with severe cGVHD involving primarily the skin followed by oral involvement, had the best response rates to ECP treatment. ECP is a safe treatment option for patients with steroid-refractory cGVHD. A good individual response against liver cGVHD was noted, however, no convincing response was seen in patients with severe pulmonary involvement. Patients who responded to the ECP treatment had a significant increase in platelet and albumin levels. Our results suggest that ECP treatment is safe and effective for patients with skin, oral and liver cGVHD. T-cell responses were tested by proliferation or IFNγ ELISA. Results: Alloreactivity was seen in total APH and NTF, but not in TF (7/7 cases). NTF responses were higher than APH responses in 5/7 cases due to higher alloprecursor frequency in NTF (median 66% αβ T-cells vs. 23% in APH). To test responses to protein antigens, TF cells, that contain professional APC (mono-macrophages MM, DC) were added to NTF. Proliferation and cytokine secretion were remarkable also in NTF, compared to APH (9/9). NTF alone contained functional B-cells that presented protein antigens as efficiently as APC from TF (MM, DC) (7/8 cases) , implying that depleted B-cells exert antigen processing in addition to antigen presentation. A high background in APH was observed when tested by proliferation (5/9 cases), but not by IFNγ. This was attributed to proliferation of cells other than αβ T-cells, which would have also produced IFNγ had they been activated. Spontaneous proliferation was confined to cells in TF which contained, in addition to gd T-cells and NK cells, also CD34 and other hematopoietic precursors likely accounting for baseline proliferation. Conclusion: In spite of extensive manipulation (apheretic procedure, centrifugation, incubation with antibodies and magnetic beads, magnetic column passage), αβ T-cells and B-cells in NTF were functionally intact. The anti-CD19 Ab on B-cells and the anti-αβ Ab on T-cells did not interfere with antigen presentation and with antigen response. We hypothesize that anti-αβ Ab either do not engage all TCR molecules or that anti-αβ Ab/TCR complexes are internalized during the incubation phase with de novo expression of free TCR molecules. These data suggest that NTF could be used as DLI surrogates. Finally, NTF cells could be further manipulated to i. remove anti-αβ Ab/bead complexes from T-cells; ii. eliminate B-cells with fresh anti-CD19 magnetic beads, after they have exerted the APC function; iii. deplete alloreactive precursors; iv. positively select pathogen-specific T-cells; v. generate pathogen-specific T-cell lines; vi. introduce suicide genes. Work is in progress to test these possibilities. Disclosure of Interest: None declared. Reduction of DMSO concentration in cryopreservation mixture from 10% to 7,5% and 5% has no impact on engraftment after auto-HSC transplantation 97%, range 90-99% in 5% DMSO; P = 0.56). The median time to leukocyte recovery, defined as the first day with WBC count exceeding 1.0 × 10 9 /L was 10 days in all groups (ranges: 9-12 for 10% DMSO; 7-11 for 7.5% DMSO; and 9-12 for 5% DMSO; P = 0.59). Similar results were obtained in case of neutrophil recoveryin all arms, the median day, when the ANC exceeded 0.5 × 10 9 /L, was 10 (ranges: 9-12; 9-11 and 9-12, respectively; P = 0,48). The day platelets level were greater than or equal to 20 × 10 9 /L (sustained without transfusion within 7 days) was similar in all groups: medians were 15 days in 10%, 7.5% and 5% DMSO (ranges: 8-20; 0-18; 0-24; P = 0.18 (3) (4) (5) (6) (7) (8) (9) (10) , blood volume/ cycle 1,000 mL (600-1,400), and blood volume processed 2.0 times (1.1-3.5) . Procedure performance was evaluated on collected products CD34+ CE using the calculation 100X (Total CD34+ product yield)/(Volume of TBV processed X CD34+/μl donors' blood). Benchmark CE was determined as the median CE of collections using same separator. Predicted absolute CD34+ was calculated using the formula (TBV to be processd (L) X Benchmark CE X CD34+(μl))/Body weight (Kg) x1000. The Donors safety was evaluated by the number of reactions during and after aphaeresis, and hemoglobin and platelet loss measured after the procedure. Predicted CD34+ counts/kg were plotted against corresponding observed counts for all collections and the scatter plots assessed for linear relationship between predicted and observed. Linear regression analyses were performed and the linear correlation coefficients (r-values) were calculated. The same analyses were performed on pediatric and adult collection sub-groups. Results: The CD34+ collection efficiency median was 67.2%. Plotted predicted versus actual CD34+cells/kg for all collections had an r-value of 0.889, which demonstrates positive linear correlation, same results were observed for pediatric and adult donors, which had strong r-values of 0.917 and 0.865 respectively, CD34+ yield was 8.6 ± 6.5 X 10E6/Kg. Donor hemoglobin and platelet loss were minimal (7.0 ± 11.7% and 14.0 ± 13% respectively). Adverse events with Amicus were mild citrate related reactions in 3 out of 20 runs which were resolved by intravenous Calcium administration. Conclusion: Regardless of the heterogeneity of the donors in our study, results demonstrated optimal efficiency using the Amicus separator for HPC aphaeresis. The device allows 1) full automated control of collection with minimum manual adjustment of parameters during the procedure, 2) Ability to predict collection of CD34+ cells/Kg and optimize the procedure to patients, and 3) Minimum donor risks particularly low weight children Introduction: American Society of Clinical Oncology guidelines recommend the use of growth factor after high-dose chemotherapy (HDC) and peripheral blood stem cell (PBSC) support. This retrospective study aims to compare the efficacy of pegfilgrastim (PEG) compared with filgrastim (FIL) after HDC. Material (or patients) and methods: We collected data on 195 consecutive adult patients who received an autotransplant (Myeloma, lymphoma and others) between January 2004 and December 2014 at two tertiary care centres. The primary end point was the duration of neutropenia both in terms of days to reach an ANC 40.5 × 10 9 /l. Filgrastim was given to 110 patients and Pegfilgrastim was given to 85 patients. Results: Time to engraftment, defined as the time to reach an ANC of 0.5 x 10 9 / L on 2 consecutive days after the day of auto-SCT, was 12.6 days with Filgrastim compared with 12.1 days with Pegfilgrastim (P = 0.126). When comparing the total days of severe neutropenia (WBC o0.1 x109/L), there were 5.5 days of severe neutropenia with filgrastim compared with 5.8 days with pegfilgrastim (P = 0.7). The duration of febrile neutropenia was an average of 5.3 days with Filgrastim and 4.6 days with Pegfilgrastim (P = 0.029). The total number of antibiotic days was shorter for the patients who received Pegfilgrastim, being 11.08 days with Pegfilgrastim and 12.1 days with Filgrastim (P = 0.184). The average cost savings per person in terms of number of days of hospitalization and number of days of total parental nutrition was 582 Rs, (P = 0.512) and 6003 Rs (P = 0.018) respectively in favour of Pegfilgrastim arm. Material (or patients) and methods: Patients with hematologic malignancies who underwent their first AHSCT using an UCB or URD graft between January 2010 and June 2015 were S140 included. The overall survival (OS), graft-versus-host-disease (GVHD)-free relapse-free survival (GRFS), relapse, non-relapse mortality (NRM), neutrophil recovery, grade II-IV acute GVHD (aGVHD), and moderate to severe chronic GVHD (cGVHD) were analyzed. GRFS events were defined as grade III-IV aGVHD, cGVHD requiring systemic immunosuppressive treatment, disease relapse, or death following AHSCT. S141 apheresis process and in final apheresis product by flow cytometry using ISHAGE protocol in FC 500 (Beckman Coulter) s cytometer. CE of white blood cell (WBC), total mononuclear cells (TMNC), CD34+ cells and platelet loss were calculated as: CE(%) = [collected cells* product volume /Pre-apheresis cell* (processed volume-ACD used)]*100. Platelet loss(%) = [Pre-apheresis count-Post-apheresis count/ Pre-apheresis count]*100. We recorded all adverse effects during the apheresis procedure. We used the Mann-Whitney test for the comparing analyzed. Results: CBC pre and post the apheresis process were comparable in both devices, except in the number of platelets post-apheresis, which was significantly higher in the OPTIA group. The hemoglobin, WBC, TMNC, CD34+and platelet count in the apheresis product were similar in both groups. There were no differences in processing volume, procedure time and amount of ACD anticoagulant used. Introduction: Optimizing donor selection and choice of cell source for hematopoietic cell transplantation (HCT) continues to evolve. Study of direct comparisons from randomized controlled trials (RCTs) permits the study of network geometry of available interventions and facilitates identification of potential indirect comparisons that can inform best practices and drive cost-effective transplant care towards optimized patient outcomes. We performed a scoping review of RCTs addressing the source of cells used in allogeneic transplant and choice of donors and created an evidence network to study the geometry of direct and indirect comparisons that shape the current evidence base supporting the selection of optimal stem cell products. Material (or patients) and methods: A scoping review addressing all aspects of care in allogeneic HCT was performed. RCTs addressing the source of cells used in allogeneic HCT and choice of donors were included. We searched Medline and EMBASE (1946 to May 2015) and extracted information regarding the donor and source of cells from each eligible study. Results: Seventeen RCTs encompassing 2176 patients were identified. Eight reports (1015 patients) compared clinical outcomes following the use of peripheral blood progenitor cells (PBPCs) with bone marrow (BM) from matched related donors. One study (551 patients) compared PBPCs and BM from matched unrelated donors. The remaining studies examined the impact of T cell depletion in BM grafts (1 study) , the method of BM infusion (1 study), the addition of PBPCs to BM (1 study), G-CSF primed BM (2 studies), surface molecule-based selection and/or depletion (2 studies), and the optimal number of units for pediatric cord blood transplantation (1 study). There were no identified published RCTs that compared different types of donors (i.e. related vs. unrelated). Evidence network geometry was analyzed to identify opportunities for novel indirect comparisons for various transplant outcomes using different cell sources. No comparisons can be made using evidence from RCTs to evaluate different donor options but opportunities to expand the network were identified. Following a search of ongoing registered RCTs with the Clinical Trials Network of the Centre for International Blood and Marrow Transplant Research and a search of prospective trials affiliated with the European Blood and Marrow Transplantation group, one actively enrolling RCT comparing double cord blood transplant with haploidentical donor transplant was identified. Conclusion: Analysis of network geometry from RCTs addressing source of cells used in HCT provides the basis for novel indirect comparisons for matched related donor transplants. RCTs comparing different donor choices will leverage the existing evidence base to expand the network. Future trials should aim to link within the existing geometry to leverage the current evidence base most effectively. Disclosure of Interest: None declared. Effect of single HLA-mismatch on outcomes following allogeneic haematopoietic progenitor cell transplantation for haematological malignancies C. Hoong 1 , F. Lim 1,* , W. Hwang 1 , Y. C. Linn 1 , C. Phipps 1 , S. Gopalakrishnan 1 , Y. Loh 1 , Y. T. Goh 1 , J. J. Lee 1 , A. Ho 1 1 Department of Haematology, Singapore General Hospital, Singapore, Singapore Introduction: Allogeneic haematopoietic stem cell transplantation remains the only potentially curative treatment option for patients with a range of haematological malignancies. However, the availability of a HLA fully compatible sibling donor (SD) is restricted to approximately 20-30% of potential recipients, with others having to rely upon Volunteer Unrelated Donors (VUD) or alternative stem cell sources. On average approximately 40% of our VUD searches lead to the procurement of a donor for transplant, in our predominantly South-East Asian population. Material (or patients) and methods: To determine if a single HLA mismatch affected the outcome of transplantation in our centre, we conducted a retrospective review of recipients who received allografts for all haematological malignancies in the last decade, focusing on the outcomes of recipients who received grafts from single HLA-mismatched donors vs fullycompatible (8:8) donors. Results: A total of 339 patients received allografts for haematological malignancies during this period. The disease indications broadly included AML (n = 152), ALL (n = 65), MDS (n = 34), Lymphoma (n = 33), CML(n = 19), AA (n = 13), MPD (n = 8), ABL (n = 9),Myeloma(n = 6). Preparative regimens were intended to myeloablative in 43% and reduced-intensity in 55%. All VUD and 7:8 SD graft recipients received ATG (Thymoglobulin™) at a dose of 4.5 mg/kg. All donor/recipient pairs were typed at high resolution. There were 230 HLA 8:8 compatible SD (median recipient age = 47 years (range 16-68)), 84 HLA 8:8 VUD (median recipient age = 43 years (range 16-66)), 6 HLA 7:8 SD (median recipient age = 38 years (range 16-56)) and 19 HLA 7:8 VUD (median age = 37 years (range 17-66)). 4 (67%) of SD grafts were mismatched at HLA-A, and 2 (33%) at HLA-DR. 5(26.3%) of VUD grafts were mismatched at HLA-A, 5(26.3%) at HLA-B, and 9(47.3%) at HLA-C. Donor selection hierarchy-first choice 8:8 SD, 8:8 VUD, followed by 7:8 VUD or 7:8 SD. All recipients who received 7:8 SD grafts did not have any potential 7:8 VUD. With a median follow-up of 50 months, the 3 year overall survival (OS) for 8:8 SD recipients was approximately 55%; 8:8 VUD, 52%; 7:8 SD, 33% and 7:8 VUD 45% (P = ns). The 3 year disease free survival (DFS) was 49%, 49%, 0% and 48% S142 respectively. The 3 year non-relapse mortality (NRM) was 21%, 23%, 34% and 34% respectively (P = ns). Conclusion: There does not appear to be any statistically significant difference in OS, DFS and TRM between 8:8 SD, 8:8VUD and 7:8VUD graft recipients. The small number of 7:8 SD recipients make valid analysis of this group difficult, although OS and TRM appear similar to the other groups. All 7:8 SD recipients had high risk disease (2ALL, 2 AML, 1 NHL, 1 MF), with 6 relapses/progressive disease in this cohort. A larger series permitting a multivariate analysis of the groups should clarify the role of single HLA allele mismatched donors in allogeneic transplants for haematological malignancies. Disclosure of Interest: None declared. The effect of stem cell viability and storage time S. Solmaz Medeni 1 , G. Özsan 2,* , C. Acar 2 , Ö. Sevindik 3 , D. Introduction: Cryopreservation is needed for hematopoietic stem cell viability and functional integrity when they are dissolved and for infusion with minimum toxicity. To evaluate viability of stem cells, trypan methylen blue staining under light microscopy or flowcytometry are used. We want to evaluate stem cell viability at 3. and 5. year after thawing, relationship between CD34+ cell level, stem cell viability and storage time and intend to contribute to the literature. Material (or patients) and methods: Stem cells product of 61 patients who are ex at study time were evaluated. Demograhic features, CD34+ cell levels at the time it was collected and after dissolution with trypan blue, flow cytometry in 7AAD CD 34 viability test and storage time were analyzed. Results: Median age was 44 with M/F ratio as 62.3/37.7. Mean CD34+ cell level at product at the collected time was 1977 μl (98-7457), but 170 μl (6-1345) after disolved. Viability was 21.46% (4-78) when they are disolved. Mean product storage period was 67.3 months (34-138). There was a inverse correlation between storage time and CD34+ cell level (at beginning and at the end of storage time). Viability was lower at long-term stored product.(p:0.004). In addition, a significant relationship was found between the product's initial post-thaw CD34 value. Correlations were found between pre-apheresis leukocyte value and the CD34 value of the dissolved product. The patient were grouped according to stored time of their product as o 48 months (n:18), 48-60 months (n:8), 460 months (n:35). CD34+ cell viability was higher in o48 months with respect to 460 months. The most lower viability was seen at 460 months group,this was statistically significant. CD34+ cell level also very lower at 460 months group (this was statistical significance). CD34+ cell viability with trypan blue also very lower at 460 months group without statistical significance. Conclusion: In conclusion; viability of stem cells and amount of CD34+ cells in the product are affected by storage time period and especially storage time longer than 60 months mostly affect the viability of stem cell in the product detected with both trypan blue staining and flowcytometric analysis. Our study results were presented for contribution to literature. Disclosure of Interest: None declared. Viability and cellular recovery of cord blood units used in unrelated hematopoetic stem cell transplantation Introduction: Umbilical cord blood (CB) are increasingly used for allogeneic transplantation as an alternative source to bone marrow or peripheral stem cell for patients who lack a human leucocyte antigen (HLA)-matched donor. Nevertheless, graft failure remains a major problem. Cell dose is one of the main determinant factor for selection of the unit. Cell doses available for selection of the unit are the ones evaluated before cryopreservation. Processing of the graft either with cryopreservation and/or thawing inevitably lead to cell loss but the data showing to what extent is very limited. In addition, their quality particularly related to viability remains uncertain. To evaluate the extent of the cell loss after thawing and the effect of preservation duration on cell loss and the viability, we retrospectively analyzed our experience in a cohort of patients receiving unrelated CBT. Material (or patients) and methods: A hundred seven CB units used for unrelated CBT in 106 patients were included in the study. One of the patient was transplanted with double CB unit. All units were obtained from allogeneic CB banks. All CB units were thawed in a 37°C water bath. None of the units except one were washed after thawing. Samples were taken from each CB unit immediately after thawing at the time of infusion. Total nucleated count (TNC) was obtained with hematology analyser. For enumaration of CD 34 cells, BD Procount Progenitor Cell Enumaration kit (BD Bio-Sciences-Sanjose,USA) was used. Flow cytometry analysis was performed with a flow cytometer (FACSCalibur,BD, Biosciences). Viability assays were performed using viability dye 7-amino actinomycin D (7-AAD) by flow cytometry in TNC. Cell recoveries were computed from values provided by each bank before freezing and values obtained in our laboratory after thawing. Results: Median number of declared TNC per kilogram of recipient body weight was 18.3 (range:2.4-69.0)x10 7 /kg. After thawing, the median TNC dose decreased to 11.8 (range:1.5-50.2) x10 7 /kg with median cell viability at 86% (range: 15-100). The median recovery of TNC was %71.3 (range:9%4170%). The median CD34+ cell count declared by the banks per kilogram body weight of recipient was 8.1x10 5 (range:0.48-48.3) . The median postthawed CD 34 cell count and recovery were 4.6x105/kg (range: 0.07-29.2) and 59% (range: 2.7-292), respectively. The median storage duration of the CB units in liquid nitrogen was 42 months (range: 12-103 months). Correlation analysis did not show any significant correlation between CB unit storage duration and cellular recoveries including both TNC and CD34. However, viability was found to negatively correlate with storage duration (P = 0.001 r = -0.31). The comparison of the cell counts of the cord blood units that were engrafted and rejected yield that both declerad and postthawed TNC and CD 34 cell counts and also viability are higher in engrafted units. However celluler recoveries are not different. Conclusion: Cryopreservation and thawing of the cord blood units results in a substantial loss of cells with TNC loss approaching to %30 and CD34 loss approaching to %40. Although cell loss does not correlate with storage duration, viability is found to negatively correlate. The extent of cell loss and also decrease in viability correlating with storage duration should be kept in mind especially CB units with critical cell counts are choosen. Introduction: As bone marrow transplantation (BMT) increases, the availability of suitable donors becomes critical. Finding a suitable match donor requires a large donor pool. We hypothesized that voluntary blood donors may can be a good resource for this large donor pool need. The aim of this study is to explore the knowledge and motivations of voluntary blood donors toward hematopoietic stem cell donation also, and to determine predictors of hematopoietic stem cell donation motivation among them. Material (or patients) and methods: The knowledge and motivations of voluntary blood donors toward hematopoietic stem cell donation were surveyed using a questionnaire with 42 items. The sample consisted of 100 voluntary blood donors that admitted to Gulhane Military Medical Academy Blood Training Center and Blood Bank Department, between May 1st and September 1st, 2015. Donors filled in a self-administered questionnaire including the measures of demographic information, knowledge and motivations about hematopoietic stem cell donation. Data was analyzed using SPSS Ver.22. Results: The survey responders comprised 17% female and 83% male. The mean age was 33.5 ± 6.8. 47% of the reponders' said "they would" and 53% said that "they would not" be volunteer for bone marrow transplantation. The most reason for having willingness to be a hematopoietic stem cell donor was the aim of saving lives (89%). The most reason of would not volunteers' was the decision as having negative impact on donors' health (75.4%). The rates of routine blood donors and would volunteers were both significantly higher in the higher (411 years) education group ( p4.001, P4.001). The majority of responders were aware that 'a donor can refuse to donate anytime they want' (n = 97). Most of the participants (95%) answered as it is wrong to the question "Donation process can be finished almost always without any problem.". Conclusion: The motivation of the voluntary blood donors for also being a donor in hematopoietic stem cell transplantation was very low. Results show that donors have an unfavourable perception of the hematopoietic stem cell transplantation compared to voluntary blood donation. A key factor in determining an unfavourable perception of hematopoietic stem cell transplantation was the information level about transplantation; in fact we found by linear regression that the information level affects the perception of safety and donation importance. These differences of perception show that there are prejudices about hematopoietic stem cell transplantation which are caused by a lack of Information. By statistical analysis, improving knowledge of the voluntary blood donors about being a hematopoietic stem cell transplantation donor also, seems to be the the most effective way to convince them. Disclosure of Interest: None declared. The impact of matched vs. mismatched nationality among donor and recipients in allogeneic stem cell transplantation: A retrospective multi-institutional study in Korea J. Youk 1,* , Y. Introduction: Recently, the number of allogeneic stem cell transplantation (ASCT) from foreign countries has been increased because of a shortage of unrelated stem cell donors in Korea. Although a number of studies have been performed to show the prognostic factors for ASCT, the prognostic impact of nationality matching status has not been evaluated before. Thus, we tried to show the impact of donor/recipient nationality matching status in Korea, in which consists of racially very homogeneous people. Material (or patients) and methods: A total of 855 patients (No. of matched nationality: 972, No. of mismatched nationality: 117) who received unrelated ASCT in five hospitals between January 1st, 2005 and April 30 th , 2015 were collected from Korea Marrow Donor Program. We retrospectively reviewed electrical medical records in each hospital. In order to control potential confounders, a propensity score using logistic regression was developed. The propensity score was composed of disease category (severe aplastic anemia, acute leukemia, lymphoma and others), HLA matching status, the number of CD34 stem cells and disease status (e.g. complete remission) right before ASCT. Then, 3:1 matching was created on the propensity scores. Overall survival (OS) and relapse free survival (RFS) were calculated using Kaplan-Meier method. Chi-square test was applied to compare acute and chronic graft versus host disease (GVHD) incidence between two groups. Results: A total of 332 patients were analyzed (Male/ Female = 209/123, median age = 36.5 years old). The number of patients with aplastic anemia, acute leukemia, lymphoma and others was 35, 197, 40 and 60. The details of donor sources were as below; Taiwan (n = 26), Japan (n = 25), Unites States of America (n = 17), China (n = 15), Germany (n = 3) and Australia (n = 1). Between two groups, major variables such as age, sex, stem cell sources (bone marrow or peripheral blood), disease status, HLA matching status, the number of CD34 and disease category were not statistically different. The median OS of matched and mismatched nationality was 24.9 [8.6, 41.2] months and 11.8 [5.4, 18.2] months, respectively (P = 0.284). The median RFS was not reached in both groups and the mean RFS was slightly longer in mismatched nationality group (144 (119-169) months versus 93 (83-103) months; P = 0.055; Figure 1 ). Chronic GVHD was not different between two groups (35.1% versus 34.5%; P = 0.584). The incidence of acute GVHD was 39.2% in matched nationality group and 41.4% in mismatched nationality group (P = 0.937). Conclusion: In this study, donor/recipient nationality mismatching status was not an inferior prognostic factor for overall survival and GVHD in allogeneic stem cell transplantation. Interestingly, RFS tended to be longer in mismatched nationality group. Thus, unrelated allogeneic donor stem cells from foreign countries could be a good therapeutic option for ASCT candidates who don't have appropriate HLA-matching donors in their countries. Introduction: The success of unrelated allogeneic bone marrow transplantation greatly depends on HLA antigen matching. It becomes evident that the certain HLA loci can be responsible for specific posttransplant complications development. In this regard it seems useful to analyze the possible role of HLA-А*01:01/HLA-А*03:01 genotype in the development serious immunologic complications after allo-HSCT. Material (or patients) and methods: We analyzed 72 patients with hematologic malignancies who received an allogeneic stem cell transplant from unrelated donors in our center between March 2012 and August 2015. In the majority of cases, peripheral blood was the source of stem cells -48 patients (67%). 48 patients were transplanted with an HSCT from a fully matched (10 of 10) at the allele level for the HLA-A, -B, -C, -DRB1, -DQB1 loci unrelated donor, whether 24 patients were transplanted from a partly matched donor (7 patients had disparities in A locus, 6 -in B locus, 5 -in C locus, 3 -in DRB1 locus, 2 -DQB1, and one patient had disparities in С and DQB1. 6 patients had HLA-А*01:01/HLA-А*03:01 genotype (group 1) and had highly negative results of allo-HSCT in comparison with the rest ones (group 2). Results: After the transplantation 5 of 6 patients in the group 1 (83.3%) developed acute graft-versus-host disease (GVHD) grade 2-4 and 4 of 6 (66.7%) were diagnosed with acute intestinal GVHD resistant to systemic and topical steroids. Chronic GVHD was observed in 5 patients of 6 in the group 1 (83.3%). All patients in this group died during the first year after allo-BMT due to complications. In the 2 group 28 patients of 66 (42%) developed acute GVHD grade 2-4, 11 patients (16.7%) had resistant form of GVHD. Chronic GVHD was observed in 26 patients of 66 in the group 1 (39,4%). 1-year mortality was 9% (6 patients) in this group, 42 patients (63.6%) are still alive at the moment of analysis (December 2015). Group 1 (n = 6) Introduction: The low content of hematopoietic progenitor cells (HPCs) is the main drawback of cord blood (CB) transplantation, leading to a higher incidence of engraftment failure and transplantation-related mortality than in bone marrow (BM) transplantation [1] . Various approaches have been set to increase the dose of HPCs in the CB graft [2] . Among these, the "ex vivo expansion" over an adherent layer of mesenchymal stromal cells (MSCs), closely mimics physiologic hematopoiesis and has been set also for clinical use [2, 3] . Considering that cellular composition of bone marrow (BM) hematopoietic niche consists, in addition to MScs, also of endothelial cells [3] , we evaluated the expansion of CB HPCs cultured in presence of adherent MSC or endothelial cell layers. Material (or patients) and methods: MSCs were obtained from healthy BM donors; endothelial cells (grown from endothelial colony forming cells, ECFCs) were obtained from CB, as previously described [4] . CB CD34+ cells were isolated by immunomagnetic method (Stemcell Technologies Ottawa, Canada) and seeded at 2X10e4 cells/ml over MSCs or ECFCs [5] in serum free medium (Stem Span, Stemcell Technologies) containing SCF, GM-CSF, Flt3-ligand (all 100ng/ml) and Tpo (50 ng/ml) (all growth factors purchased from Miltenyi Biotec, Germany). At baseline an after 7 days of cultures, adherent and non-adherent cells were recovered, counted and plated at 1x10e3 /ml in 35mm culture dish in MethoCult classic medium (Stemcell Technologies). Results: Both types of 7-day cultures supported a significant expansion of the number of cells, without significant differences between ECFC and MSC cultures (795.000+147.560 versus 672.500+113.250, P = 0250; Figure 1 ). The clonogenic cell output, however was significantly higher in MSC cultures. In particular, BFU-E and most immature CFU-GEMM were more abundantly recovered from MSC cultures. Haploidentical T-cell replete hematopoietic transplantation conditioned with thiotepa-busulfanfludarabine (TBF) followed by high-dose cyclophosphamide and tacrolimus as graft versus host disease prophylaxis in 33 patients with myeloid disease A. Esquirol 1,* , M. J. Pascual 2 , M. Ortiz 2 , J. L. Piñana 3 , C. Ferra 4 , G. Irene 1 , I. Vilades 2 , S. Brunet 1 , R. Martino 1 , J. Sierra 1 1 Hematology department, Hospital de la Santa Creu i Sant Pau, Barcelona, 2 Hematology department, Hospital Regional Universitario, Malaga, 3 Hematology department, Hospital clinico universitario, Valencia, 4 Hematology department, Hospital Universitari Germans Trias i Pujol, Badalona, Spain Introduction: Hematopoietic stem cell transplantation (HSCT) is an effective therapy for a variety of severe hematological diseases, among them myeloid malignancies. In patients without an HLA-identical sibling, the procedure is limited by the time spent to identify an unrelated and by the stem cell dose of umbilical cord blood units. In the last decade, haploidentical family transplantation followed by high dose of cyclophosphamide as GvHD prophylaxis (haplo-Cy) has been growing as a suitable option. We analyse here the outcome of this approach in a series of AML and MDS patients. Material (or patients) and methods: Thirty-three patients whit acute myeloid leukaemia (AML, n = 16) and myelodysplastic syndrome (MDS, n = 17) without HLA-matched sibling or unrelated donor, received a haplo-Cy between 04/13 and 06/15 using the same protocol in 4 Spanish hospitals. Highdose conditioning regimen included thiotepa 5 mg/kg x 2 days (-7 and -6), fludarabine 50 mg/m2 x 3 days (-5, -4 and -3) and busulfan 1 mg/kg/6 h x 3 days or its equivalent IV dose (day -5, -3 and -2). In patients older than 55 years busulfan administration was limited to two days. Graft versus host disease (GvHD) prophylaxis consisted of cyclophosphamide S146 50 mg/kg on day +3 and +4, and tacrolimus continuous iv infusion from day +5 and adjusting for blood levels 5-10 ng/l. Outcomes analysed were overall survival (OS), progression free survival (PFS); and cumulative incidences (CI) of GvHD, relapse and non-related mortality (NRM Introduction: Allogeneic HSCT from an HLA-haploidentical relative (haplo-HSCT) is now considered a suitable option for children with acute leukemia (AL) either relapsed or at high risk of treatment failure. We recently developed a novel method of negative depletion of α/β T cells, which was shown to be more effective than positive selection of CD34+ cells for S147 protection against infections. Here we report the comparison of the outcome of 80 children with AL given haplo-HSCT after α/β T-cell depletion (group 1) with that of patients transplanted from an HLA-identical sibling (group 2) or an UD (group 3) in the same time period. Material (or patients) and methods: All children were affected by AL and were transplanted at the Bambino Gesù Children's Hospital in Rome, Italy, between December 2010 and September 2014; 80 patients were enrolled in group 1, 41 in group 2 and 51 in group 3, respectively. Patients received haplo-HSCT in the absence of suitable conventional donor or if affected by rapidly progressive disease not permitting time to identify an UD. Clinical characteristics of patients assigned to the 3 groups and those of their donor are shown in Fig 1. All children were given a fully myeloablative regimen. No group 1 patient was given any post-transplantation GvHD prophylaxis, while patients of group 2 and 3 were given Cyclosporine-A and short-term methotrexate. Group 1 and 3 patients received ATG Neovis (4 and 5 mg/Kg/day, respectively) from day -5 to -3 for preventing both graft rejection and GvHD. Results: All group 2 patients had sustained engraftment of donor cells, while 1 patient in the group 3 and 2 in group 1 experienced primary graft failure. The CI of acute GvHD was 30%, 41% and 42%, respectively (P = NS). However, while none of the group 1 children had either gut or liver involvement, the incidence of visceral acute GvHD of group 2 and 3 patients was 17% and 16.3%, respectively ( p40.001). The CI of chronic GvHD was significantly lower in group 1 children than in those of groups 2 and 3 (5.4% vs. 18.9% and 23.5%, respectively, P = 0.02); none of the 4 group 1 patients experiencing chronic GvHD had the extensive form of the disease. Four, 1 and 6 children of patients assigned to group 1, 2 and 3, respectively, died for transplant-related causes leading to a CI of TRM of 5%, 2.4% and 11.8% respectively (P = NS HAPLO and SIB patients were significantly older (44 and 47 years) as compared with matched UD (35 years) mmUD (36 years) or CB. Phase of the disease was comparable in the 5 groups with patients coming to transplant with active disease ranged between 29% (SIB) and 47% (mmUD) (P = 0.1). The proportion of patients receiving a myeloablative conditioning regimen was highest in HAPLO grafts (100%) and lowest in SIB (58%) ( p40.001). Results: Results. The incidence of acute GvHD grade III-IV was 7% in SIB, 4% in MUD, 9% in mmUD and 3% in HAPLO and 1% in CB (P = 0.1); moderate-severe chronic GvHD was seen respectively in 25%, 20%, 13%, 22%,15% (P = 0.2). similarly the incidence of moderate-severe chronic GvHD was 25%, 20%, 13%, 23% respectively (P = 0.2). The cumulative incidence of transplant related mortality was 17% in SIB, 36% in MUD, 44% in mmUD, 13% in HAPLO and 37% in CB ( p40.0001); the cumulative incidence of relapse was instead significantly higher in SIB transplant (39%) as compared to MUD (22%), mmUD (23%), HAPLO (24%) and CB (29%) (P = 0.002). The actuarial 2 year survival (OS) was 60%, 50%, 40%, 63%, 44% in the 5 groups respectively. The actuarial 2 year GRFS was 37% in SIB, 33% in MUD, 30% in mmUD, 49% in HAPLO and 39% in CB (P = 0.1). The difference in 2 year GRFS between HAPLO and the other donor sources is statistically significant (P = 0.01); the difference in favour of HAPLO was particularly evident in patients with active disease at transplant (P = 0.01) as compared to patients in remissi on (P = 0.1). In a multivariate Cox analysis on GRFS, HAPLO grafts remained superior to other donor types (P = 0.002), together with disease phase ( p40.0001); donor recipient age -gender and intensity of the conditioning regimen were not predictive. Extrahematological toxicities were as usual, without significant differences between BEAM (or BEAM-like regimens) and HD-Mel groups (P = 0.59 and 0.56, respectively). Among the 21 patients with active disease before ASCT, the overall response rate was 100%, and those in CR were 5 (19%) . For the haplo phase, the stem cell source was bone marrow (n = 17) or peripheral blood (n = 9). All conditioning regimens were reduced-intensity. The 1-y cumulative incidence of transplant-related mortality (TRM) was 16% (1-30, n = 4). 4 toxic deaths occurred: viral pneumonia (n = 2), fungal pneumonia (n = 1) and multi-infection (n = 1). The 3-y overall survival (OS) of the entire cohort was 66% (95% CI: 47-86), the 3-y progression-free survival was 59% (40-79). The cumulative incidences of acute graft-versus-host disease (aGVHD) grade 2-4 and chronic GVHD were 29% (10-47) and 9% (0-22), respectively. No significant difference between BEAM (or BEAM-like regimens) and HD-Mel group was observed for OS (83% and 60% respectively, P = 0.46) or TRM (17% and 16%, P = 0.88). Conclusion: The tandem auto-haplo seems to be an effective therapeutic strategy in those lymphoma patients whose prognosis is expected to be unsatisfactory with ASCT alone and who lack a HLA-identical sibling or a matched-unrelated donor within appropriate timing. The toxicity after haplo was not enhanced. Disclosure of Interest: None declared. Introduction: Trans-placental trafficking of maternal and fetal cells during pregnancy establishes long-term, reciprocal microchimerism in both mother and child. As a consequence, the immune system of the mother may become sensitized to paternal histocompatibility antigens. In fact, antibodies directed against paternal HLA-antigens (van Rood JJ et al., Nature 181:1735 , 1958 and T lymphocytes directed against paternal major and minor histocompatibility antigens (van Kampen CA et al., Hum Immunol 62:201, 2001; Verdijk RM et al., Blood 103:1961 , 2004 were detected in multiparous women. More recently, it was hypothesized that mother's "exposure" to paternal HLA haplotype antigens during pregnancy may affect transplantation outcomes when the mother acts as donor for the child. Indeed, survival after T celldepleted HLA haploidentical haematopoietic transplantation was improved using the mother as donor (vs all other family members) (Stern et al., Blood 112:2990 . However, maternal donors were associated with increased incidence of GvHD and decreased survival after un-manipulated (i.e., T cellreplete) HLA haploidentical transplants (Wang Y et al., Blood 124:843, 2014) . Material (or patients) and methods: A retrospective EBMT registry-based study was performed in a combined series of adult (n = 333) and pediatric (n = 105) patients with acute leukemia (AML = 268, ALL = 160, Mixed phenotype = 10). Seventy-eight percent of patients received ex-vivo T cell depleted transplants, 22% received in-vivo T cell-depleted transplants. Results: As preliminary analyses showed transplantation outcomes from family members other than mothers did not differ from one another, such transplants were combined for analyses. When compared with transplantation from all other family members (n = 338), transplantation from mother donors (n = 100) was associated with better relapse-free survival (RFS) (43% vs 21%, P40.001), and trends towards lower relapse incidence (RI) (28% vs 39%, P = 0.07) and non-relapse mortality (NRM) (29% vs 39%, P = 0.08). Multivariate analyses showed transplantation from mother donors was an independent factor predicting better RFS (other donors vs mothers: HR: 1.42; CI: 1.01-2.00; P = 0.043) and lower RI (other donors vs mothers: HR: 1.85; CI: 1.12-3.06; P = 0.016). In addition, transplantation in relapse (vs remission) predicted worse RFS (HR: 2.36, CI: 1.83-3.04, P40.001), higher RI (HR: 3.29, CI: 2.29-4.73, P40.001) and higher NRM (HR: 1.70, CI: 1.19-2.43, P = 0.003). Age ≥ 18 (vs o18) adversely impacted RFS (HR: 1.40, CI: 1.00-1.95, P = 0.049) and NRM (HR: 3.01, CI: 1.73-5.23, P40.001). Conclusion: Our retrospective analyses in 438 HLA haploidentical T cell-depleted hematopoietic transplants for acute leukemia patients (pediatric and adult) show that transplantation from mother donors, when compared with transplantation from any other family member, is an independent factor predicting better outcomes, i.e., better RFS and lower RI. Mothers should therefore be preferred when selecting an HLA haploidentical family donor. Further clinical and preclinical studies are needed to unveil the mysteries underlying mother-to-child immune interaction during pregnancy and its bearing on the reproductive success of the human species. Disclosure of Interest: None declared. Recovery of CMV Specific T Cells Following Alternative Donor Allogeneic Transplant with Post-Transplant Cyclophosphamide L. Lamb 1 , A. Saad 1 , S. Mineishi 1 , S. Langford 1 , A. Di Stasi 1,* 1 Medicine, UNIVERSITY OF ALABAMA AT BIRMINGHAM, Birmingham, United States Introduction: Post-HSCT high-dose cyclophosphamide (CY) is a useful strategy for GvHD prophylaxis in patients who receive allogeneic marrow grafts from alternative donors, although delayed immune reconstitution can result, increasing risk for CMV reactivation. We compared lymphocyte recovery, CMV infection, and CMV-specific CD8+pp65+ recovery in patients who received HPC-Apheresis grafts and high-dose CY following haplo-identical (HAPLO), matched unrelated donor (MUD), and mismatched unrelated donor (mMUD) grafts to conventional matched related donor (MRD) graft recipients. Material (or patients) and methods: A total of 26 patients (median age 49; range 20-72) n = 5 (HAPLO); 6 (MRD); 15 (MUD) were evaluated. All patients received busulfan or TBI based conditioning. CY (50 mg/kg/day) was administered on days 3 and 4 following HAPLO and on day 3 following MUD transplant. Lymphocyte recovery and frequency of circulating CD8+pp65+ T cells were assessed on post-HSCT days 30, 60, and 90. Circulating anti-CMV T cell frequency was assessed by single platform flow cytometry using a phycoerythrin-tagged MHC dextramer against HLA-specific CMV pp65 peptides and mAbs vs. CD3, CD8, CD4, CD16/56, and CD19. Anti-CMV CD8+ T cell immunity is defined as a CMV-dextramer positive count ≥7cells/ml. CMV reactivation was defined as a serologic titer 4500IU/mL. Results: Day +30 T cell recovery was significantly faster in MRD than CY-treated recipients (P = 0.015) due to more robust CD8+ T cell recovery. CD4 T cell recovery remained incomplete in all groups at day +180. NK cells recovered to normal numbers at d+28 in all groups. Neither post-HSCT CY nor donor source significantly impacted recovery of anti-CMV CD8 + T cells (P = 0.8232). Excluding donors (D) and recipients (R) that were both negative, CMV+ T cells recovery was complete in 4/5 MRD, 7/14 MUD, and 3/5 HAPLO by d+100. Among MRD recipients either D+ or R+ (n = 5), 2 patients showed CMV reactivation within 40 days of transplant that was associated with o7 CMV Dextramer (CMV/DEX)+ T cells on d+30. A high (490/mL) CMV/DEX T cell response in one patient shortened the duration of viremia to 10d and 3 patients showed no CMV reactivation and a high CMV/DEX+ T cell response by d+60. For HAPLO CMV D+ and/or R+ (n = 5) recipients, 4 with o7 CMV/DEX+ T cells/mL at +30 experienced CMV reactivation within 50d while a robust CMV/DEX+ T cell response by d+60 was associated with shorter duration of viremia. One patient with o7/mL CMV/DEX+ T cells had CMV viremia for 36d. For MUD CMV D+ and/or R+ recipients (n = 14), 3 showed CMV reactivation within 50d of transplant -had o 7/mL CMV/DEX+ T cells at d+30, and robust CMV/DEX+ T cell response on d+60 predicted shorter duration of viremia. Conclusion: Neither post-HSCT CY nor donor source significantly impacted the percentage of patients that recovered anti-CMV CD8+ T cells at each time interval. Although not completely protective against reactivation, CMV/DEX+ counts47/mL were associated with a shortened duration of viremia while on antiviral therapy. Day +30 CMV/DEX+ counts o7 cells were associated with increased risk of CMV reactivation. This interim analysis suggests that CMV/DEX+ T cell enumeration is a useful biologic correlate for determining clinical response to antiviral therapy, and that donor-derived CMV specific T cell immunity is not further compromised following high-dose CY in alternative donor HSCT. Disclosure of Interest: None declared. Natural Killer (NK) cell alloreactivity has been demonstrated to play a major role in T cell-depleted haplo-HSCT, but whether, and how, PT-Cy affects NK cell reconstitution is to date unknown. Material (or patients) and methods: We analyzed the grafts and serial peripheral blood (PB) and bone marrow (BM) samples from 8 patients who received myeloablative haplo-HSCT with peripheral blood stem cell (PBSC) grafts and PT-Cy, sirolimus and mycophenolate as GvHD prophylaxis. To characterize reconstituting NK cells, we designed a multiparametric flow cytometry panel comprising 27 markers involved in target recognition, activation, maturation and exhaustion. We used intracellular staining to determine the frequency of Ki67 + proliferating cells and the expression of the Aldehyde Dehydrogenase (ALDH) enzyme, known to confer resistance to PT-Cy. Interleukin-15 (IL-15) serum concentration was quantified using the Bio-Plex Pro Human Cytokine 4-plex assay. Results: All patients received high numbers of mature donor NK cells as part of the graft (median 17x10 6 /kg), and donorderived NK cells were detectable at low counts as early as day 3 after HSCT and throughout the entire follow-up. At day 3 after HSCT, all subsets of NK cells, including single KIR + alloreactive cells, were actively proliferating (61,14% of Ki67 + cells), possibly driven by the high levels of IL-15 detected in the patients' sera after conditioning (21 ± 6 pg/ml). After PT-Cy infusion, a marked reduction in the frequency and counts of proliferating NK cells was evident, suggesting selective killing of dividing cells by PT-Cy. In line with this hypothesis, NK cells from the graft and harvested from patients at day 3 after HSCT showed no detectable ALDH expression. The phenotype of NK cells also changed upon PT-Cy administration: whereas before the infusion they resembled their mature counterparts from the graft, after PT-Cy an immature phenotype, CD62L + NKG2A + KIR -, became prevalent, suggesting derivation from donor HSCs rather than from infused NK cells. In line with these features, at day 30 we detected very low numbers of putatively alloreactive single KIR + NK cells, both in the PB and in the BM of the patients. Finally, we analyzed the impact of predicted NK alloreactivity in an extended series of 64 patients who received haplo-HSCT with PT-Cy, detecting no significant difference in progressionfree survival between patients with or without predicted alloreactivity (43% vs 52% at 1 year after HSCT, P = 0.82). Conclusion: Our data suggest that the majority of mature NK cells infused with unmanipulated grafts is eliminated upon PT-Cy administration and, as a consequence, that in this setting NK cell alloreactivity might be blunted by the elimination of donor single KIR + cells and by the competition between reconstituting NK and T cells. Still, the high levels of IL-15 detected in patients' sera at early time-points might provide a biological rationale for the infusion of mature donor NK cells early after PT-Cy administration. Disclosure of Interest: None declared. Introduction: HLA-haploidentical hematopoietic stem cell transplantation (haplo-HSCT) is increasingly offered to patients with high-risk acute myeloid (AML) or lymphoid leukemia (ALL). Unfortunately, graft manipulation employed to overcome the HLA barrier significantly delays immune reconstitution, posing the patients at risk of infections. Accordingly, non-relapse mortality after haplo-HSCT clearly extends beyond day 100 post-transplant. Over the years, different approaches have been investigated to speed-up immune reconstitution. In the absence of validated immune biomarkers, it is however difficult to evaluate the clinical impact of accelerated immune reconstitution. Material (or patients) and methods: Among AML and ALL patients in the EBMT database who underwent haplo-HSCT in the period 2001-2012, criteria for study entry were survival beyond day 100 and availability of differential immune-cell counts (CD3+, CD4+, CD8+ T cells, CD19+ B cells, CD16+/CD56+ NK cells). Of 259 patients meeting these criteria (age 2-70, median 33), 67 (26%) were children. The underlying disease was AML in 162 cases (63%), while ALL in the remaining. Fifty-two percent of patients were transplanted in CR1. The graft was manipulated in 199 patients (78%), including CD34-selection (50%), ex vivo T-cell depletion (15%) or both (13%). Results: The estimated overall survival at 2yrs was 43%. The estimated cumulative incidence of death due to relapse was 33%, while that of death due to other causes was 35% (51% of those were infections) The occurrence of grade III-IV GVHD and of chronic GVHD was 9% and 18% (7% extensive), respectively. Negative prognostic factors for overall survival were disease state 4CR1 (P = 0.002) and CMV seropositivity (P = 0.009). By day 100 post-transplant, patients reached the following median immune-cell counts: 100 CD3+ T cells (range 0-2576), 30 CD4+ T cells , 48 CD8+ T cells , 276 CD16+/CD56+ NK cells (18-3581), 21 CD19+ B cells . Importantly, CD3+ counts above the first quartile (1Q) of the entire data set (29 cells per microL) were significantly associated with a better overall survival (P = 0.0005 by Log rank) and a lower incidence of death due to causes other than relapse (P = 0.002 by Gray test). The same held true for CD8+ counts (1Q: 15 cells per microL; P = 0.003 on overall survival; P = 0.0004 on death due to other causes). None of the other immune-cell counts analyzed correlated with clinical outcome. Strikingly, when challenged in multivariate analysis taking into account age category, CMV seropositivity, graft manipulation and CR1 status, CD3+ and CD8+ counts above the 1Q adjusted to fit optimal cut-off points were still significantly associated with a better overall survival (P = 0.006 and P = 0.015, respectively), but only CD8+ values associated with a lesser risk of death due to causes other than relapse (P = 0.026). Conclusion: Contrary to what is generally accepted, these results indicate that an accelerated CD8+, but not CD4+, T cell reconstitution associates with a more favorable clinical outcome after haplo-HSCT. Moreover, they suggest that yet to be validated CD8+ cut-off points, rather than the commonly used arbitrary value of 200 CD4+ T cells per microL, should be considered as surrogate biomarkers in clinical trials. Disclosure of Interest: None declared. Introduction: Relapse of leukemia in the central nervous system (CNS) after conventional treatments is difficult to treat and associated with poor prognosis. The effectiveness and safety of human leukocyte antigen (HLA)-mismatched donor lymphocytes infusion for leukemia has been reported. For pediatric patients with cerebrospinal fluid (CSF) dissemination of medulloblastoma, intrathecal (IT) infusion of haploidentical peripheral blood lymphocyte lymphokine-activated killer cells proved effective in several cases without severe adverse events. We performed IT infusion of haploidentical non-donor lymphocytes for relapsing CNS leukemia after haploidentical hematopoietic stem cell transplantation (HSCT). Material (or patients) and methods: A 13-year-old girl (HLA-A31 + ) was diagnosed as relapsing from Philadelphia chromosome-positive acute mixed phenotype leukemia in CNS after receiving chemotherapy, tyrosine kinase inhibitors, haploidentical HSCT from her father (HLA-A31 -), and craniospinal irradiation. We received approval from our Ethical Review Board for IT infusion therapy of haploidentical non-donor lymphocytes, and obtained informed consent from her and her parents. Peripheral blood mononuclear cells (PBMCs) were obtained from her mother (HLA-A31 + ), as a haploidentical non-donor. PBMCs were administered by IT weekly using lumbar puncture. Results: Diagnostic lumbar puncture revealed a CSF nuclear cell count (NCC) of 12/μl, 29.4% of which exhibited chimerism based on XY-fluorescent in situ hybridization (FISH) analysis and 29.0% exhibited a BCR/ABL FISH signal, although no mass was observed in images. CD3 + T cells (2.1x10 7 ) were infused by IT injection. Examination of CSF before and one week after the first IT infusion showed NCC had increased from 8/μl to 303/μl, with recipient chimerism and BCR/ABL transcript signal barely detectable, decreasing from 88.9% to 0.0% and from 26.8% to 0.6%, respectively. Furthermore, HLA-A31 + /CD3 + T cells were not detected and the proportion of CSF CD3 + /HLA-DR + T cells had increased at one week post-IT. These results indicate the mother-infused lymphocytes did not survive and lymphocytes of her father, a haploidentical HSCT donor, had migrated to CSF and become activated. We repeated IT infusion seven times without increasing cell quantity. CSF contained 40-70/μl NCC, recipient chimerism less than 1.0%, and a BCR/ABL transcript signal of around 2.0%. The patient showed partial remission for two months following therapy. After a second IT infusion, persistent headache and malaise had disappeared, although a mild transient headache was observed as a side effect. No graft-versus-host disease was observed during the observation period, and good quality of life was maintained. Conclusion: To control leukemic CNS dissemination, it is important to manage the possibility of graft-versus-leukemia effects that might result from using haploidentical non-donor lymphocytes. Long-term observation is necessary to verify the sustainability and potential long-term complications of this effect. Disclosure of Interest: None declared. CD3/CD19 depleted haploidentical stem cell transplantation in sickle cell disease: How to 'cure'mixed chimerisms B. Pfirstinger 1,* , J. Föll 1 , S. Corbacioglu 1 1 Pediatric Hematology, Oncology and Stem Cell Transplantation, University Childrens Hospital REGENSBURG, Regensurg, Germany Introduction: T-depleted haploidentical stem cell transplantation (T-haplo-SCT) is a well-established treatment option in experienced transplant centers in particular for children. Sickle cell disease (SCD) is one of the most devastating hematological malignancies, an inherited disorder with an estimated incidence of over 350.000 affected newborns per year worldwide. Despite significant improvements in the supportive management of SCD-related complications, the disease cause substantial morbidity and mortality with reduced life expectancy. Allogeneic haematopoetic stem cell transplantation (SCT) is currently the only curative therapy for SCD and offered only if a matched sibling donor (MSD) or unrelated donor (MUD) is available. For the treatment of sickle cell disease (SCD) T-haplo-SCT can evolve to the treatment option of choice not only for those without a compatible donor. With transplant related mortality being an inacceptable outcome in hemoglobinopathies, graft-versus-host-disease (GvHD) is the worst transplant related morbidity (TRM). With these priorities T-haplo-SCT is can be the prioritized therapeutic option due to the lowest incidence of GvHD. But in T-haplo-SCT mixed chimerism and graft rejection remain the major drawbacks in particular in heavily pre-transfused patients. Material (or patients) and methods: Here we report of 2 male patients (age 4 & 16 years) with homozygous SCD and a high degree of disease burden, transplanted from haploidentical donors. Both patients suffered from multiple pain crises, several multifocal bone infarcts and an accelerated transcranial doppler sonorgraphy (TCD) 4200cm/s as indication for SCT. The donors were the mother in the 4 yrs old and the father in the 16 yrs old, respectively. Results: Conditioning consisted of ATG (Fresenius; upfront), Thiotepa (10), Fludarabin (160) and Treosulfan (46) both. One patient received additionally a total nodal irradiation (TDI) due to homo-/heterozygous HLA-constellationwith high risk for graft rejection. Peri-and post-transplant immunosuppression consisted of CSA and MMF in one patient and of Tacrolimus plus MMF in the other patient. One patient achieved a full chimerism at day +39 but showed an almost rejection within days with a drop in chimerism of o30% after discontinuation of MMF on day+ 40. With the restart of MMF we could observe a promptly increasing chimerism up to 96% associated with a drop of the total lymphocyte count and a slow but steady recovery of donor derived CD3+ T cells. The second patient with his very high risk of rejection received Tacrolimus/MMF as immunosuppressive agent and achieved an almost full chimerism of 98% on day +26. But also within days his chimerism dipped to 53%. Tacrolimus was then replaced with CSA and steroids were added to MMF which resulted in a steadily increasing and then stable chimerism above 85%. Conclusion: Next to viral infections mixed chimerism and rejection is the major drawback in T-haplo-SCT in particular in heavily pre-transfused patients. In paradox to the usual abrogation of immunosuppression +/-DLI we increased the immunosuppression under the impression of an immature donor T-cell graft and observed that in T-haplo-SCT a mixed chimerism and imminent graft loss can be recuperated via a prolonged or intensified immunosuppression with an unexpectedly low risk for viral complications in this particular patient population. Disclosure of Interest: None declared. Introduction: Allogeneic hematopoietic stem cell transplantation (alloHSCT) is considered the standard of care for patients with HL that relapse after autologous HSCT. Fit patients with chemosensitive disease can benefit from alloHSCT using and identical sibling (SIB) or matched-unrelated (MUD) donors; however, restricted availability of a suitable SIB or MUD limits its use. Recently, encouraging results have been obtained using haploidentical donors (HAPLO) and post-transplantation cyclophosphamide (PTCy) as GVHD prophylaxis. Because information regarding the results of alloHSCT using alternative donors is still scarce, we aimed to compare outcome of HAPLO transplants with conventional SIB and MUD for HL. Material (or patients) and methods: Information of patients older than 17y with HL who received an alloHSCT from a SIB, MUD, or a PTCy-based HAPLObetween 2010-2013 was downloaded from the EBMT and GETH databases. Results: 709 patients with HL were identified meeting the inclusion criteria. 338 received a transplant from a SIB donor, 273 from a MUD, and 98 from HAPLO. In comparison to SIB and MUD, a significant higher number of patients treated with alloHSCT from HAPLO donors received reduced intensity conditioning (RIC) regimens (69% and 69% vs. 90%, respectively, P40.001), bone marrow stem cells (10% and 11% vs. 61%, respectively, P40.001), and female donors (45% and 23% vs. 55%, respectively, P40.001). Other variables such as sex, age, performance status, refractory HL, and previous autologous SCT were balanced. Median follow-up after alloHSCT for all patients was 20.6 months (1-67). The 2-year probabilities of overall survival (OS) and progression-free survival (PFS) were 70% and 36% after SIB transplant, 61% and 44% after MUD, and 66% and 46% after UCB, respectively. The 2-year probabilities of non-relapse mortality (NRM) and relapse rate (RR) were 13% and 51% after SIB, 23% and 33% after MUD, and 16% and 38% after HAPLO. Multivariate analysis showed that, in comparison with SIB, HAPLO and MUD transplants were associated to a lower RR (HAPLO: HR 0.67 95% CI 0.5-0.9, P = 0.038; MUD: HR 0.61 95% CI 0.5-0.8, P40.001). However, while no differences in terms of NRM were observed between SIB and HAPLO (HAPLO: HR 1.6 95% CI 0.7-2.3, P = 0.36), MUD was significantly associated to a higher NRM (HR 1.7 95% CI 1.2-2.6, P = 0.005). HAPLO transplants showed similar OS (HR 1.3 95% CI 0.8-1.9, P = 0.24) and a trend to a better PFS (HR 0.8 95% CI 0.6-1.1, P = 0.14) when compared with SIB. MUD alloHSCT was associated with lower OS (HR 1.5 95% CI 1.2-2.0, P = 0.002), but with no differences in PFS (HR 0.9 95% CI 0.7-1.1, P = 0.31). Other factors associated to better OS and PFS were age lower than 40y, good performance status, nonrefractory HL, and CMV negative donor. Interestingly, acute and chronic GVHD incidence were significantly increased after MUD compared to SIB (49% vs. 33%, P40.001; 47% vs. 32%, P = 0.004, respectively), whereas no significant differences were found between HAPLO and SIB. Conclusion: This registry study suggests that in adults with advanced HL, the outcome of PTCy-based HAPLO alloHSCT may be comparable to that of conventional SIB and MUD alloHSCT across multiple centers and conditioning regimens. These findings warrant further investigations. Disclosure of Interest: None declared. Introduction: Haploidentical hematopoietic stem cell transplantation (haplo-HSCT) with post-transplant Cyclophosphamide (PTCy) is a type of hematopoietic transplantation in increasing use over the last decade. Our aim was to describe the infections developed by our patients during the first year after haplo-HSCT with PTCy. Material (or patients) and methods: We retrospectively analyzed bacterial, viral and fungal infections of 69 patients who underwent a haplo-HSCT in our center (Jacie-accredited SCT unit) between December-2007 and February-2015. We divided the analysis in two groups: the early infections (≤100 days after haplo-HSCT) and late infections (4100 to 365 days after haplo-HSCT). As graft versus host disease prophylaxis, we used PTCy on days +3 and +4 as well as Cyclosporine and Mycophenolate since day +5. All patients received G-CSF since day +5. As infection prophylaxis we administered Levofloxacin and Acyclovir from day -7, Micafungin from day -1 (later Fluconazole o Posaconazole) and Cotrimoxazole from day -7 to -2 and since engraftment. Cytomegalovirus (PCR) and Aspergillus (antigenaemia) were monitor biweekly. Cytomegalovirus reactivations were treated, as pre-emptive strategy, with con Ganciclovir, Valganciclovir or Foscarnet. In our analysis, Citomegalovirus reactivations were considered as viral infections. Results: Patient´s characteristic are included in Table 1 . We identified a total of 239 infections. This infections were developed by 62 (90%) of the 69 patients. The incidence of infections divided by pathogens is included in Table 2 . A more detailed description of etiology and localization of early and late infections can be found in Table 3 . Among the early infections, the incidence of bacterial infections preengraftment was 41% (9/22) and the incidence of fungal infections pre-engraftment was 64% (7/11). The median day of engraftment (N4500) was 17 days (range: 13-28). The median days of the first Cytomegalovirus reactivation was 32 (range: . The comparison between immune reconstitution and number of infections is illustrated in Figure 1 . With a median follow-up of 13 months (range: 3-58), the cumulative incidence of death caused by infections was 12.2% (IC 95% = 6. 3-23.5) . Conclusion: In conclusion, our data suggest that infectious profile after haplo-HSCT with PTCy is acceptable. Certainly, the rate of viral infections during the early period is remarkable, however the incidence dramatically decreases after day +100 coinciding with immune reconstitution. Our future aim will be to compare this results with other types of alogenic transplant. Disclosure of Interest: None declared. Introduction: Haploidentical hematopoietic stem cell transplantation (HSCT) serves curative alternative options with promising clinical outcome. On the other hand it is associated with a higher treatment-related mortality due to a significant delay in the recovery of the adaptive immune system. Therefore we would like to present our experience of immune reconstitution outcome in 23 children receiving haploidentical HSCT after negative depletion of TcRαβ(+). Material (or patients) and methods: Erciyes University is one of the reference centers for Haploidentical HSCT in Turkey; 35 haploidentical HSCT in 30 patients were performed between 2011 and 2015. Twenty-three patients (6 female, 17 male) whom long term immunological follow up was obtained were enrolled in the study. All of them received TcRαβ(+) haploidentical HSCT at the Pediatric BMT Unit of Erciyes University School of Medicine, Kayseri; Reconstitution of lymphocyte subsets (CD3, CD4, CD8) was monitored monthly by flow analysis until six months, followed by an assessment every 3 months. In addition regular PCR monitoring of viral agents including CMV, EBV, ADV, BK, HSV, HHV-6 were performed. [P083] Results: The median age of 23 children was 8 years (range: 2-17 years) in the study group. 13 had AML, 4 had ALL and 4 had severe aplastic anemia, 1 have HLH, 1 have Griscelli syndrome. Thirteen mothers and 10 fathers were chosen as donor. The median number of transplanted CD34 stem cells was 19.73 x 10 6 /kg (range7.58-39.85). Primary engraftment occurred in all patients. Four patients with leukemia were relapsed. Five patients experienced skin and gastrointestinal acute GVHD; however none of the patients developed chronic GVHD. The median number of CD3+ T cells was 550 (range 89-3311)/ mm 3 ;, for CD4+ helper T cells 87 (range 1-828)/mm 3 ; CD8+ T cytotoxic cells 330 (range 70-2754) /mm3; for CD19+B lymphocyte 24 (range:3-301)/mm 3; TCR αβ cells 144(range 19-182) /mm3; TCR γδ cells 200(range 107-1296) /mm3 at 28 th day of HSCT. The levels of TCR αβ subtype have increased and exceed the number of TCR γδ by the median time of 4 th months of transplantation. By the 4 th months of HSCT, we observed reduction of preclinical viral infections, and clinical viral diseases. Conclusion: Our data suggest that immune reconstitution is rapid and acceptable in the TcRαβ(+) haploidentical HSCT group. Disclosure of Interest: None declared. Introduction: In recent years, there is a remarkable trend in the use of haploidentical-related hematopoietic cell transplantations (haplo-HCT) in patients who do not have a HLA matched related or unrelated donor. Here, we report our preliminary, single-center, retrospective experience in patients who underwent haplo-HCT. Material (or patients) and methods: The medical records of all consecutive patients who underwent haplo-HCT during February 2012-July 2015 were retrospectively analyzed. All patients received same GvHD prophylaxis, which included cyclosporine-A, short term methotrexate, rabbit-ATG and post-HCT cyclophosphamide at day +3. Descriptive statistics are presented as median and range. Overall survival was calculated with Kaplan Meier analysis. Results: The study included a total of 14 patients (4 female; 10 male). The median age of the cohort was 28 (19-46). Primary diagnosis of patients were as follows: acute lymphoblastic leukemia (n:6), acute myeloid leukemia (n:5), non-Hodgkin lymphoma (NHL) (n:1), chronic myeloid leukemia in accelerated phase (n:1) and severe aplastic anemia (n:1). Nine and 5 patients received reduced intensity and myeloablative conditioning, respectively. One patient with NHL and 5 out of 11 patients with acute leukemia underwent haplo-HCT with active disease witho 10% marrow blasts. Twelve and 2 patients received peripheral blood and bone marrow derived grafts, respectively. The median infused CD34 + cell dose was 7x10 6 (1,62 x10 6 -9 x10 6 ). Grade 2-4 acute and moderate/severe chronic GvHD was developed in 5 (36%) and 3 (21%) patients, respectively. During a median follow-up of 23 months, 6 patients (4 with active disease at HCT) died. The overall survival (OS) of patients who received HCT in remission and active disease were 75% and 33%, respectively. The OS of the whole study cohort was 50%. Transplant-related mortality rates at D+100 and 1 year were 7% and 14%, respectively. Conclusion: As nearly half of our patients received HCT while in active disease, our preliminary results seems to be encouraging. Nearly universal/rapid availability of donors and relative low cost make haplo-HCT an attractive alternative donor source in selected patients in remission, who do not have a HLA matched related or unrelated donor and needs an urgent HCT. Disclosure of Interest: None declared. Neutrophil engraftment were delayed (related, MUD, HAPLO mean 14 vs 18 vs 21.5 days respectively P = 0,02) in the haplo group. We don't have any grade 3-4 acute GVHD in MUD and HAPLO donor groups compared to 9% in the related donor group (P = 0,9). Relapse rates were similar. (P = 0,64). Transplant related mortality in the first 100 days is highest in the HAPLO group (43%) compared to related (16%) and MUD (%19) group (P = 0.05) due to high percentage (41%) of refractory leukemia patients in this group. Estimated disease free survival (DFS) and overall survival (OS) were similar as shown in the table. Conclusion: According to these data HAPLOs were not inferior compared to related and MUDs in terms of transplant outcome. Increasing number of HAPLOs and in the next step, the cost effectiveness of these two alternative transplants will display the best available donor for our patients. Disclosure of Interest: None declared. Poorer outcomes in patients with treatment-related acute leukemia than in patients with de novo acute leukemia during the first remission after allogeneic hematopoietic Introduction: Therapy-related acute leukemia (T-AL) is associated with a poor prognosis after conventional therapy. Allogeneic hematopoietic cell transplantation (alloHCT) has been suggested for T-AL. However,no studies have directly compared T-AL with de novo acute leukemia (AL) after alloHCT. Material (or patients) and methods: We reviewed the data of 23 patients with T-AL and 111 patients with de novo AL during the first complete remission (CR1), who received myeloablative alloHCT from an identical-sibling donor or haploidentical donors (ISD or HID, respectively) between January 2006 and December 2014. One hundred and eleven de novo AL patients were selected in terms of a case-control ratio of 1:3-5. Results: The 3-year overall survival(OS) and leukemia-free survival (LFS) of T-AL was lower than de novo AL (58% vs. 81%, P = 0.003 and 46% vs. 81%, P = 0.001, respectively) after alloHCT. The 3-year cumulative incidence of relapse did not differ significantly between the 2 groups (25% vs. 10%, P = 0.11). The 3-year cumulative incidence of non-relapse mortality (NRM) of T-AL was apparently higher than de novo AL (25% vs.6%, P = 0.06). Conclusion: Although allogeneic transplantation remains the only curative therapy for T-AL and can markedly improved the prognosis of T-AL, its outcomes in T-AL patients are inferior to outcomes in patients with de novo AL. We expect that a new therapy for malignant tumor, especially targeted drug therapy, will reduce the need for cytotoxic management and consequently reduce the incidence of T-AL Introduction: n the era of Eculizumab, indentifying patients with PNH who may benefit from allogeneic stem cell transplantation(SCT) is challenging, especially for those who have no HLA-matched donors. Several recent studies have shown that HLA-haploidentical SCT for patients with hematological malignancy can achieved comparable outcomes with HLA-identical sibling transplantation. There are very few reports on the use of HLA-haploidentical SCT for PNH. The aim of the present study was to assess the long-term clinical outcome of HLA-haploidentical SCT in patients with PNH. Material (or patients) and methods: Total of 12 PNH patients received HLA-haploidentical SCT between Oct 2010 and Oct 2015 at our institution. The patients were aged 8 to 54(median 22.5 years). The median interval from the diagnosis to transplantation was 5 months (range 2-180). Of the 12 HLAhaploidentical donors, 6 were siblings, 2 fathers, 2 mothers, 1 son and 1 daughter. 11 patients received myeloablative conditioning regimen consisting of busulfan, cyclophosphamide and ATG (anti-thymocyte globulin), 1 patient who underwent salvage HLA-haploidentical SCT after the graft failure of double umbilical cord blood transplantation received conditioning including reduced-intensity total body irradiation, cyclophosphomide and ATG. G-CSF-mobilized bone marrow and peripheral blood stem cells were transplanted as graft. Prophylaxis for graft-versus-host disease(GVHD) consisted of cyclosporine or tacrolimus+mycophenolate mofetil+short-term methotrexate. Results: All 12 patients were engrafted successfully. The median time of neutrophils (ANC) reached to 0.5 × 10 9 /L and platelets (PLT) reached to 20 × 10 9 /L were 12 (range 11-26) days and 15 (range 11-120) days, respectively. 2 patients developed grade Ⅱ acute GVHD, 2 patients developed limited chronic GVHD. After a median follow-up time of 16.5 (range 2.0-40.0) months, the 3-year OS probability was 77.8 ± 13.9%. 2 patients died of treatment-related mortality, including severe pulmonary infection (n = 1) and transplant-associated thrombotic microangiopathy (n = 1), respectively. No patients were documented to have a recurrence of PNH clone after SCT. Conclusion: This study showed that long-term outcomes of HLA-haploidentical SCT in patients with PNH were comparable to that of HLA-matched donor SCT (the 3-year OS probability was 80.5 ± 10.2%, P = 0.02) at our institution. HLAhaploidentical SCT should be considered as a valid alternative therapeutic option for PNH patients without HLA-matched donors. Disclosure of Interest: None declared. Comparable survival for poor-risk cytogenetics AML in first complete remission after haploidentical donor or unrelated donor transplantation: a study from the ALWP of EBMT F. Lorentino 1,* , M. Bernardi 1 , M. Labopin 2,3,4 , F. Ciceri 1 , J. Esteve 5 , optimal donor choice is controversial. We recently reported that acute leukemia (ALL and AML) pts in CR1 and CR2 receiving a 10/10 unrelated donor (UD) have better outcomes than pts receiving a haploidentical (Haplo) HSCT, while no significant differences exist between 9/10 UD and Haplo (Piemontese S., EBMT 2015 abs#O010). We compared outcomes of Haplo-to those of 10/10 and 9/10 UD-HSCT in PRC-AML in CR1, to further investigate the influence of donor type in this peculiar disease entity. Material (or patients) and methods: We selected denovo PRC AML pts in CR1 undergoing unmanipulated haplo (n = 63), 10/10 UD (n = 383) and 9/10 UD HSCT (n = 122) from 2007 to 2014. PRC was defined as the presence of one of the following: complex karyotype; monosomal karyotype (one autosomal monosomy plus one monosomy or structural abnormality); inv (3)/t(3;3); -5 or del(5q); -7 or abn(7q); t(v;11)(v;q23); abn(17p); t (6;9). Results: Haplo-HSCT were comparable to 10/10 UD-and 9/10 UD-HSCT regarding time from diagnosis to HSCT (165 days(d), 157d, 169d, respectively, P = ns) and time from CR1 to HSCT (112d, 107d, 113d, respectively, P = ns). Haplo-HSCT more often received a myeloablative conditioning regimen (68% Vs 49% for 10/10 UD and 57% for 9/10 UD, P = 0.03) and bone marrow as stem cell source (37% Vs 19% for 10/10 UD and 15% for 9/10 UD, P40.01). UD-HSCT more often received invivo T cell depletion (TCD) with ATG (51% for Haplo, 74% for 10/10 UD, 96% for 9/10 UD, p o0.01). With a median follow-up of 18 months, no differences in 2 years (y) leukemia-free survival (LFS) and overall survival (OS) were found according to donor type (LFS: 46 ± 14% for haplo, 47 ± 6% for 10/10 UD and 41 ± 10% for 9/10 UD, P = ns; OS: 56 ± 14% for haplo, 54 ± 6% for 10/10 UD, 48 ± 10% for 9/10 UD, P = ns). At 2y, cumulative incidences (CI) of relapse (RI) and non-relapse mortality (NRM) were similar for all groups (RI: 30 ± 13% for haplo, 35 ± 5% for 10/10 UD, 38 ± 10% for 9/10 UD, P = ns; NRM: 25 ± 10% for haplo, 17 ± 7% for 10/10, 21 ± 8% for 9/10 UD, P = ns). Neither 100d CI of grade ≥ 2 aGvHD nor 2y CI of cGvHD differed according to donor (aGvHD: 28 ± 10% for haplo, 27 ± 5% for 10/10 UD, 33 ± 8% for 9/10 UD, P = ns; cGvHD: 39 ± 14% for haplo, 36 ± 5% for 10/10 UD, 29 ± 9% for 9/10 UD, P = ns). Multivariate analysis adjusted for pts age, time from diagnosis to HSCT, conditioning intensity, stem cell source, in-vivo TCD, donor/pts gender showed that, compared to haplo-HSCT, HSCT from 10/10 and 9/10 UD was not independently associated with better LFS (HR: 0.9, P = 0.8 and HR: 0.8, P = 0.8, respectively), OS (HR: 1, P = 0.9 and HR: 1, P = 0.8, respectively), RI (HR: 1.1, P = 0.8 and HR: 1.2, P = 0.5, respectively), NRM (HR: 0.8, P = 0.5 and HR: 0.9, P = 0.7, respectively), grade ≥ 2 aGvHD (HR: 1.1, P = 0.7 and HR: 1.5, P = 0.02, respectively) and cGvHD (HR: 1.2, P = 0.4 and HR: 1, P = 0.9, respectively). Conclusion: Outcomes of Haplo-are comparable to 10/10 and 9/10 UD-HSCT for PRC-AML pts in CR1. If confirmed in prospective trials, these results could likely influence criteria for donors selection, especially when disease biology and chemosensitivity seem most determinant for prognosis and warrant non-delayed treatment. Disclosure of Interest: None declared. Introduction: Allogeneic transplantation is the only curative option for patients with high risk hematologic malignancies. Only one third of them have an HLA identical sibling donor and around 70-80% will find an unrelated donor, that´s why HAPLO-HSCT offers a therapeutic option to most of these patients with the advantages of quick availability, easy programation and logistics, and a committed donor. Bone marrow (BM) or peripheral blood stem cells (PBSC) have been used as graft source but it´s not well established if any of them offer significant advantages. Material (or patients) and methods: We retrospectively evaluated the results of HAPLO-HSCT with reduced conditioning regimens and GVHD prophylaxis based on PTCy (50 mg/kg on days +3 and +4) and a calcineurin inhibitor plus mycophenolate from day +5 performed in GETH centers, with focus on the graft source comparison. Results: From Dec-2007, 181 patients have received an HAPLO-HSCT in 18 centers. Median age was 40 years (16-67), 63% were males and all were in advanced phases of their disease or presented high risk features (Hodgkin 37%, AML/ ALL/MDS 37%, NHL/myeloma/others 26%). Previous HSCT had been employed in 55% (autologous 43%, allogeneic 12%), and in 45% the HAPLO-HSCT was their first transplant. Disease status at HAPLO-HSCT was CR in 45%, with persistent disease in 55%. The DRI was low in 21%, intermediate in 31%, high in 37% and very high in 11%. The HCT-CI was 0-2 in 64% and 3 or higher in 37%. BM was the graft source in 54 patients (30%) and PBSC in 127 (70%), non T-cell depleted in all cases. The haploidentical donor was the patient´s mother (25%), father (9%), siblings (42%) or offspring (22%). Baltimore´s reduced conditioning (RIC) including 200 cGy was employed in 14% and RIC based on IV busulfan in 85% (37% with 3.2 mg/kg on day -2 (BUX1), and days -3 and -2 (BUX2) in 47%). Median neutrophil engraftment was reached on day +18 (13-45) and platelets 420 K at day +26 (11-150), without significant differences between BM and PBSC. Cumulative incidence (CI) of non-relapse mortality (NRM) at 1 year was 23% for both graft sources (P = 0.94). CI of grade II-IV acute GVHD at day +100 was 40% with BM and 41% with PBSC (P = 0.495), grade III-IV acute GVHD was 16% and 7% (P = 0.097) respectively. Chronic GVHD CI at 1 year was 24% vs 18% (P = 0.41), being extensive in 14% and 9% (P = 0.21) with BM and PBSC respectively. After a median follow-up of 15 months , estimated 2-year event-free survival (EFS) was 38% (95%CI: 24-52) vs 38% (95%CI: 26-50) with BM and PBSC respectively (P = 0.953), and overall survival (OS) was 47% (95%CI: 33-61) vs 46% (95%CI: 33-59) respectively (P = 0.985). CI of relapse or progression was 34% (95%CI: 23-51) with BM and 33% (95%CI: 25-45) with PBSC (P = 0.85). No significant differences were observed in terms of NRM, GVHD, EFS, OS and relapse cumulative incidence between BM and PBSC. Conclusion: HAPLO-HSCT with PTCy in the treatment of high risk hematologic malignancies, offers long-lasting remissions with manageable toxicity and GVHD, employing either BM or PBSC wich produce similar outcomes as graft source. Disclosure of Interest: None declared. Haploidentical Stem Cell Transplantationan effective rescue after alternative graft failure K. M. Toporska 1,* , J. Dykes 2 , B. Tomaszewska-Toporska 3 , S. Lenhoff 3 , J. Engellau 4 , J. Toporski 5 , D. Turkiewicz 5 1 Medical Faculty, Lund University, 2 Immunology and Transfusion Medicine, 3 Hematology, 4 Oncology, 5 Pediatric Oncology and Hematology, Skane University Hospital, Lund, Sweden Introduction: Graft failure is a relatively rare complication of hematopoietic stem cell transplantation (HSCT) that may be directly life threatening unless autologous recovery occurs. The challenge is to transplant the patient promptly and the timing depends on the availability of a suitable donor, possibly different than first one. Transplantation with a haploidentical donor offers the opportunity of to re-transplanting a patient at the optimal timing. Material (or patients) and methods: In this study, we present a series of 9 consecutive patients (7 children and 2 adults) who experienced a primary (5) or secondary (4) graft failure after an allogeneic HSCT and were rescued by a re-transplantation with a haploidentical donor. Details regarding the first transplant procedure are presented in Table 1 . The median time to the second SCT was 47 (35-156) days. If the first (failed) transplantation was a haploidentical one, a different haploidentical donor was chosen in all but one patient who only had one donor available. The first historical patients were retransplanted after a reduced chemotherapy-only conditioning regimen. The remaining patients received a uniformed conditioning regimen consisting of Fludarabine, Melphalan and 7 Gy Total Lymphoid Irradiation (TLI). In all cases serotherapy was used, preferably with a different preparation than the one used for the first procedure. The stem cell source was peripheral blood in all cases. GvHD prophylaxis consisted of ex-vivo immunomagnetic depletion (CliniMACS) either CD3 + or αβ+ T-cell and a short course of CellCept. The median dose of CD34+ and αβ+ T cells / kg was 18.1x10 6 and 9.0x10 3 , respectively. Legend to the Table 1 : Dgn, Diagnosis,; NB, Neuroblastoma; MPS VI, Mucopolysaccaridosis type VI; Flu, Fludarabine; Thio, Thiotepa; Eto, Etoposide; Mel, Melphalan; HD-MIBG, High Dose Metaiodobenzylguanidine; Bu, Busulfan; Cy, Cyclophosphamide; ATGF-F, Anti-thymocyte globulin Fresenius; Thymo, Thymoglobuline,. Results: All patients engrafted after the second SCT with a median time of neutrophil and thrombocyte recovery of 11 (6-13) and 12 (10-30) days respectively. Seven patients are alive and well, one relapsed day +120 from the second SCT, one died due to disease progression at day+189 from the second SCT. Neither acute nor chronic graft versus host disease was observed. No major transplant related morbidity. No TRM was observed. Three children received CD 45 RA depleted DLI for treatment of adenovirus reactivation and cleared the virus without developing overt disease. Conclusion: Reduced intensity conditioning +/-TLI and serotherapy followed by a haploidentical SCT of T-cell depleted graft offers a reasonable rescue for patients experiencing primary or secondary graft failure with no transplant related mortality and limited morbidity. Disclosure of Interest: None declared. Introduction: The BCR-ABL T315I mutation is reported to account for 4% to 20% of all mutations associated with TKI resistance. However, ponatinib hasn't come to market in Chinese mainland, and it's also worth noting that most patients in developing countries couldn't afford the economic burden of long-term ponatinib. Allo-SCT is regarded as the first-line option for CML patients with T315I mutation. In recent years, several published studies have also proved the effectiveness of SCT for CML patients with T315I. Nevertheless, the role of transplantation from haploidentical donors has not been well established. Material (or patients) and methods: Between Dec 1st, 2008 to March 1, 2015, 22 consecutive CML patients harboring T315I mutation received HID (16 cases) or MRD (6 cases) allo-HSCT according to donor availability at the Peking University Institute of Hematology. Results: Baseline characters and Pre-transplant information The majority of patients were males (95.5%) in the study. At the time of transplantation, 7 patients were in CP, 8 in AP or returning to CP post-AP (AP/AP-CPn), and 7 in BP or returning to CP post-BP (BP/BP-CPn). Engraftment and GVHD All patients achieved neutrophil engraftment within a median of 14 (range: 10-25) days. 19 patients got platelet engraftment, and the median time was 18 (range: 8-87) days.Thirteen of the evaluable patients (59.1%) developed acute GVHD (aGVHD), which was grade I-II in 11 (50.0%) patients and grade III-IV in 2 (9.1%) patients. Twelve (60.0%) of the 20 evaluable patients developed chronic GVHD (cGVHD, 7 limited and 5 extensive). Efficacy of transplantation and post-transplantation intervention 20 (90.9%) patients achieved a CMR at a median of 2 months (range, 1-6 months) after transplantation, whereas 2 (9.1%) patients did not respond to SCT. 6 patients lost their CMR at a median of 6 (range, 2-12) months, and 2 patients failed to achieve a MMR within the 3 months after SCT. Thus eight patients, including 4 in CP, 3 AP/AP-CPn and 1 BP pre-SCT, were administered with pre-emptive intervention at a median of 4.5 (range, 2-12) months after SCT in case of potential relapse. Relapse and transplantation related toxicities The cumulative incidence of 1-year relapse were in 0%, 12.5%, and 28.6% in CP, AP/AP-CPn and BP/BP-CPn, respectively (P = 0.047).4 patients suffered transplantation-related mortality (TRM). In detail, 2 were in BP/BP-CP2, 1 in AP-CP2 and the rest one in CP at the time of SCT. The cumulative incidence of 1-year TRM were comparable in the three groups (P = 0.808). Survival outcomes After a median follow-up of 17.3 (range, 2.7-52.9) months from SCT, 14 patients (63.6%) remained alive, and 13 (59.1%) were in CMR. 6 of 7 patients (85.7%) with CP-CML pre-SCT, 6 of 8 patients (75.0%) with AP/AP-CPn and only 1 of 7 patients (14.3%) in BP/BP-CPn were in CMR at the last follow-up. The 2-year estimated corresponding LFS were 80.0%, 72.9% and 0, respectively (P = 0.017). Conclusion: In summary, SCT was proved to be a curative tool for CML patients with T315I mutation, and HID-SCT was an alternative choice. A high rate of complete clinical and molecular response was achieved for patients transplanted in CP and AP/AP-CPn. For patients in CP/AP at the detection of mutation, timely SCT might result in promising outcomes. For patients in BP at the detection of mutation, SCT should be performed once returning to CP was achieved. Disclosure of Interest: None declared. Introduction: Haploidentical hematopoietic stem cell transplantation (HSCT) offers the advantage of being immediately applicable to virtually all patients who lack a HLA-matched donor. Graft manipulation with removal of all T lymphocyte subsets has been associated with an increased risk of lifethreatening infections and leukemia recurrence because of delayed immune reconstution. To improve the immune recovery, we have used a new T-cell depletion method that removes αβ+ T lymphocytes while retaining γδ+ T lymphocytes, NK cells and other cells in the graft. We report immune reconstitution data of acute leukemia patients transplanted with this approach. Material (or patients) and methods: We enrolled 34 acute leukemia patients. Median age was 28 years (range 18-60). The αβ+ T cells were depleted by using anti-TCRαβ-coated microbeads and the automated CliniMACS device (Miltenyi Biotec). B cell depletion was not performed. The conditioning regimen consisted of fludarabine (40 mg/m 2 , days -8 to -5), thiotepa (2 x 5 mg/kg, day -4) and melphalan (70 mg/m 2 , days -3, -2). Anti-thymocyte globulin (ATG-Fresenius) 15-30 mg/kg starting on days -12 to -9 was used to deplete remaining host T cells, avoiding graft rejection. Mycophenolate sodium was given as prophylactic immune suppression, if residual T cells in the graft exceeded 25 x10 4 / kg BW. Results: All patients received HSCT with αβ+ T cell depleted haploidentical grafts. The patients received a median number of ?12.69 x 106 CD 34+ progenitor cells per kg body weight (BW). In addition, grafts contained a median number of 4.58 × 106 per kg BW γδ T cells. The median residual αβ T cells was 11.72 × 103 cells per kg BW All but 3 patients engrafted with full donor chimerism and one of them died due to bacterial infection.The others were re-transplanted. Engraftment eventually occurred in 31 patients (91.2%). The median time to reach an absolute neutrophil count 40.5 × 109/L and a platelet count 420 × 109/L was 12 days (10-15) and 11 days (10) (11) (12) . Five pts developed acute GVHD. Eleven pts (30%) experienced grade I-VI acute GVHD. Two pts developed gut and liver grade IV acute GVHD (6%). Only 2 pts developed chronic GVHD. Four pts died because of disease relapse 7 pts died due to transplantation related mortality. Median follow-up is 191 days (range 35-933). The immune reconstitution were given Table. All patients and donor were CMV seropositive. CMV reactivaiton was seen every patients but CMV disease or EBV related post transplant lymhpoproliferatif disease were not seen. Conclusion: These data indicate that a selective graft manipulation results into effective prevention of both acute and chronic GVHD, high engraphmant rates, rapid recovery of neutrophil and platelet counts and low TRM. The use of TCRαβ-depleted stem cells together with a melphalan-based regimen resulted in improving immune recovery. Disclosure of Interest: None declared. Introduction: We analyzed the outcome of advanced HL patients who received haploidentical-HSCT (haplo-HSCT) using the unmanipulated stem cell and PT-Cy. Material (or patients) and methods: From April 2009, 62 consecutive HL patients undergoing haplo-HSCT with PT-Cy, in two institutions. The conditioning regimens were nonmyeloablative (NMAC) (TBI-based) or reduced intensity (thiotepabased). GVHD prophylaxis consisted of Cy 50 mg/kg administered on days +3 and +4, tacrolimus/cyclosporine A and MMF from day +5. Donors underwent to BM harvest or peripheral stem cell harvest. Results: Patient characteristics were reported in the Table 1 . The median day to obtain an unsupported ANC40.5x10 9 /L was 20 days (range 13-34). Fifty-one patients (82%) were evaluable for platelet engraftment and the median day to obtain an unsupported platelet count420x10 9 /L was 26 (range 11-49). The 100-day cumulative incidence (CI) of grade 2-4 and 3-4 aGVHD was 23% (95% CI 12-34) and 4% (CI 95% 0-11), respectively. The CI of cGVHD (overall) was 16% (95% CI 3-21) and only 1 patient developed severe form. aGVHD was the cause of death in 1 patient. Four patients had CMV disease (8%), 3 pneumonia and 1 gastritis, while no patients developed EBV-associated lymphoprolipherative disease. Infection was the cause of death in 3 patients: CMV pneumonitis (1), H1N1 pneumonitis (1), and interstitial pneumonitis (1) . The median follow-up was 32 months. The 3-year OS, PFS, relapse rate was 62% (95% CI 49-75), 58% (95% CI 45-71), and 21% (95% CI 10-32), respectively. The 1-year NRM was 21% (CI 95% 10-32). The 1-year NRM for patients receiving NMAC was 11% (95% CI: 2-20). In MV, the most consistent predictive factor was the disease status before haplo-HSCT while the use of PBSC enhanced the OS and PFS. the most frequently used stem cell source (n = 38, 70%) whereas peripheral blood stem cells (PBSC) were more recently used for aggressive diseases or donor ineligibility to marrow donation (n = 16, 30%). Sixty-eight% of patients (n = 37) were conditioned with myeloablative regimen consisting of thiotepa (tot. 10 mg/kg), busulfan (tot. 9.6 mg/kg) and fludarabine (tot. 150 mg/sqm). Graft-versus-host disease (GVHD) prophylaxis consisted of post transplant Cyclophosphamide on days 3 and 4 or 3 and 5 with calcineurine inhibitors and mycophenolic acid. All patients but one received biosimilar or originator granulocyte colony stimulating factor (G-CSF). Results: Fifty-three patients, for a total of 54 haplo HSCT, were enrolled. At a median follow-up of 18.4 months (range 1.5-53), median overall and event-free survivals were 24.2 and 14.5 months. Overall cumulative incidence of non-relapse mortality (NRM) was 16.7% and 21.5% at 3 and at 18.4 months respectively. Relapse incidence (RI) was 9.3% and 30.3% at 3 and at 18.4 months and was significantly lower in patients in complete remission at the time of HSCT (P = 0.04). Graft failure occurred in 2 patients, the cumulative incidence of neutrophil recovery was 92.6% at day +30, the median day of recovery was day +17 (day +18 and +15 for BM and PBSC respectively). No differences were observed between originator G-CSF versus biosimilar G-CSF (P = 0.795). Overall cumulative incidence of acute GVHD requiring systemic steroid therapy was 54.7% at day +100; acute GVHD grade III-IV occurred in 17% (5/29) of patients. Incidence of chronic GVHD was 47.4% at day +400. However, 52% of patients had stopped immunosuppressive drugs at median follow-up. Conclusion: Our single Institution experience on a very high risk population showed similar overall NRM and RI as reported by other groups though GVHD rates were somewhat higher. Moreover, biosimilar G-CSF was safely used in the setting of allografting partly reducing the cost of the procedure. Introduction: Prevention of GvHD relying solely on suppressing T cells has yielded limited success. Dendritic cells (DCs) play a pivotal role in GvHD and are impaired by proteasome inhibitors (PI). Ixazomib (IXZ) is a PI that dissociates rapidly from 20 S proteasome and is orally bioavailable. We aimed to examine IXZ's effects on dendritic and T cells in vitro, as well as in a MHC-mismatch GvHD murine model. Material (or patients) and methods: DCs were sorted from blood filters of healthy donors and cultured with IXZ at different concentrations for 4 h before stimulation with lipopolysaccharide (LPS) for 16 h. DCs were then stained with CD40, CD54, CD80, CD83, CD86, & CCR-7 antibodies and analyzed by flow cytometry. Survival of DCs cultured in the presence of IXZ was evaluated after staining by 7AAD. Proinflammatory cytokines were measured in the supernatant of DCs cultured in presence of IXZ, added before or after LPS. Standard MLR and T cell proliferation assays were used to evaluate the effects of IXZ on T cell proliferation. For the in vivo studies, a C57B6 → BALB/c pre-clinical GvHD model was adopted. Mice were transplanted with 5x10 6 BM cells after receiving 10 Grays of TBI in 2 fractions. Acute GvHD was induced by administering 5x10 6 splenocytes. Vehicle or IXZ (1.5 mg/kg) were given SQ as follows: schedule A = d -1 and +2; schedule B = d +1 and +4; and schedule C = d -1, +2, and +5. Mice were monitored and the results of repeated experiments were pooled. Alternatively, mice receiving vehicle or IXZ (schedule A or B) were sacrificed on day +7. Splenic DCs were sorted and assessed by FACS analysis after stimulation with LPS and staining with CD80, CD86, & MHCII antibodoes. The proliferation of CFSE-labeled lymph node CD4 and CD8 cells was quantified by FACS analysis after stimulation by irradiated BALB/c splenocytes. Results: IXZ suppressed the expression of DC maturation markers in a dose-dependent fashion starting at the concentration of 10 nM. IXZ also decreased the fraction of total DCs simultaneously expressing multiple markers. At the concentration of 10 nM, DC viability remained unaffected in comparison to control but dropped to 68% and 43% at 20 nM and 40 nM. When added before LPS, IXZ significantly decreased DC production of IL-6, IL-12, and IL-23 starting at the concentration of 20 nM. IL-1β was decreased at 40 nM. Importantly, there was no significant change in cytokine production by DCs when IXZ was added 4 h after LPS, except for IL-1β that increased at 30 nM. Starting at the concentration of 10 nM, IXZ dose-dependently inhibited T cell proliferation. In vivo, IXZ statistically improved mice survival with schedule B. Despite early improvement, IXZ failed to improve overall mice survival with schedule A. Schedule C showed improved survival that did not reach statistical significance, in part due a phenomenon of early death after drug administration on day +5. Schedule B showed impaired DC maturation and decreased T cell proliferation when compared to the control group. Conclusion: IXZ impairs DC maturation while partially preserving cell viability. IXZ also inhibits pro-inflammatory cytokine production by non pre-stimulated DCs and inhibits T cell proliferation. Finally, in an aggressive murine GvHD model, IXZ affects GvHD development in a schedule-dependent fashion. Our results provide support for rational design of clinical trials. Introduction: Endothelial cells (ECs) damage and endothelial microparticles (EMPs) have been recently proved to be implicated in the pathogenesis of acute graft-versus-host disease (aGVHD) after allogeneic hematopoietic stem cell transplantation (Allo-HSCT), 1, 2 however, the related mechanisms remain unclear. Material (or patients) and methods: The murine model of bone marrow transplantation (BMT) was performed by transplantation of bone marrow and spleen cells from donor mice (C57BL/6 for allogeneic BMT, BALB/c for syngeneic BMT) to recipient BALB/c mice. In recipient BALB/c mice, the apoptosis of ECs, the expression of hedgehog-interacting protein (Hip) and Fas protein, the production of reactive oxygen species (ROS) were in situ detected in aortic when aGVHD attack. The concentration of plasma EMPs and nitric oxide in recipient mice were also determined. In vitro, the cell apoptosis, intracellular production of ROS and NO, angiogenic activity in EA.hy 926 cells were examined after incubation with EMPs derived from TNF-α treated ECs. Moreover, the Hip carried by EMPs and the related Sonic hedgehog (Shh) pathway implicated in these effects of EMPs were analyzed. Results: Significantly higher numbers of EMPs at day 8 after BMT were found in mice with aGVHD compared with those without aGVHD (760 ± 79 vs 420 ± 54, P<0.05). The intensity of immunohistochemical staining for Hip protein in mice with aGVHD was determined as (++++), stronger than those without aGVHD, in which the intensity was determined as (+). The positive cells of TUNEL staining in aortic ECs with and without aGVHD were observed as (++++) and (± ), respectively. Increased ROS production in aorta and decreased plasma NO concentration were also determined in mice with aGVHD. In vitro, after treated with EMPs obtained from injured ECs, EA.hy926 cells exhibited a significantly increased apoptotic rate compared with resting cells (21.88 ±1.85% vs 7.92 ± 1.99%, P40.001) along with enhanced Fas and Bax, whereas reduced Bcl-2 expression. Meanwhile, increased ROS production and decreased NO generation were also found in EMPs treated ECs. In addition, incomplete and fluffy capillary-like structures, accompanied with decreased Ang-1 but enhanced Ang-2 expression were found in EMPs treated ECs. Hip is an antagonist for the the Shh pathway, these effects of EMPs on ECs were directly mediated by Hip to inhibit the Shh pathway, as illustrated by all effects of EMPs were reversed when the Hip incorporated in EMPs was silenced by siRNA or the Shh pathway in target cells was activated by SAG. Introduction: Acute Graft-versus-Host Disease (GvHD) is a major obstacle to allogeneic hematopoietic stem cell transplantation (HSCT). Due to the absence of diagnostic tools for predicting disease severity, a steroid-based first-line therapy is adopted, unless the poor outcome of refractory patients. We investigated if pentraxin (PTX)3, an acute phase protein locally produced in several inflammatory diseases, could represent a biomarker for improving GvHD management. Material (or patients) and methods: PTX3 plasma levels were firstly evaluated in a fully MHC-mismatched mouse model of acute GvHD. PTX3 levels were further evaluated in 115 pediatric HSCT patients with hemato-oncological diseases. Plasma samples were collected before conditioning regimen (basal), at HSCT, weekly until day 100, at GvHD onset. PTX3 was also evaluated by immunohistochemistry in GvHD lesions. To investigate its role in GvHD pathogenesis, rhPTX3 was injected in allo-transplanted mice. Results: In the GvHD mouse model, PTX3 levels, low before irradiation increased 8-folds, as effect of conditioning regimen. Thereafter, the protein decreased in both syngeneic and allogeneic transplanted mice, before raising again at GvHD onset only in the allogenic group (P = 0.03). PTX3 plasma levels were next measured in 115 patients who underwent HSCT at S. Gerardo Hospital, Monza and at the Regina Margherita Hospital, Turin. In accordance with our observation in the mouse model, PTX3 significantly increased from a basal of 7.19 up to 31.39 ng/ml at day 0, as effect of conditioning. PTX3 levels were next compared in patients experiencing or not the disease within 100 days after HSCT. At the onset of GvHD, PTX3 resulted significantly higher than in time-matched plasma samples of the No GvHD group. PTX3 expression was further analyzed by immunohistochemistry on tissues typically affected by the pathology. Interestingly, while a basal perilymphatic protein expression was detected in healthy skin and in the subepithelial extracellular matrix of healthy colon, a strong PTX3 production was highlighted in skin and colon from patients with active disease. To evaluate the prognostic significance of PTX3, we firstly correlated protein levels at disease onset with GvHD severity. PTX3 resulted significantly increased in patients with severe GvHD, compared to patients with milder disease. We next correlated PTX3 levels with therapy response. Interestingly, PTX3 resulted 3-fold higher at the time of disease onset in patients who had no response after 4 weeks, compared to patients who experienced a complete or partial response (P = 0.05). To assess whether, besides representing a GvHD biomarker, it could also play a role in disease pathogenesis, we repeatedly injected recombinant human (rh)PTX3 in our mouse model of GvHD. No differences were observed between animals treated or not with the molecule in terms of overall GvHD score and histopathology on GvHD lesions, thus ruling out a direct effect of the molecule on disease course. Conclusion: Our results candidate PTX3 as a reliable biomarker reflecting GvHD severity and therapy responsiveness. If these results will be confirmed in a larger cohort of patients, PTX3 could represent a clinically relevant tool for tailoring patient-specific anti-GvHD therapy soon as the pathology occurs. Disclosure of Interest: None declared. Validation study of MBL2 and TYMP polymorphisms and lymphocyte populations and inflammatory serum cytokines in day +28 as predictors of acute graft-versushost-disease aGvHD after allo-HSCT J. Beltran 1,* , V. Guillem 1 , P. Introduction: The pathogenesis of GvHD is related to danger signals that activateT and other immune cells, which induce production of inflammatory cytokines, recruiting immunological effector cells and destroying tissues and organs. Predictive diagnostic, incidence and risk biomarkers of GvHD are critically needed to improve treatment of this serious complication. Material (or patients) and methods: Between May 2013 and December 2014,104 patients received peripheral blood stem cell transplantation in one of five participant centers. Patients characteristics: Age, years: 49.0 (range 23-69), Sex (F/M): 49/55. Female donor/male recipient pair: 12.5%. Diagnosis: acute leukaemia (27.9%), lymphomas (48.1%), SMPC (5.8%), other haematological disorders (17.3%). Conditioning regimen was myeloablative (MA) in n = 15 (14.4%) of the pts and reduced intensity conditioning (RIC) in n = 89 (85.6%). Patients were allografted from MRD (n = 51, 49.0%), MURD (n = 22, 21.2%), MMRD (n = 2, 1.9%) MMURD (n = 14, 13.5%). Blood samples were analyzed by flow cytometry for CD3+, CD4+, Tregs, CD8+, CD4+CD8+, CD19+, total NK cells, as well as IL-6 and TNF-a serum levels cytokines by ELISA immunoassay in samples obtained pre allo-HSCT and on days 0, +14, +21, +30, +60 and +90 days post-SCT. In 65 patient-donor pairs, TYMP and MBL2 polymorphisms (rs112723255 G4A and rs930508 C4G) by Taqman s SNP genotyping assay were determined as validation cohort of a previous study. Patients were divided into nomild (n = 64, 61.5%) and moderate-severe (n = 40, 38.5%) aGVHD and no-mild (n = 87, 83.5%) and moderate-severe (n = 17,16.4%) cGvHD. Results: After a median follow-up of 13.9 months (range 0.3-30.8), 37 patients have died due to relapse (n = 7, 35.6%), aGvHD (n = 6, 18.9%9), cGvHD (n = 1, 2.7%), toxicity (n = 2, 5.4%), infection (n = 15, 40.5%), other (n = 6, 16.2%). FemaleD to maleR aGVHD risk (HR = 2.306; P = 0.036). We found significant differences in IL-6 and TNF-alpha serum levels at day +21 post allo-HSCT between patients who eventually developed aGvHD and those who not (P = 0.001 and P = 0.007, respectively). We also found differences between PB populations CD4 + CD8 + (n = 37, P = 0.024), CD 8 + (n = 36, P = 0.005), CD 56 + (n = 15, P = 0.03) and NK CD16 -(P = 0.038) at day +28 post allo-HSCT. The study confirms previous results detecting that MBL-2 heterozygous patient-donor combination genotype (C/G in patient, donor or both) was associated with a decreased risk of developing aGvHD (HR = 0.273; P = 0.036). Conclusion: T cell and NK lymphocyte subpopulations in addition to serum levels of IL-6 and TNF-alpha at day +21 predicts aGvHD after alloHSCT. In this study, we validated MBL-2 polymorphisms (rs930508 C4G) in D/R as a significant risk for aGvHD. Disclosure of Interest: None declared. Increase of endothelial progenitor cells in acute graftversus-host disease after allogeneic haematopoietic stem cell transplantation for acute myeloid leukaemia M. Medinger 1,* , D. Heim 1 , S. Gerull 1 , J. Halter 1 , W. Krenger 2 , C. Bucher 1 , P. Jakob 3 1 Hematology, 2 GMP Facility, 3 University Hospital Basel, Basel, Switzerland Introduction: Endothelial progenitor cells (EPCs; CD31 +CD34 bright CD133+CD45 dim cells) are novel markers of endothelial dysfunction and are related to inflammatory processes such as acute graft-versus-host disease (aGvHD) mediated by anti-host reactive, donor-derived T cells. Because there is no established second line treatment, markers for an early steroid response are clearly needed for translational studies that aim to treat steroid refractory acute GvHD as early as possible. Recent evidence indicates that bone marrowderived EPCs contribute to endothelial cell renewal in hypoxic and inflamed tissuesmaking EPCs a putative early marker for steroid response. Material (or patients) and methods: To study our hypothesis that EPCs are involved in aGvHD and that their cell numbers are altered under therapy, we examined 47 patients with acute myeloid leukaemia (AML), who had undergone allo-HSCT with myeloablative conditioning with PBSC as stem cell source in complete remission. Blood samples for the quantitative analysis of circulating EPC levels was drawn before the begin of conditioning (baseline), on the day of engraftment, at diagnosis of aGvHD and initiation of aGvHD treatment and day 3, 7 and 14 after initiation of steroid treatment (in patients with aGvHD) and in nearly the same time points since HSCT in patients without aGvHD. EPC were identified by flow cytometry as CD34+VEGFR2/KDR+CD133+ triple-positive cells among CD34+ cells. Results: 22 patients had at least grade II acute GvHD at a median of 20 days post-transplant. At baseline (before begin conditioning) and at the time of engraftment (median day 15), the circulating EPC level was not significantly different in patients with and without aGvHD (Figure 1a ). At diagnosis of aGvHD≥grade 2, EPC levels increased whereas in patients without aGvHD the EPC levels were significant lower (3021 ± 278 versus 2322 ± 195 cells/ml; P40.001; Figure 1a ). After a median of 6 days of steroid treatment, the number of EPCs returned close to the number EPCs of patients without aGvHD at day 14 of therapy in steroid-sensitive patients. However, in patients with steroid-refractory aGvHD, EPC levels remained at a high level until day 14 of therapy whereas responders to therapy had a decrease of EPC levels (2300 ± 155 versus 1287 ± 112 cells/ml; P = 0.04; Figure 1b ). Conclusion: Taken together, our results demonstrated that the number of circulating EPCs is increased in patients with aGvHD compared to patients without aGvHD. In patients with steroid-sensitive aGvHD, EPC levels decreased to the range before allo-HSCT whereas in steroid-refractory patient EPC numbers remained at high levels. Further studies must validate, whether EPC levels can function as biomarker for steroid responsiveness in aGvHD. Disclosure of Interest: None declared. Introduction: In a breast cancer model multi-donor transplantation (MDT) proved to suppress GVHD and augment the anti tumor activity. In this study we sought further analysis of MDT with a new tumor model and possible mechanism(s). Material (or patients) and methods: BALB/c mice received a single fraction TBI (7.5 Gy) followed a day later by BMT from either C57BL/6, C3H or both (MDT). The recipient mice were inoculated with syngeneic melanoma cells and the development of melanoma metastases was followed along with post transplant immune cell subpopulations recovery. In order to examine whether the second graft in MDT needs to be able to proliferate, retain its antigenicity or merely circulate in the recipient in order to achieve better GVT, we irradiated one of the grafts (C57BL/6 or C3H) with high dose ionizing gamma rays or by treating the cells with photosensitizing agent followed by UVA irradiation. The combination of an unmanipulated graft and a manipulated graft was then transplanted. Results: Measured by the Cooke grading system, MDT showed significantly less GVHD ( p40.01) as compare to the controls. Thirty days following transplantation from BALB/c into BALB/c (control), C57BL/6, C3H or MDT and inoculation with B16BL6 mouse melanoma tumor cells the lungs were excised lung nodules were enumerated. As expected, the C57BL/6 and C3H showed GVT effect with significant reduction of the number of metastases as compared to the control (figure 1). The MDT group, however, showed a much stronger GVT effect (P = 0.001 and P40.0001 as compared to C57BL/6 and C3H respectively, figure 1 ). We found that both gamma irradiation and photosensitization followed by UVA of one graft prevents the augmented GVT (figure 1). Post transplant immune cell subpopulations recovery post MDT showed preservation of cellular naivety, increased number of NK cells and increased APC. Conclusion: We conclude that MDT's augmentation of the anti tumor effect is not tumor specific. We also showed that the second graft in MDT needs to be able to proliferate and retain its antigenicity. The immune recovery explains the suppression of GVHD and the increased GVT. Disclosure of Interest: None declared. The Heme Oxygenase-1 gene promoter polymorphism is associated to outcome in patients receiving HLA-identical sibling allogeneic peripheral blood stem-cell transplantation ( Introduction: Graft-versus-host disease (GvHD) is the main cause of early mortality in allogeneic stem-cell transplantation (allo-SCT), even in HLA-identical allo-SCT. The heme oxigenase-I (HO-1) is an enzyme that catalyzes the heme group degradation. It has been related to inflammatory response regulation, with a potential role in the immune system regulation. Therefore, it could have a role in the development of GvHD. The HO-1 promoter region is polymorphic due to a variable number of GT repeats (higher protein expression when o25 repeats). Up to now, only one report has analyzed the role of this polymorphism in the allo-SCT setting, mainly myeloablative transplants, showing that the presence of o30 repeats in the donor associates to a higher transplant-related mortality and a shorter overall survival (OS). However, there is no information related to allo-SCT with reduced intensity conditioning (RIC). We aim to analyze the role of the polymorphism in HO-1 gene promoter in the outcome of patients undergoing HLA-identical sibling allo-PBSCT with RIC. Material (or patients) and methods: A total of 126 patients diagnosed of AML (n = 54), MDS (n = 31), NHL (n = 18), MM (n = 9), HL (n = 6) or others (n = 8) from five different Spanish centers were included in the study. Flu-Bu conditioning was used in 71% of the transplants while Flu-Mel in 29%. The analysis of HO-1 polymorphism was carried out as described (Gerbitz). According to the HO-1 genotype, individuals were grouped into S/S, S/L, or L/L, considering S any allele with ≤ 25 repeats, and L when 425 (Rueda). The analyses were carried out taking into account the receptor, donor and both profiles. Allele frequencies were estimated by direct counting and compared between groups using Fisher's test. Log-rank analysis was used to compare differences between survival curves. Results: Cumulative incidence of aGVHD at day 200 was 52%, and grades II-IV in 44% in our series. There were no statistical differences in clinical variables between HO-1 genotype groups. Recipients carrying L/L or S/L genotype showed a higher incidence of grade II-IV GVHD (49% vs. 18%, P = 0.06) as compared to S/S genotypes. Other variables influencing grade II-IV GVHD were conditioning (Flu-Mel, 64% vs. 38%, P = 0.001), intermediate/advanced disease status at transplant (53% vs. 37%, P = 0.03), and recipient age (460y, 56% vs. 38%, P = 0.04). In the multivariate analysis, Flu-Mel conditioning (HR 2.5, 95% CI: 1.4-4.3) , and recipient age 460y (HR 1.9, 95% CI: 1.1-3.3) significantly associated to aGVHD severity, while HO-1 genotype showed a trend (HR 2.6, 95% CI: 0. 8-8.4 ). There was no relation between HO-1 polymorphism and incidence/severity of cGVHD, or OS. The HO-1 donor genotype did not influence alo-TPH outcome. Conclusion: The present data suggest an association between the HO-1 polymorphism in recipients with severe aGVHD. Although confirmation in larger series is required, the HO-1 polymorphism could be used to design personalized strategies in GVHD prophylaxis patients undergoing allo-RIC SCT. Introduction: RR-GIT is the dominant complication during the pre-engraftment period and a potential acute graft-versushost disease (aGVHD) risk factor. According to our hypothesis functional variants of genes participating in DNA-damage response (DDR) may have an impact on the extent of tissue damage caused by the conditioning regimen. Material (or patients) and methods: In our single-center study, we analyzed 109 patients (female 41%) allografted between 2009-2015 for: AML (41%), MDS/MPN (19%), ALL (17%), CLL (8%), NHL (6%), CML (4%), and other hematological disorders (5%). The median age of the cohort was 50 (20-63) years. Patients were allografted after myeloablative (14%), non-myeloablative (20%) and reduced intensity/toxicity (66%) conditionings from HLA identical donors (matched-related 40%). GVHD prophylaxis was done by solo cyclosporine-A (76%) or cyclosporine-A and mycophenolate mofetil (20%) or low-dose methotrexate (4%). "In vivo" T-cell depletion with thymoglobuline was used in 69% of recipients. The patients were genotyped for five SNPs (rs4585 T/G, rs189037 A/G, rs227092 T/G, rs228590 C/T, and rs664677 T/C) of the ATM gene. SNPs genotyping was performed with Sequenom MassARRAY platform using allele-specific MALDI-TOF mass spectrometry assay (Sequenom, San Diego, CA, USA). Primers were designed using the Sequenom SNP Assay Design software version 3.0 for iPLEX reactions. Logistic regression was used in the study with significance level set to 0.05. Results: 35 patients (32%) died (21 patients died of transplantrelated complications). The median post-transplant follow-up was 1.3 (0.1-5.7) years. RR-GIT grade III and IV was documented in 30 patients (28%), RR-GIT grade I and II in 48 patients (44%), and no RR-GIT in 31 patients (28%). The conditioning regimen, ATM gene haplotypes (rs4585*T, rs189037*A, rs227092*T, rs228590*C, and rs664677*T), female sex, and EBMT score were associated with RR-GIT grade III-IV [OR = 3.0, 95%CI(1.8-5.1) P = 0.004; OR = 9.7, 95%CI(2.0-47.0), P = 0.006; OR = 2.4, 95%CI(1.0-5.6), P = 0.05; OR = 2.1, 95%CI (1.4-3.2) , P = 0.0001, respectively]. The effect of the ATM haplotype was maintained after accounting of the above mentioned risk factors in the multivariate analysis (P = 0.006). There was no effect of the ATM haplotype on aGVHD occurrence. Conclusion: This is an update of our recently published preliminary single-center study analyzing ATM gene SNPs in allografted patients after reduced conditioning. After the inclusion of myeloablative conditioning, the significance of the ATM gene haplotype in the prediction of high-grade RR-GIT has been confirmed. The linkage with aGVHD has not been proven, suggesting ATM gene SNPs seem to be more tightly associated with the initial tissue damage whereas there are more clinical and pathophysiological variables involved in aGVHD. Of note, the rs189037*A is known to be associated with decreased ATM mRNA expression. Thus, we hypothesize that patients defective in DDR mechanisms due to insufficient ATM production may be rendered at higher risk of conditioninginduced tissue damage. Disclosure of Interest: Supported by IGA MZCR NT12454 and IGA_LF_2016_001. None declared. Introduction: Steroid refractory acute graft versus host disease (SR-aGVHD) is associated with poor outcome and no commonly accepted salvage therapy is available for its treatment. Bone marrow derived mesenchymal stromal cells (MSCs) might be an option to tackle this problem. Material (or patients) and methods: All patients had grade III-IV SR-aGVHD after allogeneic hematopoietic stem cell transplantation (HSCT) or donor lymphocyte infusion. MSCs target dose was 1x10 6 cells/kg bodyweight intravenously once weekly (maximum 3 doses). Treatment response was evaluated on day 28 after the first MSCs infusion. Healthy donors' bone marrow derived MSCs for therapeutic use were obtained by directly loading donor bone marrow or after 1 passage in flasks on the closed automated device bioreactor (Quantums Cell Expansion System, Terumo BCT, Inc, Lakewood, CO, USA) in media containing 5% platelet-lysate. The samples taken from the final product were analyzed for viability, cell count, phenotype, microbial growth, mycoplasma DNA and endotoxins. MSCs showed capability to differentiate into adipocytes, chondrocytes, osteocytes. Harvested MSCs batches were aliquoted into doses of 25 or 50 x 10 6 cells and cryopreserved. The data were collected prospectively. Results: 17 adult SR-aGVHD (14 III* and 3 IV*) patients received MSCs as salvage treatment. 16 patients had gastrointestinal, S164 5 patients had skin and 5 patients had liver involvement. Number of organ systems involved were 1 (9 patients), 2 (7 patients) and 3 (1 patient). The median time between the onset of aGvHD and the first infusion of MSCs was 20.5 days (range, 5-126 days). 8 patients received three, 8 patients two and 1 patient one infusion. Median MSCs dose was 1.05x10 6 cells/kg bodyweight (range, 0.83-1.33). The median observation time was 2 months (range, 0-23). 6 patients (35%) achieved CR, 1 patient (6%) achieved PR, 4 patients (24%) showed SD, 6 patients (35%) could not be evaluated (4 patients died before reaching day 28 because of infectious complications (all were non-responders at the last evaluation before death), 1 patient had hematologic disease relapse, 1 patient was missing to follow-up). In the group of patients who reached CR/PR on day 28, median survival was not reached (range, 1-23) vs 1 month (range, 1-12) in non-responders. At the last follow up, 7 of 17 (41%) patients were alive and 10 died (59%) (the causes of death: infection -7, hematologic disease relapse -3). No acute infusion-related reactions associated with MSCs were observed. There were 3 CMV and 2 EBV reactivations requiring treatment after the MSC infusion. Conclusion: MSCs are safe and well tolerated therapy option for patients with SR-aGVHD. Responders on day 28 may have a better prognosis. New strategies to increase as well as maintain response to MSCs are needed. Disclosure of Interest: None declared. Introduction: Ocular complications develop in a substantial percentage of patients after allogeneic stem cell transplantation (HCT). This has the potential to lead to blindness and impairment of quality of life. Ocular graft -versushost disease (GVHD) develops in 40-60% of patients after allo-HCT and 60-90% of patients with acute or chronic GVHD. Certain complications such as dry eyes (keratoconjunctivitis sicca (KCS) is reported to be the most common ocular complication in allo-HCT survivors and quite commonly the presenting symptom of chronic GVHD in most patients. Material (or patients) and methods: Aim. · To assess the incidence and risk factors of ocular complications after HCT · To determine the morbidity associated with ocular complications. Patients and methods: After obtaining IRB approval, all consecutive adult patients who underwent HCT from 01/ 2005 until 06/ 2014 were included. Patient records were reviewed for type of transplant-allogeneic vs. autologous, conditioning chemotherapy, complications after HCT. Ocular complications were recorded using ICD-9 codes. Results: A total of 3557 patient records were reviewed. 288(43%) of the 677 allo-HCT and 258(9%) of the 2878 auto-HCT patients had ocular complications. Among the allo-HCT patients 578 patients received peripheral blood, 64 bone marrow and 34 umbilical cords as the stem cell source. Only 10 among the auto-HCT received bone marrow, the others received peripheral blood. Diagnostic Groups are shown in Table 1 . Other diagnoses in autologous group include Amyloidosis(358), CNS tumors (16), Neuroblastoma(4), Testicular cancer(24). Other malignancies were 29 in the autologous cohort and 1 in the allogeneic cohort. Among the auto-HCT, 86 developed cataracts (32 of them had posterior subcapsular cataract), 26 glaucoma, 21 retinopathy and 14 corneal ulcers. In the allo-HCT patients, cataract was the most common eye complication accounting for 93 patients (51 had posterior subcapsular cataract). Corneal ulcers and keratitis accounted for 41 patients and KCS was seen only in 17 patients. (Figure 1 ). Conclusion: To our knowledge this is the largest study of ocular complications in both allo and auto-HCT survivors. Cataract was the most common eye complication encountered in the post-HCT setting accounting for more than 50% of eye conditions. [P105] Disclosure Introduction: Despite recent advances in haematopoietic stem cell transplant (HSCT) resulting in improved outcomes, graft-versus-host-disease (GVHD) remains a leading cause of morbidity and mortality. Extracorporeal photopheresis (ECP) is an alternative second line therapy for treatment of GVHD (1). There is scant information regarding use and safety of ECP in low bodyweight paediatric patients. We reviewed use of ECP in the management of acute (a) and chronic (c) steroid refractory/dependant GVHD in paediatric patients post-HSCT. Material (or patients) and methods: Hospital records of 19 patients were reviewed. Patients with a/c GVHD post-HSCT treated with ECP were included. Clinical details recorded included demographics, underlying diagnosis, details of the HSCT and conditioning regime, immunosuppressive agents used, type of GVHD and stage, response to ECP and overall outcome. Results: 13 primary immunodeficiency (PID) and 6 oncology patients (13 males) received ECP from December 2012 -December 2015; 4 patients still receive therapy. Median age at HSCT was 54 months. 2 had cGVHD, and 17 had aGVHD (11 skin only, 3 skin and GIT, 1 skin and lung, 1 liver and GIT, 1 skin, liver and GIT). Among the PID group, median duration from onset of GVHD to commencement of ECP was 79 days (range 22-327) and median weight was 14.8 kg (range 6.8-29.5 kg). Among the oncology group, the median duration from onset of GVHD to commencement of ECP was 91 days (range 17-276) and median weight was 55 kg (range 12-86 kg). All patients received corticosteroids plus at least one other immunosuppressive agent prior to ECP, with most patients on multiple agents. Within both groups, adverse side effects included fainting due to use of an infected line and bleeding from the central line site, but otherwise ECP was well tolerated. Immune suppression was reduced in 14 patients and 2 were off immune suppressive therapy completely while being treated with ECP. All 6 oncology patients had a partial response to ECP, but 3/6 (50%) patients died from intercurrent infections. Among the 13 PID patients, 5 (38.5%) had a complete response, 7 (53.8%) had a partial response and 1 (7.7%) had no response. 5/13 (38.5%) died, 2 from intercurrent infections, 1 from PTLD, 1 from inflammatory pneumonitis and 1 due to GVHD. Conclusion: ECP is safe and well tolerated in low bodyweight patients. Partial and complete responses were seen in the majority of patients and ECP treatment helped to allow reduction in immune suppression. There are remaining questions to be answered regarding the use of ECP in the management of GVHD including optimal timing of initiating treatment, how to reduce concomitant immune suppression and how to identify those patients who will respond to ECP treatment. Introduction: There is no validated consensus for treatment of steroid refractory chronic graft-versus -hostdisease(cGVHD). Extracorporeal photopheresis, rituximab, imatinib, mTOR inhibitors, methotrexate and more recently, ruxolitinib are used as secondary line treatment of chronic GVHD.The choice of the therapy will be determined by the transplantation team experience and patient GVHD characteristics. In 2011 J. Koreth and collaborators published the results of their phase 1 study using low-dose interleukin-2 to suppress clinical manifestations of chronic GVHD. Preclinical evidence for such strategy relied on IL-2 potency to enhance regulatory T cells function in vivo. Material (or patients) and methods: We are presenting the efficacy of a treatment with low dose of IL 2 for glucocorticoid refractory cGVHD. Subcutaneous IL2 was administred daily at the dose of 1X 10 6 UI /m 2 for 28 days followed by 14 days without treatment. The cycle was started again with the same schedule. Results: 15 patients received IL 2 for chronic GVHD from January 2013 to September 2014. The median age was 30 years. There were 9 females and 6 males. Regarding the stem cell source, 12 patients received peripheral stem cell transplantation, 1 had bone marrow transplantation and 2 cord blood unit. The median number of targeted chronic GVHD sites per patient was 3. Patients have received a median of 5 prior treatments for chronic GVHD before the administration of low dose IL2. The IL2 treatment was initiated with a median of 2 years after the first treatment for chronic GVHD. Median time of IL2 treatment duration was 3 month. We are reporting 1 complete response. 5 patients presented a partial response. Most of the observed responses concerned the skin localization of chronic GVHD, with improvement of the sclerotic features and ulcerations. 3 patients had stable disease with the IL-2 treatment. 4 patients continued to worsen their condition. One patient had to prematurely stop the treatment because of arthralgia. One patients presented fever each IL2 infusion and could'nt continue the treatment. 6 patients presented minor side effects. Only 4 patients could taper the other immunosuppressive agents. 9 patients presented infections during IL2 treatment. We counted 8 death mainly related to infection and/or GVHD progression. There was no relapse of the hematological disease. The Treg were studied in 6 patients. We observed an increase of absolute number of Treg one month after the beginning of the treatment. Conclusion: Low dose IL2 treatment could be one of the treatment offered to the patients with glucocorticoid refractory GVDH. The treatment is well tolerated, but most of the responses were partial. The exact schedule of the treatment infusion is not known. The results of the phase 2 study of the J. Koreth team were presented in ASH 2014, including 35 patients. Treatment seems to be more efficient if started rapidely after the chronic GVHD onset. Supplementary studies are required to confirm the efficacy of IL2 treatment in setting of GVHD. The place of IL2 in GVHD treatment strategy should be assessed, in comparison with JAK2 inhibitors and extracorporeal photopheresis. Disclosure of Interest: None declared. Sirolimus, tacrolimus and low-dose methotrexate as GVHD prophylaxis is safe and associated with promising results in patients receiving reduced-toxicity allogeneic hematopoietic cell transplantation A. M. Avila 1,2,* , P. Barba 2,3 , C. Introduction: The combination of sirolimus, tacrolimus and low dose methotrexate (MTX) has been increasingly used as graft-versus-host disease (GVHD) prophylaxis in patients receiving reduced-toxicity conditioning (RTC) allogeneic hematopoietic cell transplantation (allo-HCT). However, the impact on non-relapse mortality (NRM) and mortality due to GVHD of this combination have not been fully assessed. Material (or patients) and methods: We evaluated all consecutive patients with lymphoid malignancies undergoing RTC allo-HCT and receiving sirolimus, tacrolimus and low dose MTX as GVHD prophylaxis in our center from 04/2008 to 12/2014. Tacrolimus and Sirolimus doses were adjusted to maintain serum levels of 3-12 and 5-10 ng/mL, respectively, until day +60 and then tapered on an alternated schedule. MTX was used at 5 mg/m2 on days+1, 6 and 11. Results: A total of 135 patients with a median age of 54 years (range 23-75) were included. Most frequent diseases included diffuse-large / transformed B-cell non-Hodgkin lymphoma (n = 32, 24%), follicular lymphoma (n = 22, 16%) and chronic lymphocytic leukemia (n = 22, 16%). Eighty-seven percent of patients had chemosensitive disease at transplant (50% CR, 37% PR). Patients received non-myeloablative (n = 99, 73%) or reduced-intensity (n = 36, 27%) conditioned allo-HCT mostly from HLA-matched related (n = 58, 43%) or unrelated (n = 65, 48%) donors (URD). ATG was used in 56 (41%) patients. Median follow-up time was 34 months (range 0-88). Compliance with the planned GVHD prophylaxis scheme was achieved in 109 (81%) patients. Main reasons for treatment discontinuation included acute renal failure (n = 18, 70%) and liver abnormalities (n = 2, 8%). Cumulative incidences of 1-year grade 2-4 and 3-4 acute GVHD were 36% (95%CI 28-44) and 6% (95%CI 3-11), respectively. Patients450 years (P = 0.002) and those receiving mismatched URD (P = 0.02) were more likely to develop grade 2-4 acute GVHD. No risk factors could be identified for the development of grade 3-4 acute GVHD due to low number of events. The use of ATG and conditioning intensity were not associated with higher risk of acute GVHD. One and 2-year cumulative incidences of chronic GVHD were 24% (95%CI 18-32) and 42% (95%CI 34-51), respectively. Forty-three patients (32%) had died at last follow-up (21 from disease relapse, 19 from GVHD and 3 from other causes). NRM at 1 and 3 years was 8% (95%CI 4-14) and 15% (95%CI 9-22), respectively while OS at 1 and 3 years was 84% (95%CI 78-91) and 71% (95%CI 64-80). No risk factors could be identified for the development of NRM and OS among pre-HCT variables. Patientso 50 years had a promising outcome with 3-year NRM and OS of 5% and 76%, respectively. Using a landmark analysis at day +100, patients with acute GVHD grade 2-4 had higher 3-year NRM (34% vs. 10%, P = 0.01) ( Figure) and a trend to lower 3-year OS (60% vs. 76%, P = 0.1) than patients with acute GVHD grade 1 or no GVHD. Conclusion: GVHD prophylaxis with sirolimus, tacrolimus and low-dose MTX is feasible, relatively safe and associated with low incidences of GVHD, especially grade 3-4 acute GVHD. Introduction: Background. Patients with chronic GvHD (cGvHD) have a lower incidence of leukemia relapse, and this has been called the graft versus leukemia effect, or GvL. However, cGvHD is also the first cause of late non relapse mortality, due to the involvement of several vital organs, together with prolonged immune deficiency. The question therefore is, how do these two opposing effects balance out long term. Aim of the study. To assess 20 year survival of patients with or without chronic GvHD. Material (or patients) and methods: Patients. Eligible for this study were 1621 patients with hematologic malignancies, alive on day +100 after an allogeneic transplant grafted between 1978 and 2013. Patients were divided in acute and chronic malignancies. Acute malignancies (n = 769) included AML (n = 471) and ALL (n = 298); chronic malignancies (n = 852) included chronic myeloid leukemia (CML) (n = 372), myelofibrosis (MF) (n = 78), lymphoproliferative disorders and myeloma (n = 199), myelodysplastic syndromes (MDS) (n = 164), other (n = 43). [P108] Results: Results. The number of patients with no, minimal, moderate and severe cGvHD was 365, 737, 397, 122, respectively, which corresponded to 22%, 45%, 25%, 8%. The incidence of moderate/severe cGvHD was higher in the chronic disorders (36%) as compared to the acute leukemias (26%) ( p40.0001). Acute Leukemias. The cumulative incidence (CI) of transplant related mortality (TRM) at 20 years was 8%, 10%, 28%, 69% for cGvHD no, minimal, moderate, severe. The CI of relapse related death (RRD) was 56%, 37%, 24%, 19%. The actuarial 20 year survival was 34%, 52%, 47%, 16% respectively. Minimal cGvHD produced best survival at 20 years. Chronic malignanices. For chronic malignancies the CI of TRM at 20 years, was 21%, 15%, 38%, 45% for cGvHD no, minimal, moderate, severe. The CI of RRD at 20 years was 42%, 25%, 18%, 20%, and the actuarial survival 35%, 56%, 42%, 31%. Again minimal cGvHD produced best survival at 20 years. In multivariate Cox analysis on RRD, minimal cGvHD reduced the risk of relapse related death to less than 50% as compared to no cGvHD. The results can be summarized as follows: (1) minimal cGvHD protects from relapse, and produces best 20 year survival; (2) severe cGvHD produces worst survival at 20 years, both in acute and chronic malignancies; (3) survival of moderate cGVHD is worse than minimal cGvHD; (3) . the absence of cGvHD (no cGvHD) is better than severe cGvHD. Conclusion: Conclusions. Minimal cGvHD has a protective effect on leukemia relapse as compared to no cGvHD, and low TRM; therefore this is the best option for patients with hematologic malignancies undergoing an allogeneic stem cell transplant. Moderate and severe cGvHD must and can be prevented, as shown by 4 prospective randomized trials, using antithymocyte globulin (ATG) in the conditioning regimen. Disclosure of Interest: None declared. Introduction: Despite remarkable progress in the field of allogeneic hematopoietic stem cell transplantation (allo-SCT), the rate of transplant-related mortality remains high. Steroidrefractory acute graft versus host disease (SR aGVHD) is a specific complication of allo-SCT which lacks effective management and bears poor prognosis. No randomized trials were conducted and only few studies comparing extracorporeal photopheresis to anticytokine therapy (ACT) were reported. Material (or patients) and methods: A total of 128 allo-SCT recipients were included in the study. All developed SR aGVHD that was defined as no response after 3 days of methylprednisolone or aGVHD flare during the corticosteroid tapering. ECP was used as further therapy in 68 (53%) patients and ACT in 60 (47%) patients. ACT consisted of daclizumab (monoclonal antibody (MAB) to interleukin-2) in 17 patients, infliximab (MAB to necrosis factor alpha (TNF-α)) in 19 patients and etanercept (a fusion protein against TNF -α) in 24 patients. Etanercept was administered s.c. 0.4 mg/kg twice a week for 3 weeks, infliximab 10 mg/kg i.v. weekly for 3 weeks and daclizumab 1 mg/kg, i.v. twice a week for 3 weeks. The schedule of ECP was individual, ranging from 1 to 2 procedures per week. The median number of ECP procedures was 6 (2-23). Both groups (ECP and ACT) were homogenous in terms of such characteristics as median age (17 vs. 19 years), sex (male 57% vs. 53%), diagnosis (acute leukemia 77% vs. 65%), disease status (remission 52% vs. 60%), donor type (unrelated donors 79% vs. 84%), transplant source (bone marrow 62% vs. 51%), respectively. But ECP group and ACT group were heterogeneous in terms of aGVHD stage (III-IV stage 38% vs. 70%), number of affected organs (≥2 organs 37% vs. 60%) and conditioning regimen (myeloablative 60% vs. 31%), respectively. Response to therapy, overall survival (OS) and cumulative incidence of relapse were analyzed. Results: Overall response to therapy was observed in 45 patients (66%) after ECP and in 37 patients (62%) after ACT (P = 0.53). Complete responses were registered in 21 patients (31%) and 20 patients (33.5%), respectively. Partial responses were diagnosed in 24 cases (35%) and 17 cases (28.5%), respectively. Three-year OS in ECP group was 33% and in ACT group 20% (P = 0.05). The cumulative incidence of relapse at 3 years after ECP was 12% (95% CI 5-23) and 14% (95% CI 6-25) after ACT (P = 0.6). The relatively low cumulative incidence of relapse in SR aGVHD may be associated with the possible presence of graft versus tumor effect in this cohort of patients. Conclusion: ECP and ACT demonstrate comparable rates of response to treatment and are associated with low cumulative incidence of relapse. Patients after ECP show statistically significantly better OS than after ACT but as the study is limited by the heterogeneity of analyzed groups, further randomized studies are needed. Disclosure of Interest: None declared. Introduction: Extracorporeal photopheresis is a cell-based immunomodulatory therapy that permits the direct targeting of psoralen-mediated photochemotherapy to circulating pathogenic T cells. Limited data are currently available regarding overall response rates and durability of responses among children with chronic graft-versus-host disease (cGVHD), which constitutes a major challenge of successful hematopoietic cell transplantation. Material (or patients) and methods: We conducted a retrospective analysis of children after allogeneic hematopoietic stem cell transplantation (allo-HSCT) treated with extracorporeal photopheresis (ECP) during a 7-year period at a single university hospital. All patients received ECP on two consecutive days every 2-4 weeks (one procedure). ECP was given to 32 patients, median age 11 (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) , with steroid-refractory chronic GVHD (cGVHD severity based on NIH 2014 criteria, mild (n = 2, 7%), moderate (n = 13, 40%) and severe (n = 17, 53%). Cutaneous cGVHD was found in 26 patients (81%) and was accompanied by visceral (hepatic, gastrointestinal) cGVHD (n = 7, 22%), mucous involvement (n = 10, 31%, lung involvement (n = 7, 22%). All had failed conventional immunosuppressive therapy or could not be adequately controlled only by corticosteroid treatment. The primary efficacy end points were response rate to ECP and overall survival. Results: Overall response rate was 72% (n = 23). The complete response (CR) rate was 9% (n = 3) and the partial response rate was 63% (n = 20). Nine patients (28%) had no response (NR). The median duration of ECP was 32 months (5-91) and the median number of ECP procedures in each patient was 7 . OS was 77%. 7-year OS in patients with PR/CR was 84% and 7-year OS in non-responders was 38% (P = 0,001). Responses, both complete and partial, were obtained in 1 (50%), 11 (85%), and 11 (65%) of patients with mild, moderate and severe cGVHD, respectively. Responses, both complete and partial, in the mucus lining, skin, lungs and visceral organ (hepatic, gastrointestinal) were 80, 73, 57, and 43%, respectively. Conclusion: We infer that ECP is an effective treatment of steroid-refractory cGVHD in children, especially with cutaneous and mucous involvement of cGVHD. Those who respond to ECP have better outcome and favorable prognosis. Disclosure of Interest: None declared. Introduction: Chronic graft-versus-host disease (cGVHD) is a leading cause of morbidity and mortality after allogeneic hematopoietic stem cell transplant (SCT). Duration of systemic immunosuppression (IS) for cGVHD is about 2 years; however, 15% of patients require IS 7 years after SCT. Previous studies identified factors associated with longer duration of IS. Our aim was to identify clinical and immunologic factors associated with cGVHD that is persistent ≥ 7 years after cGVHD diagnosis (pcGVHD). Material (or patients) and methods: Patients were drawn from a cross sectional study of the natural history of cGVHD at the National Institutes of Health. A cohort of patients who enrolled ≥ 7years from time of cGVHD diagnosis was compared to those who enrolled o1 year from diagnosis. Further analysis was performed to evaluate cytokine profiles and candidate cGVHD biomarkers among patients with pcGVHD, control patients, and healthy volunteers as a second control group. The IFN-g, IL-6, IP-10, and MCP-1 high sensitive assays were parts of V-PLEX from Meso-Scale Discovery (MSD). The BAFF, CXCL9, and ST2 high sensitive assays were developed and customized for clinical testing using MSD electrochemiluminescence immunoassay technology with antibody pairs obtained from R&D Systems. Univariate analysis was performed, and significant factors were used in multivariable logistic regression. All p-valueso0.005 were considered statistically significant; 0.005 ≤ P40.05 represented strong trends. Results: There were 38 patients with pcGVHD and 83 control patients o1 year from cGVHD diagnosis; 62% of pcGVHD patients were still on systemic IS. Factors that were significantly associated with pcGVHD included: bone marrow (BM) stem cell source (result divergent from previous studies), myeloablative conditioning, greater lung involvement and lower FEV1, more eye and mouth symptom burden, lower ferritin, lower IS, +ENA autoantibody, higher platelets, CD19 cells, CD4 cells, and immunoglobulin levels. Interestingly, there were no differences in factors associated with cGVHD disease severity, distance from home to the transplant center, nor factors identified in previous studies. In a multivariable analysis that included 11 variables, BM stem cells, +ENA, higher NIH lung score, higher platelets, and higher IgA levels were independent predictors of pcGVHD. There were 26 pcGVHD patients, 52 controls, and 8 healthy volunteers for whom cytokine analysis was performed. Significant differences were seen between pcGVHD and control patients in levels of IP-10, MCP-1, CXCL9, and BAFF ( Figure 1 ). Conclusion: Although cGVHD is often self-limited, late forms requiring long duration of IS exist. Our analysis of pcGVHD patients showed more lung involvement, less inflammatory signs, and higher B cells, immunoglobulins, and autoantibodies. In addition, plasma analysis revealed significant differences in a subset of cytokines and candidate cGVHD biomarkers. However, standardly accepted measures of disease severity were not associated with pcGVHD, suggesting that further tools are needed to differentiate accumulated damage from active disease. These findings suggest that pcGVHD may reflect irreversible damage rather than an active immune process. Our results contribute to hypotheses in the pathogenesis and may have clinical implications relative to IS intensity management in patients with pcGVHD. Disclosure of Interest: None declared. Material (or patients) and methods: We enrolled 10 paediatric patients (median age 10 years, range 4-15) that underwent allogenic HSCT for malignant hematopoietic disorders. The donor was a sibling for 4 patients, a matched unrelated donor for 5 patients and a haploidentical parent for 1 patient. 11 samples were taken at different time points (median day +400, range +90 -+1682) and clinical status at the time of the sample was recorded and classified as active cGvHD, inactive cGvHD, active aGvHD or no GvHD. After cell separation by Ficoll-Hypaque, we analysed by flow cytometry the intracellular expression of IFN γ, IL17, IL4 and IL22 on CD4-positive T-cells. Th1 cells were defined as CD4 +/cyIFNγ+ cells, Th2 as CD4+/cyIL-4+, Th17 as CD4+/cyIL-17+ and Th22 CD4+/cyIL-22+/cyIFNγ neg/cyIL-4 neg/cyIL-17 neg cells. Statistical comparison was performed with the non parametric Mann-Whitney U test for independent populations. Results: The proportion of T-helper cells expressing any intracellular cytokines were small (median 663/10 6 lymphocytes, range 450-2410) but detectable. Significantly higher numbers were found in patients with active cGvHD (median 2204 cells/10 6 lymphocytes) or active aGvHD (median 2520 cells/10 6 ) than in patients with inactive GvHD (median 582 cells/10 6 lymphocytes) or no GvHD (median 582 cells/10 6 lymphocytes) (P = 0,04). As compared with patients without GvHD or with resolved GvHD, patients with active GvHD showed an expansion of Th1 (median = 451 cells/10 6 lymphocytes versus 66 cells/10 6 lymphocytes, P = 0,04) and Th17 (median = 560 cells/10 6 lymphocytes versus 160 cells/10 6 lymphocytes, P = 0,04). No significant difference was observed in Th2 or Th22 populations. The expression of cytoplasmic cytokines in T-helper lymphocytes of patients affected by GvHD was analysed. Results show a higher number of T-helper cells expressing intracellular cytokines in patients with active GvHD, mainly due to the expansion of Th1 and Th17 cells. Differently from other previous reports, we did not stimulate the cells before the analysis. Although the population analysis is more difficult due to the rarity of cytokine expressing cells, this method is feasible and the results obtained are probably more adherent to the biological process. Moreover, without the prior in vitro stimulation, the absolute number of cytokine expressing T-helper cells can be determined. The evaluation of the absolute number of cytokine expressing T-helper cells could be a promising marker of GvHD activity and needs to be validated in a higher number of patients. Cells producing IFNγ and IL17 seem to be the major contributor to the expansion of activated cells in active GvHD. Disclosure of Interest: None declared. Laboratory Medicine, Therapeutic Immunology, 2 Dept. of Laboratory Medicine, 3 Dept. of Dental Medicine, Karolinska Institutet, 4 Center for Allogeneic Stem Cell Transplantation, Karolinska University Hospital, 5 Dept. of Obstetrics and Gynecology, Karolinska Institutet, Stockholm, Sweden Introduction: Placenta-derived decidual stromal cells (DSCs) are a novel therapy for acute graft-versus-host-disease (GVHD) and hemorrhagic cystitis (HC) after allogeneic hematopoietic stem cell transplantation (HSCT). Material (or patients) and methods: In this retrospective study, we assessed the adverse events of this treatment. The cohort study included 44 treated patients and 40 controls. Age was median 49 (range 1-68) years and 48 (4-63) years in the two groups, respectively. The DSCs given were in passage 2 (n = 21), 3 (n = 31), or 4 (n = 30). The viability of infused DSCs was median 93 (69-99)%. The dose given i.v. was median 1.5 (range 0.9-2.9) x 10 6 DSCs/kg. The patients were given median 2 (range 1-5) doses of DSCs with a total of 82 infusions. Monitoring started 3 days before DSC infusion and ended 3 months after the last DSC infusion. Results: The frequency of adverse events in the two groups was compared using two-sided t-tests and the frequencies of thromboembolism, stroke, and infusion-related adverse reactions were compared using Fisher's exact test. Three patients had transient reactions during DSC infusion. The patients suffering from acute GVHD treated with DSCs had higher levels of albumin than the controls (P = 0.03). Cytomegalovirus reactivation was less common in the GVHD DSC group than in the controls (P = 0.0003). The DSC-treated group had the same frequency of other adverse events (invasive fungal infection, leukemic relapse, and death) as the controls. The DSC group had a lower incidence of non-invasive fungal infections than the controls (P = 0.039). Introduction: Imatinib is effective in steroid-refractory GVHD (SR-cGVHD); however doses 4200 mg/day are associated with relevant toxicity.Nilotinib (NIL) has higher affinity than Imatinib to intracellular tyrosine kinases (IC50 = 20 nM for p210BCR-ABL, IC50 = 71 nM for PDGFR) and has a more favourable tolerability profile. Material (or patients) and methods: We conducted a phase I-II study (standard 3+3 design) aimed to define MTD and activity of NIL in SR-cGVHD (NCT01810718). A secondary endpoint was pharmacokinetic (PK) evaluation of NIL, to be related with efficacy and toxicity measures. Clinical outcomes (OS, EFS and PFS) according to response rate (ORR) have been also evaluated, according to the old and new NIH response criteria and to the clinical improvement (CI) specifically defined for this protocol. Results: Twenty-one patients were enrolled (Tab.1). After dose-escalation up to 600 mg/day, the MTD was not reached. Six patients received NIL at 200 mg/d, 6 at 300 mg/d, 6 at 400/mg/d and 3 at 600 mg/d (300 mg BID). Extrahematological AE with a frequency 420% were: asthenia, headache, nausea, pruritus, cramps, constipation. The most frequent hematological AE was anemia (grade 3 in only 1/21 pts); the other hematological AE were mild. During the first 6 months of NIL treatment, there were only 4 grade 3 AE and none grade 4 AE.PK evaluation was planned 15 days and 1, 3, 6 months after start of NIL. Samples were drawn at least 8hrs after last NIL administration; data were available for 17 pts. Mean and median plasma concentrations of NIL (C-NIL) were 817 (SD ± 450) and 773 ng/ml in all patients; C-NIL were significantly different by NIL dosing (P = 0.016, Anova). We also checked for associations between lower C-NIL and lack of response (according to NIH criteria) but we did not find significant results; or between higher C-NIL and more frequent or severe AE, but again we did not find significant correlations. 18/21 patients received NIL for ≥ 30 days and were evaluable for ORR and clinical outcome at 12 months; we did not find discordances between the new and old NIH response criteria evaluation. We observed 5 PR;9 unchanged;2 progressions and 3 failures (2 SAE, and 1 addition of new Immunosuppressive treatment:ISTX), while ORR evaluated according to the definition of "objective improvement" (OI) was: 9 OI;3 unchanged;3 progressions and 4 failures(2SAE;1 new ISTX;1 relapse of the underlying disease). With a median follow-up of 25 months 15 patients are alive (11 are still on NIL), with an OS of 78% and the Event Free Survival (EFS) was 51% (72% and 44% on the ITT basis respectively); PFS was 56% ( Fig. 1 ). NIL was well tolerated in this cohort of SR-cGVHD patients, even at higher dosages. The favorable safety profile of NIL makes this drug an attractive option in SR-cGVHD patients unable to tolerate Imatinib. Disclosure of Interest: None declared. Comparison of cyclosporine/ methotrexate and cyclosporine/ mycophenolate combinations for prevention of acute graft-versus host disease after allogeneic stem cell transplantation from unrelated donors: A single center retrospective study R. Yerushalmi 1 , I. Danylesko 1 , N. Shem-Tov 1 , A. Nagler 1 , A. Shimoni 1,* 1 Division of Hematology and Bone Marrow Transplantation, Chaim Sheba Medical Center, Tel-Hashomer, Israel Introduction: Acute graft-versus-host disease (GVHD) is the major treatment-related complication after unrelated donor stem cell transplantation (uSCT). Several GVHD prevention regimens have been explored, but no regimen has shown superiority. In this analysis we compared outcomes following two of the most frequently used regimens, cyclosporine A and a short course of methotrexate (MTX group) and cyclosporine A and mycophenolate (cellcept, CC group). Material (or patients) and methods: The study included 472 consecutive patients (pts) given uSCT in a single institution. We retrospectively analyzed pt charts for SCT outcomes. Results: The median age was 55 years (range, 18-76). The conditioning regimen was myeloablative (MAC, 19%), reduced-intensity (RIC, 39%) and reduced-toxicity (41%). All pts were given ATG (Neovii, 15 mg/kg) during conditioning. 71% had 10/10 and 29% had ≤ 9/10 HLA matching. In all, 314 pts were given MTX and 158 CC. The CC group included older pts (median age 57 Vs 52 years, respectively, P = 0.002). A lower percentage had MAC (9% Vs 24%, P = 0.005). More pts had lymphoproliferative disease and less acute leukemia (P = 0.02). More pts had advanced disease at SCT (65% Vs 39%, P = 0.001) and more had a high comorbidity score (18% Vs 11%, P = 0.05). Engraftment was faster after CC, day 11 and 14, respectively (P = 0.001). Acute GVHD grade II-IV occurred in 47% (95% CI, 40-56) and 27% (22-33, P40.0001) and grade III-IV in 28% (95% CI, 21-37) and 12% (9-17, P40.0001), respectively. Chronic GVHD was 30% (26-34) after both regimens. With a median follow-up of 52 months (5-151), the 5-year non relapse mortality (NRM) was 44% (37-53) and 24% (20-30), ( p40.0001) and overall survival (OS) was 29% (22-36) and 40% (22-36), after CC and MTX, respectively (P = 0.0006). However, the CC group included pts with higher risk for acute GVHD and NRM. To minimize this potential bias we analyzed these outcomes on the basis of an intention to treat like analysis. During the years 2008-9, the leading GVHD prevention regimen for uSCT in our institution included CC ( Introduction: Hematopoietic stem cell transplant recipients appear to be one of the most vulnerable populations for the development of Clostridium difficile Infection (CDI) infection. There are significant changes in the gastrointestinal tract when a patient has undergone allogeneic hematopoietic stem cell transplantation. These changes occur for several reasons including the immunosuppression patient receives, gastrointestinal (GI) mucosal damage from conditioning chemotherapy and change in the GI microbiome. Due to these changes, it was felt that the risk of C. difficile was much more likely in those patients who developed acute graft versus host disease (aGVHD). Material (or patients) and methods: Aim: 1. Study the correlation between the incidence of aGVHD and those who developed CDI. 2. To evaluate the incidence of those who developed CDI with or without developing aGVHD. After due IRB approval, all patients who underwent allogeneic hematopoietic stem cell transplant, with either myeloablative or non myeloablative conditioning regimen and those who developed aGVHD were identified. Patients were grouped in 4 groups-those with aGVHD of GI tract and CDI, those with any type of aGVHD and CDI, those without CDI but had aGVHD and those with no aGVHD but had CDI. Statistical analysis was done using Pearson's chi square analysis for categorical variables and One way analysis of variable (Anova) Summary fit to analyze occurrence rates. Results: 424 patients were evaluable for final analysis-214 myeloablative and 196 non-myeloblative. Of these 317 patients had aGVHD with no CDI and 51 patients had aGVHD of GI tract with CDI. CDI occurred in 19 patients without aGVHD of the GI tract. CDI recurrence rates could not be addressed due to incomplete data. There was no significant difference between the myeloablative and non myeloablative groups in terms of developing CDI. 25% of allogeneic stem cell recipients developed CDI without underlying aGVHD. Introduction: PTCy in combination with calcineurin inhibitors (CNI) plus mycophenolate mofetil (MMF) after haploidentical allogeneic hematopoietic cell transplantation (alloSCT) has been shown to be effective prophylaxis against GVHD. In myeloablative (MA) matched related and unrelated bone marrow donor settings, single-agent PTCy has produced encouraging results. However, some reports have suggested that PTCy could not be adequate as sole GVHD prophylaxis in the context of peripheral blood stem cells (PBSC) and/or reduced-intensity conditioning (RIC) alloSCT. The objective of this study was to investigate the outcome of unrelated peripheral blood alloSCT using PT-CY-based GVHD prophylaxis. Material (or patients) and methods: We retrospectively analyzed data from 23 consecutive adult patients who received a PBSC unrelated alloSCT using PTCy-based GVHD prophylaxis from May/2014 and Oct/2014. Ten patients received an 8/8 HLA-matched alloSCT, and 13 a 7/8 HLAmismatched transplant. Results: Patients were transplanted for AML/ALL (70%), MDS (13%), MPD (13%), and LPD (4%). Disease status was 74% CR, 4% PR and 22% active disease. Fludarabine-based conditioning regimens were used in all patients: 17 (74%)MAC and 6 (26%) RIC. GVHD prophylaxis consisted on PTCy 50 mg/kg i.v. on days 3 and 4 after transplant followed by tacrolimus and MMF (n = 15, 65%) or CNI alone (n = 8, 35%). All patients engrafted, with a median time to neutrophil count of 0.5 of 16 (12-23) days, and median time to platelet count 20 x 10 9 /L of 19 (11-33) days. Two patients, both in the MAC HLA-matched group, presented a secondary graft failure. One patient developed VOD and 5 cases of hemorrhagic cystitis were observed (4 mild, 1 moderate). Incidence of severe infectious complications was low: pulmonary aspergillosis, n = 1; bacterial infection, n = 2; and toxoplasmosis, n = 1. Thirteen (59%) patients presented CMV reactivation; no EBV reactivation was observed. Five (22%) patients died of TRM, all of them during the first 100 days after transplant: VOD (n = 1), hepatic toxicity and multiorgan failure (n = 1), toxoplasmosis (n = 1), toxic leukoencephalopathy (n = 1), and cerebral hemorrhage (n = 1). Cumulative incidence of acute grade II-IV and grade III-IV GVHD were 29% and 15%, respectively, at 1 year. Cumulative incidence of chronic GVHD was 15% at 1 year (2 mild, and 1 moderate). Median follow-up after alloSCT was 247 days (9-618 Introduction: Extracorporeal photopheresis (ECP) is a second line therapeutic intervention for steroid refractory, dependant or intolerant chronic GVHD (cGVHD). The serum proteins Elafin, Reg3α and ST2 have previously been found to predict GVHD grade and transplant related mortality in acute GVHD [1] [2] [3] . The aim of this study is to investigate, for the first time, the relationship between these biomarkers and cGVHD in patients referred for ECP, with the future aim to evaluate the use of these factors for predicting and monitoring response to ECP. Material (or patients) and methods: Serum from 20 patients taken immediately before the start of ECP treatment was analysed. The patients had a range of sites affected by GVHD (skin = 16, gut = 4, mouth = 4, liver = 1, joints = 2, eyes = 1) and had varied lengths of time between transplant or subsequent DLI and ECP start (median 257d, 85-1812d). Healthy controls (HC) were also analysed (n = 7). Serum Elafin, ST2 and Reg3α levels were measured by ELISA, and vitamin D determined by competitive immunoassay. The skin score of each patient was determined on the date of referral (up to 2 weeks before ECP start). The scores used were Modified Rodnan's skin score and the NIH skin score. Statistical analysis was performed using Prism 6. Tests performed were Mann-Whitney and Pearson correlation coefficient. As well as differences between patients and controls, relationships were also found between the biomarkers and disease severity. Elafin levels correlated with the NIH skin score (P = 0.0012, r = 0.7, CI 0.35-0.88) in keeping with previous studies showing the association of Elafin with skin GVHD. No relationship was found between either skin score and Reg3α, which has previously shown an association with gut GVHD. Although most patients had skin involvement, only 4 had signs of gut GVHD and so a relationship with gut score is yet to be established. Association was also seen with the soluble decoy receptor for IL-33, ST2, and NIH skin scores (P = 0.019, r = 0.52, CI 0.10-0.78). ST2 can be induced by the important immuno-modulator vitamin D 4 but no relationships between vitamin D and the other biomarkers examined were found. However, 85% patients were found to be deficient in vitamin D ( o50nmol/l). Conclusion: This pilot study in patients with cGVHD referred for ECP agrees with previous studies in aGVHD showing raised levels of the biomarkers Elafin, Reg3α and ST2 in patients compared to controls. It also shows a correlation between Elafin and ST2 and the level of skin involvement. No relationship however was found between skin scores and Reg3α. We aim to confirm these results further with an increased cohort size and also to follow these patients to determine how ECP affects the levels of these markers and how this relates to GVHD response and the ability of the patients to come off steroids. Introduction: Extracorporeal photopheresis (ECP) is an effective cell-based therapy for the treatment of acute and chronic graft versus host disease (aGVHD, cGVHD). In the"offline" system, the cell dose is critical in terms of mononuclear cells(MNC). Our primary objective was to analyze retrospectively the mean cell dose of MNC and lymphocyte subpopulations (LS) infused per procedure. The second objective was to investigate the impact of MNC dose on ECP secondary response. Material (or patients) and methods: From November 2009 to November 2015 we performed 580 ECP procedures in 33 patients(pts).23 patients with aGVHD were registred, median age was 41 years(30-52) and weight 62Kg (50-80).6 pts were diagnosed of AML,2 ALL,5 MDS and10 other high risk hematological neoplasias (NHL, myelofibrosis(MF) or MM). ECP was the 2 nd line treatment for aGVHD in 30%, 3 rd line in 56% and 4 th line in 14% of patients. We analyzed 10 pts with cGVHD. Median age was 45 years(37-56) and weight 55Kg(45-63).3 pts were diagnosed of AML,2ALL,1MDS and the rest other pathologies.ECP was the 2 nd line treatment for cGVHD in 40%,3 rd line in 40% and 4 th line in 20% of patients. "Offline"system ECP were performed with a cell separator(Spectra Optias) for the CMN collection; after 8-methoxypsoralen was added, UVAMATIC-Macopharmas was used for product ultraviolet irradiation. ECP procedures were performed for two consecutive days, initially weekly (aGVHD), or every two weeks (cGVHD) and afterwards monthly according to clinical response. Before reinfusion, the cellularity was determined in terms of TMNC/kg and LS (CD3,CD4,CD8, CD19 and NK) using the Unicell DHX800 automatic counter Beckman Coulters and 5 colour multiparameter flow cytometry FC500 and Navios Beckman Coulters respectively. The association between the response and cellularity dose infused was measured using Pearson's lineal correlation coefficient and Spearman's non-parametric correlation coefficient. Statistical analysis was performed using version 21.0 of the SPSS program. Results: The median of procedures of ECP was12 (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) in pts with aGVHD and 21(11-33) in cGVHD. The median duration was 55 days (26-125)in aGVHD and 129(69-200) in pts with cGVHD. Cell dose infused in each process in patients with aGvHD and cGvHD is detailed in Table 1 . Cellularity in terms of MNC and LS was higher in cGVHD than in aGvHD. CD3+CD8+ LS was mostly infused in both groups. For responding pts there were no stadistical differences in MNC and LS infused. The relationship between infused cellularity and response is detailed in table 1. Conclusion: In our study, the total MNC infused in all pts exceeded the minimum reported as effective (4 100 x10 6 /kg per cycle). Cellularity infused in terms of MNC and LS was higher in cGVHD than in aGVHD. In responding pts, aGVHD and cGvHD, MNC and LS dose infused was higher than in nonresponders (no statistical significance). More studies are warranted in order to rule out the potential influence of other factors. Introduction: ECP is an immunomodulatory therapy that has proven effective in the treatment of acute and chronic refractary GVHD. The aim of this study was to evaluate the clinical effect ofECP, analyze the impact in reducing steroid dose, safety and tolerance of the procedure. Material (or patients) and methods: All patients with refractory acute and chronic GVHD were retrospectively studied, from November 2009 to November 2015. We have performed 587 ECP procedure with the "off line" system to 33 patients (aGVHD: 23, cGVHD: 10). Diagnosis of aGVHD and cGVHD and assessment of organ involvement were performed according to the criteria of Glucksberg and the NIH consensus criteria, respectivily. The clinical characteristics of the patients are shown in Table 1 . Apheresis procedures were perfomed with COBE Spectra system (Terumo BCTs, Lakewood, CO, USA; version 7.0) by processing 1.5-2 times the patient blood volume. The product was transferred to a UVA-permeable bag (UVA, Macopharma, France), added 5 mL (0.1 mg) of 8-methoxypsoralen (8-MOP) saline solution (S.A.L.F.s, Cenate Sotto, Italy), exposed to UVA irradiation (Macogenic G2, Macopharmas), and then reinfused. ECP procedures were performed for two consecutive days, initially weekly (aGVHD), or every two weeks (cGVHD) and afterwards monthly according to clinical response. Response was evaluated by clinical assessment, and reduction of steroid dose at the end of treatment was recorded. Descriptive data were summarized using median values, ranges and percentages for continuous variables. Overall survival (OS) since start of ECP treatment was analyzed using the Kaplan-Meier method. Results: In the overall analysis, in aGVHD group, there was a 48% complete response, 17% partial response and 34% non response. In cGVHD group there was a 30% complete response, 50% partial responses and 20% of non response. Muco-cutaneous involvement showed the best response rates, 76% in aCVHD and 88% in cGVHD, 66% in gastrointestinal involment, 50% in liver involment. The response rate in lung involment in cGVHD was 75%.The reduction in steroid dose was 100% in the responders. he tolerance to the procedure was excellent and the unique complications seen were related with two central venous catheter infections.At last follow-up the overall survival in the aGVHD and cGVHD group was 54% (confidence interval: 26-74%) and 52% (confidence interval:16-79%) at 9 months.Patients with aGVHD and cGVHD achieving a complete or partial response showed a significant improved OS after ECP (P = 0,0158 y 0,0049) respectively. Conclusion: In our experience, the ECP was associated with excellent tolerance, significant response rates and successful reduction of steroid dose in patients with refractory GVHD. Lung cGVHD 75% partial response rate, is remarkable because of the severity of this disease. Responders has survival advantage The small sample size of our study did not allow us to draw firm conclusions. Disclosure of Interest: None declared. [P120] P122 Different clusters of immunologic variables are associated with chronic GVHD and relapse in a dynamical model these immunological reactions. Changes in pre-and posttransplant B-lymphopoietic microenvironment and imbalances in the subsets of B-cells may influence GVHD and GVT. The atypical association of regulatory T-cells with GVHD may be explained by the relative efficiency of different subsets of regulatory T-cells (naïve4memory), as shown in some experimental models. The correlation of CD8+TEMRA at +90 with chronic GVHD may early indicate a persistently activated and dysregulated immune system. The validation of these clusters of immunological parameters as specific early predictors of GVHD or GVT, even before SCT, could potentially allow the development of pre-emptive and targeted therapies. Disclosure of Interest: None declared. Parameters of protein metabolism and thyroid function as predictors in a scoring system for acute and chronic Graftversus-host disease C. Skert 1,* , A. Introduction: Some "classical" patient-, donor-and transplant characteristics, such as age, gender disparity, donor type, HLAmatch, and source of stem cells, have been reported as predictors for acute and chronic GVHD. However, no studies analysed these "classical" variables together with parameters of metabolic and endocrine functions that may potentially influence the immune system. Thus, patient-and transplant variables together with index of liver and thyroid function, and some parameters of protein and lipid metabolism were retrospectively evaluated at different time points after transplantation, in order to identify possible predictors of acute and chronic GVHD and to calculate a risk score. Material (or patients) and methods: Clinical and transplant characteristics, number and type of infections before and after SCT were analysed in 194 patients. The following variables were also analysed pre-SCT, at day +7,+14,+21,+28, at+3 and +6 months: LDH, parameters of liver function; parameters of protein and lipid metabolism; thyroid function tests; autoimmune parameters; body mass index. A 2-step multivariate analysis was performed using principal component analysis and Cox regression analysis. Based on the regression coefficient of Cox analysis for each significant predictor, a scoring system for acute and chronic GVHD was calculated. Results: In multivariate analysis, diagnosis of Myelodisplastic Syndrome or Chronic Myeloid Leukemia (P = 0.0004), conditioning regimen including Total Body Irradiation (P = 0.0003), and pre transplantation urea434 mg/dl with +21 day urea454 mg/dl (P = 0.0008) were predictors for acute GVHD. Score values for each factor are 2, 1, 1, respectively and the system had a score range between 0 and 4. The probabilities of acute GVHD according to the sum scores ranged from 8% (score 0) to 98% (score 4). Female donor (P = 0.0008), pre-SCT TSH values ≥ 2 mU/L with +28 day urea ≥ 39 mg/dl (P = 0.02), +6 month total proteino 5,5 g/dl with gamma-GT ≥ 347 U/L (P = 0.0001) resulted predictors for moderate/severe chronic GVHD. Risk of chronic GVHD at +6,5 month ranged from 3% (score 0) to 97% (score 4). Conclusion: Our study evidenced that factors other than those "classical" may be associated to GVHD. The scoring system included routine-parameters, which are easily available in clinical practice. Urea levels depend on the balance between protein intake, endogenous catabolism and urinary excretion. The inflammatory microenvironment of GVHD promotes muscle catabolism and hence, increased urea levels. Increased urea levels could be indirect index of increased uremic toxins as well, which may stimulate the production of pro-inflammatory cytokines and the activation of leukocytes. Increased urea levels and uremic toxins could also derive from a dysregulated metabolism of the gut microbiome that may influence immune system. Our findings suggest the usefulness to study in deep the complex network between metabolic/ endocrine functions and immune system for a holistic approach of the transplant management. Introduction: There are no approved therapies for cGVHD, a serious complication of allogeneic stem cell transplantation (SCT). Both B-and T-cells play a role in the pathophysiology of cGVHD. Ibrutinib (ibr), a first-in-class inhibitor of Bruton's tyrosine kinase (BTK), was first approved in 2013. In mouse models, ibr reduces the severity of cGVHD mediating its effects via BTK and IL-2-inducible T-cell kinase (ITK) (Dubovsky JCI 2014). Ibr may also improve donor chimerism and/or graft-vsleukemia effects while reducing cGVHD in post-SCT patients (pts) (Ryan ASH 2014, Ryan BMT 2015). This ongoing phase 1b/ 2 trial evaluated the safety and efficacy of ibr in pts with active steroid-dependent/refractory cGVHD who need additional therapy. Material (or patients) and methods: Eligible pts had ≤ 3 prior regimens for cGVHD and either 425% BSA erythema or an NIH mouth score 44. Pts received daily ibr until cGVHD progression or unacceptable toxicity. The primary endpoint was cGVHD response per 2005 NIH consensus response criteria. The effect of ibr on B-and T-cell signaling pathways was evaluated. Results: No dose-limiting toxicities were observed in phase 1b (n = 6); the recommended phase 2 dose was 420 mg. At data cut-off, 28 pts who underwent SCT were enrolled. Median age was 57 yrs (range, 27-70). Median duration of cGVHD prior to study entry was 15.8 mo (range, 1.4-63.2). Pts had received a median of 2 prior regimens. Six pts have not reached their first response assessment. Of 22 pts who had a response assessment or discontinued early, 12 responded to treatment (ORR, 55%; 1 CR, 11 PRs). Three of 3 pts who were treated for 49 weeks and 3/3 additional patients who were treated for 23 weeks maintained response. Eleven of the12 responders had reduction (n = 10) or discontinuation (n = 1) of corticosteroids. Serum BAFF levels decreased over time supporting an effect of ibr on B-cell-driven cGVHD. A decrease in CD4 T cell pPLCγ1-Tyr783 levels, a marker of ITK, indicated a decrease in CD4 T-cell activation in examined pts. The most common AEs were fatigue (50%), bruising (25%), diarrhea (25%), and nausea (21%). Grade ≥ 3 AEs occurring in 41 pt were fatigue (in 5 pts) and diarrhea, pneumonia, and headache (each in 2 pts). Six pts discontinued therapy for AEs (fatigue [n = 2] and atrial fibrillation, brain abscess, oral pain, and tongue ulceration, each in 1 patient) and 3 discontinued for progressive cGVHD. Conclusion: With an ORR of 55%, ibr shows promising efficacy with initial signs of durability for the treatment of steroiddependent/refractory cGVHD pts. Biomarker changes support an ibrutinib effect in both B and T cells in cGVHD. Reported AEs were consistent with those for ibr in B-cell malignancies. Efficacy in this pretreated, high-risk population supports further study of ibr in the treatment of cGVHD. Introduction: The use of antithymocyte globulins (ATG) as part of the conditioning regimen for acute leukemia or myelodysplastic syndrome (MDS) is still controversial in the setting of matched or mismatched donor allogeneic hematopoietic stem cell transplantation (alloHSCT). We have opinions that the dose of ATG is critical issue and could be a major determinant of outcome after alloHSCT. Here, we investigated the impact of ATG on cumulative incidences of graft versus host disease (GVHD), relapse free survival rates (RFS) and overall survival rates (OS). This study included patients who diagnosed acute leukemia or MDS received induction chemotherapy followed by alloHSCT. They underwent conditioning regimen based on fludarabine (30 mg/m 2 daily, for 5days) or cyclophosphamide (50 mg/m 2 daily for 2days), intravenous busulfan (3.2 mg/kg daily for 2-4days) and r-ATG (total dose of 2.5 -9 mg/kg, for 1-3days) prior to alloHSCT. All patients received from peripheral blood stem cell as graft source and they received calcinurin inhibitor as GVHD prophylaxis started at day-1. Results: Data from patients at four university hospitals in South Korea between January 1999 and February 2014 were collected retrospectively. We analyzed 314 consecutive patients with a median age of 39 years (range, 14-68) and the male to female ratio was 1.21:1.0. The 59.6% patients were received myeloablative conditioning regimen and 40.4% received reduced intensity of conditioning. The 33.4% patients were received ATG and 66.6% patients were not. The 53.2% patients were received stem cell from identical sibling donor, 13.1% were from mismatched related or unrelated, 33.7% were from matched unrelated donor. The cumulative incidences of grade II-IV acute GVHD were higher in patients received less than 2 days of ATG than 2 days or more than that days of ATG (48.0% and 34.4%, P = 0.022) and the those of more than extensive chronic GVHD were not different (28.7% and 28.4%, P = 0.887) according to receiving days of ATG, respectively. The cumulative incidences of grade II-IV acute GVHD and the those of more than extensive chronic GVHD were shown trend high in patients received ATG ofo5 mg/kg, ≥ 5 mg/kg and o7.5 mg/kg and ≥ 7.5 mg/kg (47.1%, 36.7% and 34.4%, P = 0.131 and 39.5%, 28.1%, and 25.4%, P = 0.151). There was no difference in the 5-yesr RFS and OS in patients with and without ATG (61.6% vs. 57.0%, P = 0.263 and 50.3% vs. 43.9%, P = 0.228). However, The 5-yesr RFS and OS were shown significant difference between ATG used and unused in patients received stem cell from mismatched donor (48.8% vs. 24.2%, P = 0.048 and 30.6% vs. 11.4%, P = 0.023). Conclusion: The longer use of ATG (≥ 2 days) or higher doses of ATG ( ≥ 5 mg/kg) reduce cumulative incidences of acute or chronic GVHD. Especially, the 5-yesr RFS and OS of patients who received ATG were shown significant higher than those of not received in patients received stem cell from mismatched donor. Further comparative prospective studies could be helpful to better identify the optimal ATG dose in different setting of allogenic stem cell transplantation. Disclosure of Interest: None declared. Introduction: Effective treatments are lacking for the treatment of steroid-refractory graft versus host disease (GVHD), a major cause of morbidity and mortality following allogeneic hematopoietic cell transplantation (HCT). Mesenchymal stromal cells (MSCs) have demonstrated promise but there is uncertainty regarding their clinical effectiveness. Material (or patients) and methods: A systematic scoping review of the literature was performed to characterize the heterogeneity of published studies using MSCs to treat and/or prevent GVHD and to identify opportunities for standardization of future studies. Actively recruiting registered clinical trials were also identified to provide insight regarding the extent to which heterogeneity has been addressed. Results: Thirty studies were identified, including 19 studies (507 patients) addressing the treatment of acute or chronic GVHD and 11 prevention studies (277 patients). Significant heterogeneity was observed in the age and diagnoses of study subjects, the intensity and specifics of the conditioning regimens, degree of HLA-matching and source of hematopoietic cells. MSCs were derived from bone marrow (83% of studies), cord blood (13%), or adipose tissue (3%) and were cryopreserved from third party allogeneic donors in virtually all studies. Culture conditions and media supplements were highly variable and characterization of MSCs did not conform to ISCT criteria in any study. MSCs were used at passage 1-7 of cell culture and the median dosage of MSCs ranged from 1-10x10 6 /kg using non-uniform schedules of administration. Treatment response criteria were not standardized and effectiveness in controlled treatment studies (5 studies) could not be demonstrated convincingly. Details of actively recruiting registered trials suggest heterogeneity will persist with only 53% of trials describing the use of standard GVHD response criteria and few detailing methods of MSC manufacturing. Conclusion: Future studies will need to make substantial coordinated efforts to reduce study heterogeneity and clarify the role of MSCs in GVHD. Disclosure of Interest: None declared. Results: Age, diagnosis, stem cell source and conditioning regimen intensity were well balanced between both groups (Table 1) . For the ATG group, conditioning regimen consisted of fludarabine combined with busulfan (80%) or melphalan (12%), and TBI plus cyclophosphamide (8%). All patients received ATG 2 mg/kg from day -4 to day -2 followed by methotrexate and cyclosporine (CsA). Conditioning regimen for all patients from the PTCy group included fludarabine and busulfan, with addition of treosulfan in 20%. GvHD prophylaxis consisted of PTCy 50 mg/kg on days +3 and +4, combined with: CsA plus MMF in 50%, CsA alone in 40%, or tacrolimus plus MMF or sirolimus in 10%. All patients showed neutrophil engraftment in a median time of 15 days for the ATG group and 18 days for the PTCy group (P = 0.07). Platelet engraftment rates were similar between both groups: 90% vs 92% (P = 0.08), however slower in the PTCy group (15 vs 26 days). Three patients in the PTCy group presented secondary graft failure (15%). Only one patient developed veno-occlusive disease in the PTCy group. However, hepatic toxicity grades I-IV was more frequent in the ATG group (60% vs 15%, P = 0.03). Haemorrhagic cystitis rate was higher in the ATG group (32% vs 5%, P = 0.03). CMV reactivations were similar between both groups (84% vs 65%, respectively, P = 0.9). Three patients from the ATG group presented EBV reactivation (one lymphoproliferative disease). Incidence of acute GvHD grades II-IV was higher in the ATG group (P = 0.03) as well as acute GvHD grades III-IV (P = 0.05, Figure 1 ), with no differences in moderate/severe chronic GvHD incidence (16% vs 19%, P = 0.6). With a median follow-up of 48 months (IC 25-75%: 45-61) for the ATG group and 9.5 months (IC 25-75%: 6-15) for the PTCy group, one-year OS was 60% vs 84% (P = 0.05), EFS was 59% and 75% (P = 0.14), relapse rate was 12% vs 8.6% (P = 0.4), and TRM was 28% vs 15% (P = 0.2), respectively, with no significant differences. However, results showed a tendency of better survival rates for the PTCy cohort. Conclusion: In our experience, albeit differences in follow-up and limited number of patients, PTCy combined with additional immunosuppression after MUD allogeneic HSCT offers lower rates of acute GvHD together with less toxicity and infectious complications compared to ATG-based prophylaxis. Further analysis including larger series and follow-up are needed to confirm these observations. Disclosure of Interest: None declared. Introduction: Graft versus host disease (GVHD) is the main cause of morbi-mortality after allogeneic stem cell transplantation (allo-SCT). Despite considerable advances in our understanding of the pathophysiology, nowdays anticipation of GVHD is an unresolved matter. Several single-nucleotide polymorphisms (SNPs) in cytokine genes have shown to be associated with donor-recipient alloreactivity and, ultimately, with SCT outcome. In this study, we propose a novel predictive model based on both clinical and genetic (SNP) variables applying an innovative estimation linear regression model, the least absolute shrinkage and selection operator (LASSO), in a large cohort of HLA-identical sibling donor allo-SCT. Material (or patients) and methods: The study evaluated 25 SNPs in 12 genes (Table 1) in genomic DNA obtained from PB samples from 273 patients included in the DNA Bank of the Spanish Group for Hematopoietic Stem Cell Transplantation (GETH) and their HLA-identical sibling donors. Each SNP was assessed for 4 different models of transmission. Clinical variables known to influence the development of GVHD were also considered ( Table 1) . Multivariant analysis was made with LASSO, which is able to select a set of optimal predictors from a large set of potential predictor variables. The best model is chosen by maximizing the area under ROC curve (AUC) and the correct classification rate (CCR). The statistical model was fitted by randomly selecting the 85% of the data and the predictive ability was computed with the remaining 15%. In order to evaluate the performance and the prediction ability of each model, training and testing samples were randomly selected a total of 100 times. For prediction purposes, we considered a cut-off value according to the proportion of Y = 1 in the sample. Results: The best clinical and genetic model to anticipate aGVHD II-IV included 11 SNPs with a CCR for patients who developed (CCR1) aGVHD II-IV of 63.6% ( Figure 1 ). The best model to anticipate aGVHD III-IV included 20 SNPs and 7 clinical variables with a CCR1 for aGVHD III-IV of 100%. To anticipate extensive chronic GVHD the model included 10 SNPs and 3 clinical variables with a CCR1 for extensive cGVHD of 80%. Predictive models with only clinical variables showed a poorer CCR1 for patients who developed aGVHD II-IV, aGVHD III-IV and extensive cGVHD (55.6%, 50% and 66.7% respectively; Figure 1 ). Based on the results from LASSO multivariate analyses, a risk score was calculated for grades II-IV and III-IV aGVHD as well as for cGVHD and extensive cGVHD. Patients were categorized into two groups: low risk (below the cut-off value) and high risk (above the cut-off). Such risk model was able to stratify patients who develop grades II-IV aGVHD ( p40.001), grades III-IV aGVHD ( p40.001) and extensive cGVHD ( p40.001) more consistently than models only considering clinical variables ( Figure 2 ). Conclusion: Identification of biomarkers useful for the estimation of the risk of GVHD constitutes an unmet need in the clinical management of GVHD. The novel predictive model proposed here, based on clinical and genetic factors, allows significantly improved anticipation of aGVHD III-IV (100% accuracy) and extensive cGVHD (80%) after HLA-identical sibling donor allo-SCT. This approach would allow a personalized risk-adapted clinical management of patients after transplantation. Disclosure of Interest: None declared. The urine proteomic pattern aGvHD_MS17 predicts acute GvHD and overall survival after allogeneic HSCT Introduction: Acute graft-versus-host disease (aGvHD) is a major complications after allogeneic hematopoietic stem cell transplantation (HSCT). We have developed a proteomic urine pattern "aGvHD_MS17" capable to predict aGvHD (1). We initiated a multicenter, randomized, placebo-controlled, double blind clinical trial to evaluate the proteomic based prediction of aGvHD in 2009. Patients after 1st HSCT in complete or partial remission for acute leukemia (AL) could be included. Patients were randomized upon positivity of the proteomic urine pattern aGvHD_MS17 prior to clinical signs of aGvHD. To date, data of 210 patients have been analysed for incidence of aGvHD grade II-IV and overall survival post-HSCT. Material (or patients) and methods: Urine was collected weekly from day +7 to +35, on days +50 and +80 post-HSCT and analysed within 72 h (1). The Pre-GvHD trial is based on the positivity (classification factor: 0.1 or more) of aGvHD_MS17. Patients with samples positive for aGvHD_MS17 were randomized and received either prednisolone (2 mg/kg BW) or placebo. The majority of the patients had AL (n = 120; 57%), were not in CR/CP (60%) and were transplanted from matched (n = 184;88%) donors. Reduced intensity conditioning regimens were used for 151 patients (72%) and 177 patients (84%) received immunosuppressive antibodies as GVHD-prophylaxis and a calcineurin-inhibitor (CSA/MTX n = 78; 37.4%; CSA/MMF n = 93 44.2%) based prophylaxis afterwards. Results: Prospective and blinded evaluation of the aGvHD_MS17 classifier revealed in this analysis the prediction of aGvHD II-IV in patient urine samples with a sensitivity of 71% and a specificity of 80%, if this classifier is combined with clinical parameters (2) . Data collection and data base clearing are still ongoing. Thus, we cannot evaluate the influence of the pre-emptive therapy yet. Instead, we analysed the patternpositive patients for the incidence of acute GvHD grade II-IV. Patients in the aGvHD_MS17-positive group have a 2.76 fold higher risk of developing aGvHD II-IV than those in the aGvHDnegative group (P = 0.0001. In the case of aGvHD_MS17 positivity, immune-supressive therapy-requiring aGvHD is predicted about 7 days (range 1-21) prior to clinical or biopsy based diagnosis of aGvHD. Both groups reach a plateau around day +50. Overall survival is significantly different for patients in the aGvHD_MS17 positive group (p o 0.0001, hazard ratio: 4.19), 40% of those die within the 1 st year after HSCT (day +80 and +250). In contrast, 90% of patients who never had aGvHD_MS17-positive samples survive the first year. Conclusion: Taken together the analysis of aGvHD_MS17 for aGvHD-prediction is highly reproducible. Especially in the early days +5 to +35 aGvHD is predicted with high accuracy. Additionally, aGvHD_MS17 accurately separates patients with good or bad overall survival after allogeneic HSCT. In fact 40% of patients with 1 sample positive for aGvHD_MS17 do not survive the first year after HSCT. notable interest due to their known capability to induce GVL without GVHD. Material (or patients) and methods: This analysis is a single center prospective study performed at the University Hospital Regensburg investigating the possible correlation between the regeneration of NK cell subsets and the incidence of acute GVHD (aGVHD) and vice versa during the first 200 days following alloSCT. Data were collected from 2009 to 2012, and 342 samples of 107 patients were analyzed by FACS with focus on immature CD56 high , mature cytotoxic CD56 dim NK cells, and the ratio between these two populations (CD56 dim :CD56 high ). Statistical analysis was performed using both logistic and B-spline linear regression. Results: While 45 (42.05%) patients did not develop any GVHD during the first 200 days, 62 (57.9%) patients developed aGVHD (grade 1 n = 22, grade 2 n = 21, grade 3 n = 14, grade 4 n = 5) in median at day 28 (11-182) after alloSCT. Interestingly, a lower number of CD56 high cells correlated with aGVHD (P = 0.045). In the longitudinal analysis aGVHD showed an impact on immune reconstitution after alloSCT. There was a clear association between incidence of aGVHD and delayed expansion of the total NK cell population (P = 0.032), in particular the CD56 high NK cells (P = 0.001). Remarkably, we observed a significant correlation between the severity of aGVHD and the lower recovery of CD56 high NK cells during aGVHD. Conclusion: It is well known that NK cell reconstitution following alloSCT goes along with higher numbers of CD56 high that further differentiate into cytotoxic CD56 dim CD16 high NK cells. In our study, aGVHD not only correlates with reduced numbers of total NK cells, but specifically with an impaired early expansion of CD56 high NK cells. In sum, these data let suggest a negative impact of aGVHD on early NK cell immune reconstitution, maturation and NK subset distribution. In addition, monitoring of early NK cell reconstitution after SCT may help to identify patients at risk for the development of severe aGVHD. Disclosure of Interest: None declared. Introduction: The outcome of patients underwent to allogeneic stem cell transplantation (allo-SCT) is closely related to graft versus host disease (GvHD) and graft versus leukemia (GvL) effects which, can be mediated by mHAgs. Twenty-three mHAgs have been identified and reported to be differently and variably correlated with GVHD or GVL, but a simultaneous method to genotype a so large panel of mHAgs has never been employed. The aim of this work was to develop a feasible method to genotype all the 23 mHAgs described so far and to test them for their correlation with GVHD and GVL in a group of donor/ recipient pairs submitted to allo-SCT. Material (or patients) and methods: For a multi-genotyping of 23 mHAgs we used iPLEX Gold Mass Array technology (3 multiplex) . We tested the antigens in 46 donor/recipient pairs full-matched that underwent allo-SCT (sibling or MUD) because of Philadelphia positive CML-Ph+ (n = 29) or ALL-Ph + (n = 17). Results: IPlex Gold Mass Array technology is accurate, highly automated and rapid. Our data show that sibling pairs have less mHAgs disparities despite MUD pairs. Notably, donor/ recipient genomic mismatch on DPH1 was correlated with an increased risk of grade ≥2 acute GvHD and LB-ADIR1 mismatch on graft versus host direction was related to improvement of RFS with no increase of GvHD risk. Conclusion: Our work provides a rapid, accurate and highly automatable methodology to genotype mHAgs and confirm the role of mHAgs in addressing the immune reaction between donor's lymphocytes and host. Work supported by Lions Club "Bassa Bresciana" and BCC di Pompiano e Franciacorta Founds. Disclosure of Interest: None declared. Introduction: Acute graft-versus-host disease (aGVHD) is the main cause of morbidity and is associated with non-relapse mortality (NRM) after allogeneic stem cell transplantation (HSCT). Systemic corticosteroids (CTC) remain the standard first line treatment of aGVHD, however several patients (pts) may require second line therapy. Moreover, CTC are associated with side effects and increased toxicity highlighting the need for CTC-sparing strategies. MTX is effectively used for GVHD prophylaxis. We hypothesize that this drug can also be effective in the curative setting. Material (or patients) and methods: We report here the use of low dose MTX as salvage therapy for aGVHD in 25 pts treated between January 2012 and July 2015. MTX was administered intravenous at weekly intervals either at the total dose of 5 or 3 mg/m2, according to pts hematopoiesis. Reasons for administrating MTX were aGVHD either refractory after systemic CTC (n = 18) or recurring after CTC tapering (n = 7). Results: Median age at HSCT was 53 (range 20-76) years and median follow-up for survivors was 26 (range 13-35) months. Diagnosis were AML (n = 12), MDS (n = 4), ALL (n = 3), MPD (n = 3) and LNH (n = 3). Eight of 16 pts with acute leukemia were in complete remission at HSCT. One pt had a previous HSCT. Type of donor was HLA identical sibling (n = 5), unrelated donor (n = 13), haploidentical donor (n = 5) and double cord blood (n = 2); stem cell source was peripheral blood in 22 pts. Reduced-intensity conditioning regimen (RIC) was used in 18 cases. Eight pts received a sequential regimen. ATG was administrated in 21 pts (18 with a total dose of 5 mg/Kg). Initial GVHD prophylaxis consisted of cyclosporine (CSA) alone in case of HLA identical donor (with short course MTX for ABO major incompatibility, n = 2), CSA plus mycophenolate for unrelated donors and post-HSCT cyclophosphamide was added in haploidentical HSCT. Among the 25 treated pts, 5 had grade II, 17 grade III, and 3 grade IV aGVHD. All pts had gut involvement (9 at stage 2, 12 at stage 3 and 4 at stage 4). In 1 pt aGVHD occurred after DLI administration. Concomitant GVHD treatments were CTC and CSA (in all but 5 pts, due to CSA-related toxicity). In case of skin involvement most pts received also weekly ECP. Overall, 10 pts experienced MTX toxicity: 9 hematological (3 mild, 5 moderate, 1 severe) and 1 moderate hepatic. Median number of administrated MTX doses was 6 (range 1-27); 8 pts received less than 4 doses (3 in 2 pts, 2 in 4 pts and 1 in 2 pts) due to hematological toxicity (n = 3), early death from acute GVHD (n = 4) or early complete response (CR, n = 1). Fourteen pts responded to MTX administration achieving CR; initial improvement was observed after the second infusion. In addition, MTX allowed a significant reduction of CTC dose with a median time to CTC interruption of 99 (range 49-298) days. Inolimomab was administered as third line treatment in 7 pts. Five pts relapsed under immunosuppressive treatment in a median time of 6 (range 1-19) months after HSCT. Overall 16 pts died, 11 of NRM (of these, 7 of GVHD related complications) and 5 of relapse. In our population the global CR rate to MTX was 56% with low toxicity profile in refractory aGVHD; 1-year overall survival was 48%. Conclusion: In conclusion, low dose MTX is a well-tolerated and a steroid-sparing agent. Our results support prospective controlled trials of MTX in refractory acute GVHD. Disclosure of Interest: None declared. Introduction: Graft-versus-Host Disease (GvHD) is a major cause of morbidity/mortality after allogeneic transplantation, occurring in 30-40% of matched unrelated donor (MUD) transplants. Steroids and cyclosporine A (CSA), the most common therapies, lead to high incidence of infections. Extracorporeal photopheresis (ECP) controls GvHD by inducing apoptosis of activated lymphocytes and increasing donor T-regulatory cells 1 . It involves infusion of autologous peripheral blood mononuclear cells, after incubation with the photoactive drug 8-methoxypsoralen (8-MOP) and exposure to ultraviolet irradiation. Main indication to ECP is steroidrefractory GvHD and the most common schedule is twiceweekly. This study adopted thrice-weekly ECP in order to suspend steroid treatment sooner and reduce the risk of infections. Material (or patients) and methods: Our Pediatric Transplant Unit recruited 6 patients (median age: 11 years; range: 5-21; 5 males, 1 female; 3 acute lymphoblastic leukemias; 2 severe aplastic anemia, SAA; and 1 acute myeloblastic leukemia secondary to Fanconi's anemia). Donors were: 5 MUD and 1 matched-sibling. CSA and short-term Methotrexate constituted GvHD prophylaxis. They developed acute GvHD, 4 grade III and 2 grade IV, which was treated with 2 mg/Kg/day steroids and mycophenolate mofetil, and/or tacrolimus, and/or sirolimus. Four patients developed chronic GvHD (2 limited, and 2 extensive). All were given thrice-weekly ECP for a median of 3 months (range: 2-7). ECP was then gradually reduced until suspension. The median follow-up was 16 months (range: 3-31). A 7 th patient who was at very high risk of GvHD (male, 18 years old, SAA, MUD transplant with peripheral HSC and two courses of ATG before transplantation) received thrice-weekly ECP as GvHD prophylaxis. Results: Steroids were discontinued to under 1 mg/Kg/day in all 6 patients within a median of 36.5 days (range: 9-56). Longterm GvHD control was achieved in all. ECP was reduced too rapidly in one poorly compliant patient who experienced GvHD exacerbation, which was resolved with further ECP and no other immune suppressive drugs. One patient developed invasive aspergillosis, which resolved with anti-fungal agents. The patient who received ECP as GvHD prophylaxis did not develop GvHD. In all patients peripheral blood T-cell subpopulation counts rose rapidly than patients who received standard GvHD treatment. CD4 + and CD8 + T-cells reached 100/ L medianly on, respectively, days 65 (range, 50-85) and 40 (range, 15-87); 200/L medianly on days 150 (range, 55-170) and 70 (range, 30-100). Naïve T-cells increased significantly after the fifth month post-transplant. CD20 + cells reached a mean of 120/L three months after transplantation. Specific CD4 + and CD8 + T-cells against pathogens such as A. fumigatus, C. albicans, CMV, Adenovirus, HSV, VZV, and T. gondii emerged significantly sooner (2 months vs 4 months in patients who received standard GvHD treatment) and remained stable over time. Conclusion: Thrice-weekly ECP was safe, feasible, controlling GvHD and steroid withdrawal. It was associated with a low incidence of infections and better immune reconstitution. ECP suspension needs to be slow and gradual to ensure good response. Introduction: Biomarkers present before the onset of chronic graft-versus-host disease (cGVHD) may aid in the clinical prognostication of later risk for cGHVD development following allogeneic hematopoietic stem cell transplant (HSCT). We investigated whether subsets of regulatory T cells (CD4 +CD25 high CD127 Lo ) (Tregs) may be prognostic biomarkers for the later development of cGVHD. Material (or patients) and methods: Fifty adults from the Vancouver cGVHD consortium study undergoing HSCT had blood drawn at day +100 (+/-2 weeks), before the onset of cGVHD. Multiparameter flow cytometry for Treg subsets was S182 performed and correlated to the later development of cGVHD. Participants were divided into four groups depending upon the presence or absence of acute (grades 1-4) and/or chronic GVHD. Data is presented as means +/-standard deviations. Statistical analysis was done by one-way ANOVA and receiver operator curve (ROC) analysis. Results: At day +100, patients with a history of both acute and later development of chronic GVHD (n = 7) had, as a proportion of total regulatory T cells, significantly lower percentages of CD31+ CD45RA-Tregs (5.5% +/-3.2%) compared to patients with acute GVHD only (18.9% +/-6.7%; n = 11), chronic GVHD only (18.5% +/-6.7%; n = 10), or no GVHD (19.4% +/-12.9%; n = 22) ( Figure) (P = 0.0009). Lower proportions of CD31+ CD45RA-Tregs were predictive for the later development of chronic GVHD in patients with a history of acute GVHD (ROC area under the curve = 0.97). Conclusion: CD31+ CD45RA-Tregs are a poorly defined subset of Tregs that have recently emigrated from the thymus (CD31+), have encountered antigen, and may be an early memory Treg cell (CD45RA-). Low proportions of CD31+ CD45RA-Tregs at 100-days following HSCT could be a valuable prognostic biomarker for the later development of chronic GVHD in patients with a history of acute GVHD, and may reflect thymic damage from acute GVHD post-HSCT. Further validation in larger cohorts of patients is warranted. Disclosure of Interest: None declared. Introduction: Chronic GVHD (cGVHD) is the first cause of transplant-related mortality and steroid-refractory (SR-cGVHD) represents an unmet clinical need. In 2005 NIH cGVHD Consortium recommended a standardized scoring and response criteria. We conducted a trial on Imatinib for SR-cGVHD to assessing feasibility of NIH response criteria and to validate their prognostic impact on survival. The new NIH criteria for cGVHD can now offer a shared framework, however, these criteria are not immediately suitable for use in clinical trials and a learning period is needed to train the practicing physicians and to highlight and correct possible criticalities emerging in the real life. Material (or patients) and methods: We have built a platform for data collection, based on a software prototype developed for internal use by our center for management of cGVHD. This software has already been structured according to new NIH 2015 recommendations and has been integrated with algorithms that automatically determine: organ-specific and global severity and organ-specific and global response according to the new NIH criteria (Complete: CR, Partial: PR, Mixed: MR, Unchanged or Progression: NR). Collection of data related to cGVHD reflects all the informations required in the standardized case report forms (CRF) redacted by the NIH consensus: clinical scales and measures reported by clinicians (FORM A) and patient-reported measures (FORM B); the evaluation of quality of life (SF-36v2) questionnaire; other ancillary measures. The platform also collects data about primary treatment and topical treatments. Event-driven collection of adverse events is requested, completed with grading according to CTCAE 4.0.Each step is mirrored by a dedicated CRF included in the software. The prototypic software has been developed by a small-medium IT (Information Technology) company (Bimind S.A.S., Jesi, Italy, http://www.bimind.it/en). While the main package (Dossiers) is integrated in the informatics environment of the Ancona Hospital, a stand-alone package has been created for cGVHD. Quality data control processes are integrated in the software. The software has been developed for daily clinical practice and is meant to be used onsite, in real-time during patient's clinical assessment; it is already compliant with all requirements for data protection, traceability of all activities, data backup. Results: We have successfully tested the performance of this software in a study evaluating the safety and activity of Nilotinib in 21 patients with SR-cGVHD, comparing the response (ORR) measured with this new tool with the old NIH criteria and with the specific end points of the study design. The ORR calculated with the software exactly corresponded to the ORR calculated by an expert physician, according to the updated NIH criteria. Both baseline and reassessment evaluations required a median of 15 minutes by using this software during the outpatient visit. Conclusion: We expect with this approach to enhance the quality of data and to reduce bias bringing the collection of data closer to their source. At the same time, an electronic tool structured according to the most recent evidences can improve daily clinical work and promote education. A brief demo of this software will be available for oral presentation. Disclosure of Interest: None declared. Introduction: the outcome of selected donor transplants for SAA has improved over the past decades. in this study, we analyzed the results of unmanipulated haplo-identical transplant and matched unrelated donor transplant for severe aplastic anemia with ATG (Genzyme, France) in conditioning regimen in our center. Material (or patients) and methods: Between January 2011 and August 2015, 42 consecutive SAA patients received unmanipulated PBSCT and BMT from haploidentical family donor and matched unrelated donor transplants (21 MUD,21 haplo-identical family donor). The conditioning regimen for MUD is Flu(30 mg/m2/d from D-10 to D-5) +CY(50 mg/kg/d from D -4 to D-3)+ATG(2.5 mg/ kg/d, from D-4 to D-1), for Haplo-identical is BU(4 mg/kg/d, from D-12 to D -1)/CY+Flu+ATG(the dose of CY/Flu/ATG as above). Both patients received cyclosporine (CsA), methotrexate (MTX), mycophemolate mofeil (MMF) for graft-versus-host disease (GVHD) prophylaxis, another mesenchymal stem cells (MSCs) derived from unrelated umbilical cord or the donor's bone marrow were cultured, and a total dose of 1.27 x 106/kg was infused on Day 0 to prevent graft rejection and reduce GVHD. Additionally for haplo-identical transplant, basiliximab Basiliximab was given at 20 mg/day on Days 1 and 4. Characteristics of patients N = 42 Median age: 9 (2-47Y) Median time from diagnosis to alloHSCT: 12 months(3-120 m) Rate of previous IST: 21.4%(n = 9) Transfusion dependency: 67%(n = 28). Results: the results as following table: Conclusion: seleted donor transplants included ATG in conditioning regimen for SAA can decrease middle and severe GVHD. The rate of GVHD is lower, hematopoietic recovery is more rapid, the rate of CMV infection is lower in MUD than in haploidentical transplant. Disclosure of Interest: None declared. Introduction: Extracorporeal photochemotherapy (ECP), as a safe and non-toxic immunotherapeutic method that is able to mediate patient's immune system without generalized immunosuppression, has been used in our center since 2007. After the establishment of multidiciplinary Center for cGVHD and long-term follow-up at University Hospital Center (UHC) Zagreb in 2013. all patients referred to ECP were scored according to established cGVHD-related grading scales and measurements in collaboration with the National Cancer Institute, USA, and patients previously treated with ECP were rescored. Material (or patients) and methods: The aim of the study was to report our 9 year experience with clinical and immunomodulatory effect of ECP, as well as adverse reactions associated with ECP.The influence of the EPC treatment on the levels of T-lymphocyte subsets, B-lymphocytes, and NK cells in blood was evaluated. Study was performed throughout the period of 2007-2015 in UHC Zagreb on group of 13 patients (7 male, 6 female) who were treated with off-line ECP. MNC were collected with Cobe Spectra and Optia cell separators. After the addition of psoralen products were irradiated with UVA on Macogenic device. Patients' median age was 32 years (range, 12-69 years). The patients suffered from generalized sclerodermatous skin changes, impaired join mobility and joint pain. In 3 patients the symptoms of oral disease have developed. Peripheral blood samples taken before and after leukapheresis, and samples from leukapheresis bag were analyzed for WBC, Htc, MNC, and platelet counts. Number of T-lymphocyte subsets (CD3+, CD3+4+, CD3+8+, CD4+8+ ratio) B-lymphocytes (CD 19+), and NK cells (CD56+) in patient's peripheral blood were taken monthly. Blood counts and parameters of MNC were measured by means of analyzer Advia 120, Bayer, USA. The levels of B, T lymphocytes and NK cells were evaluated by the use of flow cytometry technique, Becton Dickinson, Facs Calibur, USA. Results: In 13 patients with cGVHD, scored according to NIH classification as 11 severe and 2 moderate, 664 ECP procedures were performed. The median number of ECP procedures performed per patient was 49 (range 12-131). Clinical response to ECP is typically delayed until 2 to 3 months. Overall response, defined as either a complete response (CR) or a partial response according to NIH criteria, was obtained in 8 of 13 patients (61,5%), and CR, in 4 of 13 (30,7%). The effect of ECP in patients with skin and joint involvement was mostly benefitcal, as well as in all patients who suffered from the oral disease. At last follow up, 10 patients were alive and well and 3 patients died. In patients who responded well to ECP, CD4+/CD8+ ratio and number of NK cells were normalized. In general, ECPs were well tolerated, and main issue was adequate venous access for long term ECP treatment.. No increased incidence of infections and no serious adverse reactions have been observed so far. Conclusion: ECP is safe procedure that may be beneficial in treatment of cGVHD and can be recommended for patients who do not respond to conventional therapy. The specific influence of ECP on T-cell subsets leads to the suggestion that interactions between T-cell subsets may participate in the process of ECP. Disclosure of Interest: None declared. First and second generation TKI as salvage therapy for sclerotic chronic graft-versus-host disease I. Sánchez-Ortega 1,* , R. Parody 1 , O. Servitje 2 , C. Muniesa 2 , M. Arnan 1 , B. Patiño 1 , R. F. Duarte 3 , A. Sureda 1 1 Hematology, Catalan Institute of Oncology, Hospital Duran i Reynals, 2 Dermatology, Hospital Universitari de Bellvitge, Barcelona, 3 Hematology, Hospital Universitario Puerta de Hierro, Madrid, Spain Introduction: Sclerotic chronic graft-versus-host disease (ScGVHD) has limited and disappointing treatment options. Imatinib mesylate (IM) has been proposed as a valuable salvage therapy. Moreover, Dasatinib (DASA) is secondgeneration PDGF-R kinase inhibitor with a greater inhibitory potency and proven clinical efficacy in the treatment of CML patients refractory or intolerant to IM. We hypothesized that first and second generation TKI may be an effective therapeutic alternative for patients with refractory ScGVHD. Material (or patients) and methods: We describe a series of eight consecutive patients with ScGVHD who failed at least two previous immunosuppressive treatment lines and went on to receive salvage therapy with IM. Five patients were intolerant and/or refractory to IM and went on to receive DASA treatment. Median age was 53 (27-67) years, 5 male, 6 AML/MDS, 7 matched related donors, 5 myeloablative conditioning. Seven patients presented de novo chronic GVHD. Before the start of TKI treatment, all eight patients had severe chronic GVHD (Jagasia, BBMT 2015). Severe scoring was attributable to multiple severe organ involvement (skin, joints and fascia and/or lungs) in three patients and single skin involvement in five. All patients had sclerotic features including deep tissue sclerosis and hidebound lesions unable to pinch, two patients had severe ulcerations and three, generalized pruritus. Additional cGVHD targets included eyes (n = 3), nails (n = 3), hair (n = 3), mouth (n = 3), gastrointestinal (n = 3) and genital tract (n = 1). Results: Seven patients discontinued IM treatment (one case achieved complete response 49 months after IM initiation, five patients were resistant and/or intolerant after a median of 3 [1] [2] [3] [4] [5] [6] [7] [8] months and one case presented grade IV neutropenia two weeks after starting treatment that recovered after IM withdrawal). There was one non-relapse death (pneumonia of unknown origin) with a partial response of ScGVHD 35 months after starting IM treatment. Three of five patients on DASA achieved partial responses (decrease in clinician overall severity score, NIH skin, joints and fascia, mouth and eyes scores, increase in Photographic Range of Motion scores, improvement in Karnofsky Performance Status and resolution of skin ulcers) and could discontinue DASA treatment 57 (26-75) months after initiation. There was one non-relapse death (progressive lung GVHD) 28 months after starting DASA treatment. Finally, one patient previously intolerant to IM maintains stable disease after 2 months on DASA. At last follow-up, five patients have been on DASA for ScGVHD for a median of 28 (2-75) months with no grade 3-4 adverse events. There has been only one infectious complication (one pneumococcal meningitis) during TKI treatment. Overall, four patients (50%, one on IM and three on DASA) achieved partial responses and could discontinue TKI treatment and additional immunosuppressive treatments except one patient who continues on low doses MMF. Conclusion: TKI may represent a valuable option for patients with ScGVHD. Advantages include a well-established safety profile and oral route of administration. Our series suggests that DASA may be better tolerated and a more effective therapeutic alternative for patients with refractory ScGVHD. The evaluation of its role in this clinical setting needs further investigation. Disclosure of Interest: None declared. Introduction: There is a growing evidence of safety and efficacy of post-transplantation cyclophosphamide (PTCy) in related and haploidentical bone marrow (BM) transplantations, but the data regarding unrelated and peripheral blood stem cell (PBSC) transplants is limited. So we conducted a prospective trial of risk-adapted graft-versus-host disease (GVHD) prophylaxis with PTCy that included different types of donors and graft sources. Material (or patients) and methods: 200 adult patients (median age 32 y.o., range 18-62) with hematologic malignancies, including AML (47,5%), ALL (26,5%), CML (12%), MDS (4%), and lymphomas (10%), were enrolled in NCT02294552 trial. 23% of patients were classified as salvage. 26% received the graft from matched related (MRD), 65% from matched/ mismatched unrelated (MUD/MMUD), and 9% from haploidedntical (haplo) donor. 43% received BM graft and 57%4 PBSC graft. 18,5% had myeloablative conditioning and 81,5%reduced-intensity conditioning. GVHD prophylaxis for matched BM graft consisted of single-agent PTCy 50 mg/kg days+3,+4, for matched PBSC graft -PTCy+ tacrolimus+ mycophenolate mofetil (MMF) 30 mg/kg days 5-35, and for any mismatched graft -PTCy+ tacrolimus+ MMF 45 mg/kg days 5-35. Median follow-up was 7 months (range 2-27). Results: Grade II-IV, III-IV acute GVHD, and moderate-severe chronic GVHD in MRD, MUD/MMUD and haplo groups were respectively 10% (95% CI 5-24%) vs 18% (95% CI 12-25%) vs 11% (95% CI 7-58%), P = 0.37; 4% (95% CI 1-16%) vs 6% (95% CI 3-12%) vs 0%, P = 0.59; 12% (95% CI 5-28%) vs 11% (95% CI 6-20%) vs 9% (95% CI 1-59%), P = 0.88. Non-relapse mortality (NRM) and relapse incidence were 10% (95% CI 4-27%) vs 15% (95% CI 9-25%) vs 17% (95% CI 6-48%), P = 0.39, and 44% (95% CI 30-66%) vs 22% (95% CI 16-34%) vs 57% (95% CI 20-100%), P = 0.03 for MRD, MUD/MMUD and haplo group, respectively. 2-year overall survival (OS), event-free-survival (EFS), and GVHD-relapse free survival (GFRS) were 74% (95% CI 53-87%) vs 63% (95% CI 51-74%) vs 32% (95% CI 5-65%), P = 0.001; 45% (95% CI 28-62%) vs 62% (95% CI 50-71%) vs 26% (95% CI 1-63%), P = 0.02; 36% (95% CI 19-53%) vs 54% (95% CI 43-63%) vs 21% (95% CI 2-56%), P = 0.25 for MRD, MUD/MMUD and haplo groups, respectively. There was no statistical differences in OS, EFS and GFRS between MRD and MUD/MMUD (p40.05). In a multivariate analysis only salvage status (HR 3.6 95%CI 2.2-5.9, P40.0001) and occurrence of sepsis (HR 1.8 95%CI 1.0-3.0, P = 0.04) were predictive for EFS, while type of donor was not a significant factor , P = 0.24) (figure 1). The incidences of complications were: hemorrhagic cystitis -23%, sepsis -24%, severe sepsis -8%, invasive mycosis -8%, CMV reactivation -45%, venoocclusive disease -2.5%, transplant-associated microangiopathy -3.5%, grade 3-4 liver toxicity -14%, grade 3-4 kidney toxicity -1%. Conclusion: With relatively short follow-up we have demonstrated that the risk-adapted PTCy GVHD prophylaxis is safe and results in very low incidences of NRM, acute and chronic GVHD, and it could be used with any kind of graft source. It also alleviates the GVHD risk of MMUD and haplo donor. Relapse of underlying malignancy with this prophylaxis still significantly influences the outcome. Disclosure of Interest: None declared. Introduction: Refractory or relapsed chronic graft-versus-host disease (cGVHD) is complication of allogeneic HCT that significantly impacts quality of life, may be associated with morbidity and mortality and has limited treatment options. Two of the most promising treatment options are extracorporeal photopheres (ECP) and iterleukin-2 (IL-2) therapy. Here we report single-institution experience with these types of therapy. Material (or patients) and methods: 62 adult patients (pts) were included in analysis. 46 received ECP (MacoGenic, Macopharma, France), meadian 12 courses (range 3-38). 16 pts received IL-2 (Roncoleukin, Biothech, Russia) 1 MU 3 times a week, for 2.5 months median (range 1-7). Baseline severity of cGVHD (NIH) were not different between groups (P = 0.80). Overall 65% of patients had severe and 34% moderate GVHD, 71% had primary refractory and 29% relapsing cGVHD after steroid therapy. 66% of patients received steroids before the start of ECP or IL-2. 43% of patients had lung cGVHD. Response was assessed by physician assessment (clinical response), improvement in NIH scores and Karnofsky score. Any objective positive response was defined as improvement in either NIH scores or Karnofsky. Results: At 12 weeks there was no statistical difference between ECP and IL-2 in percentage of clinical response (72% vs 50%, P = 0.27) and improvement of Karnofsky (35% vs 19%, P = 0.35), but higher response in ECP group assessed by NIH scores (65% vs 31%, P = 0.023) and any response (70% vs 38%, P = 0.036). With a median follow-up of 40 months the best response was higher in ECP group when assessed reduction of NIH scores (70% vs 38%, P = 0.036), Karnofsky (57% vs 25%, P = 0.046) or presence of any objective response (76% vs 44%, P = 0.029), but not when assessed by physician (61% vs 50%, P = 0.53). In ECP group 86% of patients were off steroids, at median 6 months (range 3-9 months), while in IL-2 group only 3/6 patients were off steroids. Complete resolution of cGVHD was observed in 17% and 13% (P = 0.53) of pts in ECP and IL-2 groups, and 22% vs 13% (P = 0.71) were off all GVHD treatment. The non-relapse mortality was observed in 38% in pts with no objective response and 14% in responders S185 (P = 0.001). Relapse incidence was 15% in the whole group. Pts with lung cGVHD had worse probability of any objective response (79% vs 58%, P = 0.068) and a mean decrease of FEV1 -12 ± 27 l/s after second line therapy. Only 12% of lung cGVHD pts had improvement of lung severity score. Conclusion: With limited data on IL-2, ECP seems to be the more effective therapy of refractory cGVHD, but only a small proportion of patients have a complete resolution of cGVHD symptoms. Lung cGHVD has worse response to either ECP or IL-2, novel approaches are required for this complication. Disclosure of Interest: None declared. Introduction: Chronic graft-versus-host disease (GvHD) is a leading cause of morbidity and mortality after allogeneic hematopoietic stem cell transplantation. Our laboratory has previously found an association between an increased number of B cells primed for a TLR9 response and the onset of extensive cGvHD. TLR-9 is a Toll-like receptor, which recognizes unmethylated CpG dinucleotide sites in bacterial and viral DNA. Mitochondrial (mt) DNA, similar to bacterial DNA, is rich in unmethylated CpG motifs and can induce an immune response via activation of the TLR9 signaling pathway. Sustained elevation of free plasma mtDNA is postulated to be linked to increased cell necrosis in chronic inflammatory conditions. We hypothesized that the onset of chronic GVHD was associated with a significantly higher level of plasma free mitochondrial DNA. Material (or patients) and methods: Plasma cell free mitochondrial DNA (mtDNA) levels were measured in 39 adult patients post HSCT from 6 centers enrolled in an international cGvHD biomarker study. Sixteen normal adult controls were obtained from our local institution. In addition, 16 patients had cGvHD as diagnosed by NIH criteria with a mean onset of 209 ± 86 (S.D.) days post-HSCT and 31 who developed neither aGvHD nor cGvHD samples were obtained from 23 patients and measured at 3 (n = 10), 6 (n = 15), and 12 (n = 6) months post HSCT with 4 patients who had samples drawn at all three time-points. mtDNA was isolated from plasma and quantified by Q-PCR amplification of mitochondrial-specific COX1 DNA. The results for patients' post-HSCT were expressed as a foldincrease in the level of mtDNA compared to the average adult non-HSCT normal value. We also measured levels of other previously identified cGVHD biomarkers by ELISA and Luminex assay to correlate with mtDNA levels. Results: The levels of free plasma mtDNA were elevated in patients without cGVHD compared to normal non-HSCT adults at 3 months (2.36 ± 1.00 [mean ± S.D.] fold increase; P = 0.03), 6 months (4.19 ± 3.25 fold increase; P = 0.007) and 12 months (5.4 ± 4.66 ; P = 0.005) post-HSCT. The 6 month time point for no GvHD controls was chosen for comparison with cGvHD affected patients. This was based on a comparison of 3, 6, and 12 months post HSCT post HSCT controls with no GvHD controls which revealed no statistical difference at any time point. The 16 cGvHD onset samples were then compared to the 6 month post HSCT control and had a significantly higher level of free plasma mtDNA associated with the onset of cGVHD (6.49 ± 2.81 versus 4.19 ± 3.25 fold increase; P = 0.04). mtDNA concentrations were correlated to known cGvHD plasma markers on the identical samples. There was a significantly positive correlation between free mtDNA levels post-HSCT in both cGvHD index cases as well as the 6 and 12 month post HSCT no GvHD controls with paired samples for CXCL10 (P = 0.003), ICAM-1 (P = 0.007), CXCL9 (P = 0.03), sCD25 (P = 0.05) and sBAFF (P = 0.05). Conclusion: This pilot study highlights plasma free levels of mtDNA as a promising biomarker for cGVHD that may act as an inflammatory stimulus to TLR-9 expressing immune populations such as B cells. Further larger prospective clinical studies are required to confirm the role of recipient derived mtDNA as part of the onset of cGvHD. Disclosure of Interest: None declared. Introduction: Hypoxia-inducible transcription factors' (HIF) new role in coordinating the balance between regulatory T cell (Tregs) and T H 17 has been described lately [1] . It has been proposed for HIF-1 alpha to inhibit differentiation toward the Treg lineage by targeting FOXP3 fo proteasomal degradation [1, 2] . FOXP3 is a crucial transcriptional complex for differentiation and function of Tregs [3] . Tregs play a central role in suppressing autoimmune responses in graft versus host disease (GvHD) [3] , one of the most serious transplantation complications. On this backround, we found interesting if HIF-1 alpha expression in Tregs population may be associated with poor prognosis of GvHD patients. Material (or patients) and methods: We analyzed a 20 patients group, all with extensive chronic GvHD after hematopoietic stem cell transplantation (median age 40 (21-69), F/M -8/12, sibling/unrelated donor -9/11)). All patients were transplanted from fully matched donors and were refractory to corticosteroids thus qualified for second line treatmentextracorporeal photochemotherapy (ECP). The median time from transplantation at the treatment switch was 16, 5 months (4.5-39.7). All patients underwent MNC collection on 2 subsequent days (every two weeks) with the Optia AutoPBSC procedure and further UVA exposure of the collected cells was performed with UVA-PITs system. After 8 ECP procedures we classified 11 patients as non-responders (NR) and 9 of the original group as responders (R). We collected blood samples before the first and after eighth ECP procedure. All samples were analyzed for CD4+CD25+FOXP3+HIF-1α+ cells and CD4 +CD25+FOXP3+HIF-2α+ cells by flow cytometry. The criteria for diagnosis, staging and therapeutic response of chronic GvHD were used according National Institute of Health recommendations. Results: The median number of HIF-1 alpha positive T regs before ECP in NR group was 0.179x10 3 /ml (0-8.8x10 3 /ml) and in R group 0.057x103/ml (0-0.62x103/ml) and there was a significant difference between the groups (P = 0.04, Fig. 1 ). Similar results were obtain with the number of HIF-2 alpha positive T regs. The median number of HIF-2 alpha positive T regs before ECP in NR group was 0.134x10 3 /ml (0.077-3.57x10 3 /ml) and in R group 0.038x103/ml (0-0.26x10 3 /ml) and t was a significant difference between the groups (P = 0.01, Fig. 2 ). We did not find any difference between the groups after ECP procedures. Introduction: Haploidentical HSCT without in vitro T cell depletion, an alternative to treat haematological malignancies, has obtained comparable relapse rates, TRM, LFS, overall survival compared with the well-matched HSCT at Peking University 1 . The development of cGVHD affected the prognosis and life quality of patients 2 . Thus it is significant to find a novel marker to predict the development of cGVHD. Material (or patients) and methods: A total of 129 patients who received partially matched related donor HSCT at Peking University were enrolled under ethical committee approval and institutional procedures including written informed consent from each patient. After 90-180 days post-transplant, 46 of the patients developed cGVHD, according to which patients were classified into two groups. Peripheral cells from patients were collected on day 30, 60, 90, and 180 after allo-HSCT, and stained. Many factors were involved in the analysis, including underlying disease type, disease status, the age of recipients, the gender of donor and recipients, degree of HLA matching, the time post-transplant, age of recipients, HCT-CI score, the infused cell number, the development of aGVHD and cGVHD. The associations of these factors and selected cell populations were analysed using bivariate correlation analysis. Excluding other factors, independent t test was used to compare the number of CD4 + CD25 + CD45RO + cells from patients with or without cGVHD later. Results: When all possible factors were involved in the analysis, the number of CD4 + CD25 + CD45RO + cells assessed on day 30 post-transplant showed a significant correlation (Pearson correlation coefficient 0.228, P40.01) with subsequent development of cGVHD using bivariate analysis. The peripheral blood of patients who suffered cGVHD between day 90 and day 180 post transplant was found to possess more CD4 + CD25 + CD45RO + cells on day 30 post-transplant ( p40.01, mean difference 0.0177*10 9 /L, std. error of difference 0.006*10 9 /L, 95%CI 0.005*10 9 /L-0.030*10 9 /L), compared to patients without experiencing of cGVHD, independent of clinical parameters. Introduction: Rabbit anti-thymocyte globulin (ATG) is applied to reduce the incidence and severity of acute and chronic graft versus host disease (GVHD) after allogeneic hematopoietic cell transplantation (HCT). However, ATG did not improve overall survival (OS) in most prospective and retrospective studies. Recently, homozygosity for HLA-C group 1 (C1/1) killer-cell immunoglobuline-like receptor ligands (KIR-L) was revealed as risk factor for severe acute GVHD. Moreover, C1/1 recipients had favourable outcomes after ATG-based transplants. Here we test the hypothesis that C1/1 recipients benefit in terms of survival from an intensified GVHD prophylaxis incorporating ATG. Material (or patients) and methods: We retrospectively analysed 733 allogeneic (excluding haploidentical) HCT using peripheral blood or bone marrow grafts, for adults with predominantly malignant hematological diseases at two Austrian HCT centers. ATG-Fresenius or Thymoglobulin were applied in 227 and 87 HSCT, respectively, while 419 HSCT were performed without ATG. Median follow-up of survivors is 44.3 months (range, 3.3 -253.5). Results: Adjusted for recipient (R) and donor (D) age, disease status, HLA-match, R/D relationship, sex match, R/D cytomegalovirus serostatus, conditioning intensity, and type of postgrafting GVHD prophylaxis, Cox regression analysis revealed a significant association of ATG with a reduced the risk for nonrelapse mortality (NRM) (risk ratio/RR, 0.62; P = 0.009), but not overall mortality. In C1/1 recipients, (n = 274), ATG significantly reduced both non-relapse (RR, 0.42; P = 0.005) and overall mortality (RR, 0.6; P = 0.04). In HLA-C group 1/2 recipients (n = 342), ATG reduced NRM with borderline significance (RR, 0.61; P = 0.05), but had no impact on OS. In both, C1/1 and C1/2 recipients, there was no significant impact of ATG on relapse incidence and progression free survival (PFS). Considering only lower ATG doses (Fresenius ≤ 30 mg/kg and Thymoglobulin ≤ 5 mg/kg), ATG reduced the risk for overall mortality in C1/x (C1/1 or C1/2) recipients of HLA-matched PBSCT for malignant disease (RR, 0.68; P = 0.02; n = 369). By contrast, ATG neither reduced NRM (RR, 1.39; P = 0.24), nor overall mortality (RR, 1.54; P = 0.15) in C2/2 recipients (n = 117), but increased their risk for relapse (RR, 4.1; P = 0.04). Conclusion: These results suggest that ATG, particularly at lower doses, may provide a survival benefit by reducing NRM in recipients with at least one C1 group KIR-L allele. Introduction: Here, we present data from the long-term follow-up (median 8.6 years) from the prospective randomized phase III multicenter trial comparing a standard GvHD prophylaxis with cyclosporine A and methotrexate with additional pretransplant ATG-FRESENIUS S (given 20 mg/kg/ day, days -3 to -1) in unrelated donor hematopoieitc cell transplantation after myeloablative conditioning for 201 patients included with leukemia/ MDS in CR1/early (107 pts) or advanced (94 pts) disease. The previous results demonstrated a significant reduction of acute and chronic GvHD without compromizing relapse rate and survival. [1, 2, 3] . Material (or patients) and methods: Apart from chronic GvHD, NRM, relapse risk, DFS and OS, we analysed the effect of ATG-F vs control on the composite endpoint severe GvHD (acute GvHD III-IV, extensive chronic GvHD) and relapse-free survival, and on time under immunosuppressive therapy. Furthermore, we analysed conditional survival, i.e. the OS probability after having survived 1 and 2 years after transplantation. Results: Beyond 2 years late events were rare and main reasons for death were relapse with 10 and 8 patients in the ATG-F and control arm, respectively, and GvHD with 3 patients in the control arm, other reason 2 patients in ATG-F and control arm, each. In total, GvHD-related death occured in 1 (acute) patient in the ATG-F group and in 15 (9 acute and 6 chronic) patients in the control group. The incidence of extensive chronic GvHD after 8 years was 13.5% in the ATG-F group vs 51.8% in the control group ( p40.0001). The 8-year rates with respect to outcome were: NRM 20.5% vs 34.0% (P = 0.15), relapse 35.2% vs 29.9% (P = 0.54), relapse mortality 30.8% vs 28.8% (P = 0.90), DFS 44.3% vs 36.1% (P = 0.60), and OS 48.7% vs 36.8% (P = 0.31), ATG-F vs control, respectively. For patients transplanted in CR1 or advanced disesase the 8-year survival rates were 50% and 33%, respectively. ATG-F substantially increased the combined severe GvHD/ relapse-free survival rate. The rates were 48.5% vs 20.4% after 1 year and 33.6% vs 13.0% after 8 years (P = 0.0003), ATG-F vs control, respectively (see figure) . The probability of being alive and free of immunosuppressive therapy was 46.8% in the ATG-F group and 11.2% in the control group at 8 years (P = 0.0002). The conditional 8 years-survival probability increased in the ATG-F group from 48.7% (unconditional) to 70.6% and 80.9% (conditional on having survived 1 and 2 years after transplantation), and in the control group from 36.8% (unconditional) to 58.5% and 71.7% (conditional on having survived 1 and 2 years after transplantation). Conclusion: The sustained protective effect of ATG-F in addition to standard CSA/ Mtx GvHD prophylaxis is observed beyond 8 years without increasing relapse and compromising survival and results in significantly improved severe GvHD/ relapse-free survival. The use of ATG-F in unrelated donor transplantation after myeloablative conditioning substantially increases the probability of surviving free of immunosuppressive therapy. Introduction: Several genetic variants have been investigated in association with allogeneic haematopoietic stem cell transplantation (allo-HSCT) outcomes. Acute graft-versus-host disease (aGvHD) remains one of the most significant complications of the procedure. Cytokines play a pivotal role in the mechanism of aGvHD and many of the implicated cytokines relay their signals through the Janus kinase/signal transducer and activator of transcription (JAK/STAT) pathway. Material (or patients) and methods: The aim of the present study was to examine whether recipient and donor JAK2 46/1 haplotypes could have an impact on allo-HSCT outcomes in this cohort, including 124 recipient/donor pairs. The patients underwent HSCT for acute myeloid leukaemia in complete remission between January 2007 and December 2013 at our single centre. Data were collected and analysed retrospectively. For identification of JAK2 rs12343867 from genomic DNA LightCycler melting curve analysis (LightCycler 480II, Roche Diagnostics) was performed. Results: In our cohort both, recipient and donor JAK2 46/1 haplotypes were significantly associated with aGvHD grades II-IV development (P = 0.006 and P = 0.031, respectively), and their influence seemed to be additive (aGVHD rates were 14% in non-carrier patients and donors, 29% if the donor, 35% if the recipient and 46% if both, the recipient and the donor were 46/1 carriers). Including donor type (sibling versus unrelated, P = 0.031), conditioning (myeloablative versus reduced intensity, P = 0.036), recipient age at transplantation (P = 0.046) in multivariate analyses, the role of the recipient haplotype was independently related to moderate/severe aGVHD (P = 0.012), while for the donor only a tendency was found (P = 0.08). Underlining the independent role of both, recipient and donor haplotypes, the highest aGvHD grades II-IV incidence (46%) was found among those patients, who themselves and their donors also carried the haplotype, whereas a single 46/1 haplotype carriership either of the recipient or the donor resulted in an intermediate rise in aGvHD (35% if recipient and 29% if donor was a carrier) compared to double negative cases (14%). Recipient haplotype favourably affected relapse rate (P = 0.004), but not the donor haplotype (P = 1). Overall survival (OS) did not differ according to recipient haplotype (P = 0.732), but there was a tendency for worse OS if donors were carriers for the 46/1 haplotype (P = 0.056). Conclusion: The common JAK2 46/1 haplotype represents a well-established susceptibility locus for myeloid neoplasms. The exact pathomechanism has remained unexplored, but inappropriate myelomonocytic response to cytokine stimulation, increased risk of inflammation, and impaired defense against infection have been suspected in the presence of 46/1 haplotype. Our findings suggest that the recipient and donor JAK2 46/1 haplotype might be involved in the development of aGvHD. Promising results of JAK2 inhibitors in the management of aGvHD provide further supporting evidence for the involvement of the JAK2 pathway in aGvHD. . The aim of this study was to compare the validity of these two TCE assignment methodologies to inform clinical practice. Material (or patients) and methods: We previously demonstrated a significant impact of non-permissive HLA-DPB1 donor-recipient disparity on the risk of mortality according to the functional TCE assignment (Pidala et al, Blood 2014) in a cohort of 2730 HCT patients. Here we re-analyzed the outcomes when DPB1 disparity was assigned according to the FD TCE assignment. The associations with clinical endpoints overall survival (OS), transplant related mortality (TRM), disease free survival (DFS), relapse, acute GvHD (aGvHD) and chronic GvHD (cGvHD) were studied using multivariate proportional hazards methods. Results: 2005/2730 (73.5%) HLA-DPB1 mismatched pairs were classified concordantly as permissive (N = 1402) or nonpermissive (N = 603) by the two models. Of the remaining 725 pairs, 676 classified as permissive mismatches in the functional TCE assignment switched to the non-permissive group in the FD TCE assignment, and 49 pairs vice versa. These differences were predominantly due to a shift from TCE group 3 into TCE group 2 for alleles that were undefined by the functional TCE assignment, but could be assigned according to the FD TCE assignment. In multivariate analysis using the respective permissive groups as reference, the previously reported significant associations of non-permissive mismatches according to the functional TCE assignment were confirmed with the FD TCE assignment for OS (P = 0.0015 vs P = 0.028) and TRM ( p40.001 vs P40.001) (Table) . Discrepant findings were seen for DFS (P = 0.007 vs P = 0.159), aGvHD II-IV (P = 0.635 vs P40.001) and cGvHD (P = 0.959 vs P40.001). However, when comparing the subgroup of 676 pairs permissive pairs according to the functional but not the FD TCE assignment to the overlapping permissive pairs as reference, significant differences were found only for aGvHD II-IV (P = 0.107 vs P40.001 vs) and cGvHD (P = 0.513 vs P40.001). Conclusion: The predicted FD TCE assignment yields consistent results with the original functional TCE assignment for OS and TRM, however appears to be better predictive of aGvHD II-IV and cGvHD. The FD TCE assignment will facilitate the clinical use of TCE group matching for all unrelated donor searches regardless of the presence of rare or new HLA-DPB1 types. Disclosure of Interest: None declared. (14), Escherichia coli ESBL (10), carbapenem-resistant Pseudomonas aeruginosa (7), other (5), 2 or more pathogens (9). Previous colonization with alert pathogens was significantly correlated with MDR infections (P = 0,00001). There was higher incidence of infections with MDR bacteria in patients with aGvHD (P = 0,014), chronic GvHD (P = 0,002) and the older age-group (P = 0,002). The analysis of MDR-colonized group showed statistically higher rate of mortality (P = 0,034) and deaths due to infection (P = 0,00076) in contrast to the noncolonized group. Conclusion: 1. Patients colonized with alert pathogens are more susceptible to develop life-threatening infections caused by MDR microbes. Therefore, to lower the risk of fatal outcome in case of FUO without or preceding microbiological identification, MDR bacteria colonizing the patient must be treated as the potential etiological factor and antibiotic therapy should be adjusted, especially in GvHD group. 2. More frequent routine cultures (twice a week) are essential to detect the MDR-colonization prior to development of serious infectious complications. 3. Preventive strategies for eradication of MDR bacteria must be considered. Disclosure of Interest: None declared. Introduction: BSIs are known to be an important cause of morbi-mortality among pts undergoing allo-HSCT. The reported incidence has been up to 36%. Aim: to describe the incidence and characteristics of BSIs among pts who underwent allo-HSCT at the U. H. Donostia during the last decade. Material (or patients) and methods: a total of 210 pts admitted into the hospital for allo-HSCT were included (table 1). The vast majority of pts received prophylactic ciprofloxacin plus an azole (fluconazole or an extended spectrum azole if high risk for invasive mold infection). Blood cultures (BC) were repeated in a daily basis in persistently febrile pts. When a positive (+) culture was considered to be due to contamination during the collection or processing of the sample, it was classified as "false BSI". "True BSI" was considered when a microorganism that does not usually contaminate BCs was isolated and when a microorganism which is frequently contaminant was isolated in, at least, 2 samples. Although both groups are reported, the analysis is focus on the true BSIs. "Breakthrough bacteremia" was defined as a BC(+) despite appropriate antibiotic therapy, after a previous BC(-). Results: a total of 1907 BCs were performed (9.1 BC per pt on average). 1637 BCs (85.8%) were (-), 134 (7%) were false BSIs, and 136 (7.1%) were true BSIs. The characteristics of the isolations are shown in table 2. The 136 isolations corresponded to 69 true episodes in 57 pts (incidence: 27.1% BSI per pt, on average). There was a unique episode of breakthrough bacteremia (caused by E. coli), which was successfully treated. Five of the E. coli and the K. pneumoniae isolated were extended spectrum beta lactamase (ESBLs) producers. There were also 2 cases of E. cloacae carbapenemase (NDM type) producers, and 1 carbapenem-resistant P. aeuruginosa (no carbapenemase). Mortality related to the BSI episode was 1.4% (3pts). Conclusion: 1) BSIs was frequent among pts undergoing allo-HSCT (27.1%), although related mortality was relatively low (1.4%); 2) Gram+ bacteria represented two thirds and Gramone third of the isolations, while fungemia was a rare event. 3) One third of the isolated Gram-bacteria showed some kind of multi-resistant antibiotic susceptibility pattern, as expected in a population multitreated with antimicrobials. 4) The number of BCs performed per pt was very high (9.1), while it was detected a single case of breakthrough bacteremia; this fact suggests that it is feasible and reasonable to reduce the frequency of BCs performed in clinical stable febrile pts with no infectious focus. Disclosure of Interest: None declared. Introduction: Human adenovirus (HAdV) infections are frequently observed in pediatric allogenic SCT patients and seem to follow reactivation of latent virus from the gastrointestinal tract during the early post transplantation period. Subsequent dissemination with high virus loads in the peripheral blood results in a high lethality of up to 50%. HAdV types of species C account for about 80% of these cases. By contrast, the incidence of HAdV infections in adult allogenic SCT recipients is lower but the case fatality rate is in the same range. More than 40 types of species HAdV-D have been described so far, representing the biggest of the seven HAdV species. Recombination between different HAdV-D types, probably during (co-)infections of the gastrointestinal tract with more than one type, promotes their high genetic diversity. Several HAdV-D types were isolated from faeces of terminal AIDS patients in the 1980 s suggesting their role as opportunistic pathogens. The most recently published HAdV-D type 70 was isolated from an adult SCT patient with chronic diarrhea more than 100 days post SCT (1). Material (or patients) and methods: Patient: A 54 year old male presented on day 93 post transplantation with symptoms of encephalitis (somnolence, vertigo and reduced reaction time). The patients had received allogenic peripheral blood stem cells from an unrelated HLA mismatched donor (HLA-DRB1 and HLA-DQB1) after myeloablative treatment with Fluarabin, ATG und TBI. Lymphocyte counts had recovered to 1250/μL but the patient was immunosuppressed with Cyclosporine A and Prednisone. Because of increasing somnolence, the patient was transferred to the ICU and was intubated. Encephalitis with an oedema of the Hypothalamus and Thalamus was diagnosed by MRI. In spite of antiviral therapy with cidofovir the patient's encephalitis remained refractory. ICU treatment was discontinued and he died on day 122. Virological methods: quantitative (real time) PCRs and next generation sequencing (NGS) were performed. Results: Diagnostic virology. Neurotropic viruses were not detected in CSF but a high concentration of HAdV DNA (3e5 c/ml). In contrast to typical acute infections, HAdV DNA was not detected at other body sites (throat swab, stool, urine). In peripheral blood, only a low virus load (below the level of quantification, o1e3 c/ml) was found, far too low for a disseminated infection. Therefore, HAdV encephalitis, probably caused by a HAdV reactivation, was diagnosed. HAdV genome analysis. Complete genomic sequencing (by "NGS", directly from CSF) demonstrated a multiple recombinant HAdV-D genome with a hexon sequence (including the neutralization epitope) almost identical to type 25 whereas other parts of the genome were either related to types 47 or 56 or 59 or were novel sequence stretches, for example in the E3 region which codes for proteins interfering with the immune system. According to the rules of HAdV taxonomy, the pathogen should be labelled as a new HAdV type. Conclusion: Reactivations of species D HAdV may occur in adults as late complications after allogenic SCT. These can cause difficult to diagnose infections limited to a single organ (e. g. encephalitis) although many infections may be less severe. By contrast, species D HAdV types have hardly ever been detected in the recent systematic monitoring programs of faeces and blood samples from pediatric SCT recipients. Introduction: Hemorrhagic cystitis (HC) remains a common complication after allogeneic hematopoietic cell transplantation (alloHCT). Its pathogenesis, risk factors and effects on survival are not well known. Recent studies have shown that polyoma BK viruria interacted with regimen intensity and donor type leading to a very high risk of hemorrhage after transplant. The objective of this study was to evaluate the prevalence of HC in patients undergoing alloHCT and to analyze the different risk factors involved in the duration and resolution of each HC episode. Material (or patients) and methods: Retrospective study of patients submitted to alloHCT (n = 189) who had an episode of HC (n = 38) in a single centre between November 2007 and July 2015. The following variables were collected and analyzed: patient and treatment characteristics (gender, age, diagnosis, graft source, donor type, conditioning regimen intensity and GVHD prophylaxis) as well as the characteristics of each episode (grade of hematuria, duration and resolution, platelet count, immune reconstitution, polyoma BK viruria and urinary infection), the complications after alloHCT (GVHD, During the first 100 days after HSCT, 807 infections requiring treatment were recorded (median 3 per patient, range 0-10), of which 49% of bacterial, 37% of viral and 11% of fungal origin. IRM cumulative incidence (CI) at 100 days was 14% (10%418%), in particular due to bacterial (9%), viral (3%) and fungal (2%) infections, while at 2 yrs it was 25% (CI 20% 431%). The cut-offs of biochemical variables identified by ROC analysis and significantly associated with 100 day IRM were: ANC o1050 cells/uL (P = 0.019), IgA o1.11 g/L (P = 0.012), IgM o0.305 g/L (P = 0.003), serum iron 4106.5 ug/dL (P = 0.023), ferritin 41473.5 ng/mL (P = 0.044) and CRP 417.5 mg/L (P = 0.036). In multivariate analysis taking into account all variables, only age 460 yrs (P = 0.003), CMV seropositivity ( p40.001), IgM o0.305 g/L (P = 0.020), IgA o1.11 g/L (P = 0.004) and CRP 417.5 mg/L (P = 0.041) were able to significantly predict CI of IRM. Noticeably, this association was independent from disease type and status, use of alternative donors, intensity of conditioning, use of ATG and TBI, AB0 incompatibility, previous MDR-GN colonization or infection. Conclusion: In addition to previously recognized clinical (age, CMV seropositivity) and biochemical pre-transplant variables (CRP), IgM and IgA levels may predict the risk of death from infections after HSCT. These data suggest a rationale for the prophylactic infusion of polyclonal immunoglobulins to protect patients from IRM and potentially improve overall HSCT outcome. Disclosure of Interest: None declared. Introduction: Correct identification of risk factors for IFD is considered of paramount importance to assist clinicians in determining which patients may require more aggressive antifungal therapeutic strategies. The hematopoietic cell transplantation comorbidity index (HCT-CI), based on the assessment of 17 comorbidities, has been found to influence the non relapse mortality (NRM) in hematopoietic cell transplantation (HCT) recipients. In the present study we sought to investigate the association between pretransplant HCT-CI, development of IFDs and subsequent mortality. Material (or patients) and methods: Between January 2009 and March 2015, 312 consecutive patients with complete HCT-CI risk score assessment were analysed. Overall, 36% of the patients were ≥ 55 years old and 63% had an acute leukaemia. The donor was a matched sibling in 33% of the cases, a matched unrelated donor (MUD) in 51% of the cases and a haploidentical relative in 16% of the cases. A myeloablative preparative regimen was used in 78% of the patients. Overall, 55% of the patients had a low or intermediate HCT-CI risk score (0-2) whereas 45% had a high HCT-CI score (≥3). Antifungal prophylaxis included fluconazole or mold-active agents in 83% and 12% of the patients, respectively, while 5% of the patients received secondary antifungal prophylaxis. Results: The cumulative incidence (CI) of proven-probable IFD at 1 year from HCT was 8,5%, with a significant higher incidence among patients with high HCT-CI (12%) as compared with patients with HCT-CI 0-2 (5%; P = 0,006). The univariate analysis confirmed that an high HCT-CI was associated with a higher incidence of IFD (SDHR 2,57; P = 0,010). Molds were responsible of IFDs in all cases except one. Overall, 136 patients died, 71 of progressive disease and 65 of transplant related complications including 9 IFDs. The 1year CI of NRM was higher in patients with high HCT-CI (30%) and patients with IFD (49%) compared to patients with low HCT-CI (12%) and patients with no IFD (16% S194 NRM, and highlight its additional prognostic value on the outcome of HCT recipients who develop IFDs. These findings may be considered of clinical relevance, since they could pave the way for appropriate strategies aiming to implement diagnostic interventions and maximize antifungal prophylaxis in patients with high HCT-CI. Disclosure of Interest: None declared. The role of galactomannan test in bronchoalveolar lavage in the diagnosis of invasive aspergillosis in children after hematopoietic stem cells transplantation Introduction: Bacterial infections are the most prevalent infectious complication in Hematopoietic Stem Cell Transplantation (HSCT). Among them, the incidence of gram-negative rods (GNR) infection is increasing, mainly due to resistant forms. The aim of this study is to analyze the incidence and characteristics of GNR infections in HSCT, in the absence of antibacterial drug prophylaxis. Material (or patients) and methods: GNR documentations from blood and urine cultures during the first 2 years of 382 HSCT (164 autologous and 218 allogeneic), performed in the period from 2002 to 2014 in our hospital, were retrospectively analyzed. The incidence of GNR infections and their differences were analyzed and evaluated according to the type of transplantation (autologous vs allogeneic), age (over or under 50 years) and the condition for which the HSCT was performed, by chi2 and U-mann-Whitney tests. The patient characteristics are shown in Table 1 . Results: 47% of transplants (181) presented infections by GNR, with a total of 407 isolates: blood culture (168) and urine culture (239). The most isolated microorganisms were E. coli (n = 180; 44.2%), E. cloacae (n = 57; 14%), K. pneumoniae (n = 43; 10.6%) and P. aeruginosa (n = 30; 7.4%). Among autologous transplants, 63 patients (38.4%) were infected by GNR, compared to 118 (54.1%) in allogeneic ones (P = 0.005). These infections present earlier in the autologous transplant than in allogeneic HSCT (day +58 vs day +8, P40.0001). In the age group ≥ 50 years, no differences in the source of isolation (P = 0.863) or in the median day of isolation (P = 0.546) were found. When analyzed by disease, patients with multiple myeloma had fewer GNR infections (33%) than other diseases; acute leukemia/MDS (55%), lymphoma (51%) and others (60%) (P = 0.007). The median day of isolation was +55 in AL/MDS, +8 in lymphomas, +19 in MM and +75 in others ( p40.0001). Introduction: CYTOMEGALOVIRUS (CMV) infection has long been suspected of causing myelosuppression in a subset of marrow transplant recipients. This supposition is supported by in vitro studies, which have identified several different mechanisms by which CMV may interfere with hematopoiesis.In the current report, a distribution analysis was used to test the hypothesis that gB type will influence the manifestation of CMV infection in patients receiving marrow transplants. The endpoint of interest was death from infectious complications of neutropenia, which appeared to be significantly associated with the relatively rare gB genotypes 3 and 4. Material (or patients) and methods: The study included 110 bone marrow transplant recipients with CMV infection, defined as positive with highly sensitive nested-PCR.This number is composed of 3 samples from 110 consecutive patients with positive virus nested-PCR from the plasma, Buffy coat and urine,with or with out CMV disease, who were categorized to estimate the gB frequency distribution among all the infected patients. Likewise, with few exceptions, the patients studied were not enrolled in early treatment protocols involving ganciclovir or foscarnet.DNA was extracted from viral isolates using a polymerase chain reaction (PCR)-compatible lysing buffer.The Samples were digested with proteinase K at 55°C for 1 hour. PCR was conducted using primers gB1319 and gB1604, which amplify a region of high peptide variability in the gB gene corresponding to a portion of gp55. Depending on the genotype of the virus, these primers yielded amplimers of 293 to 296 bp. The amplimers were then digested with Hinf1 and Rsa1 in two separate reactions. The digested DNA was resolved into informative banding patterns on 12% polyacrylamide gels. Results: gB type was included as an explanatory variable in addition to other pre-and post-transplant factors known to impact on neutropenia. These included patient age, donor CMV serostatus, and occurrence of acute GVHD. Similarly, Cox regression was used to examine the association of gB type with the hazard of acute GVHD.The Mean of 25% of transplant donors were infected totally with HCMV infection. The decline pattern of the prevalence of UL55 genotypes in the plasma, Buffy coat and Urine samples were as follows:gB24gB34gB14gB4, gB24gB14gB34gB4, gB24gB34gB14gB4.PCR amplification and restriction digestion of the gB gene yielded informative banding pattern for all the 300 isolates studied.The similar frequency of death due to myelosuppression in patients with gB types 1 and 2 was compared with 3 and 4,which also suggested functional similarities between CMV strains with gB3 and 4 provided with rational for combining type 1with 2,and 3 with 4 for the multivariable analysis.A simple comparison of these proportions leads to a difference that is highly significant. HC was observed in 11/51 (21.5%) and 2/54 (3.7%) among MAC and RIC HSCT, respectively ( p40.025); 11/13 (85%) pts with HC were BK positive in the urine in comparison to 52/92 (56%) without. JC urine positivity associated to TRM ( p40.005), whereas both urine JC ( p40.01) and urine BK ( p40.05) associated to any cause mortality. Furthermore, JC urine activation was more often observed following a RIC regimen ( p40.005) but associated to a lower acute grade II-IV GVHD rate ( p40.03). Conclusion: Our data support the view that HC is not simply caused by BK reactivation but is rather the result of multiple risk factors, with alternative donor and MAC regimens reaching statistical significance. The relatively low rate of JC plasma reactivation and the lack of PML cases point out to possible geographical and ethnical differences. JC rather than BK urine reactivation proved to be a hallmark of immunosuppression with notable effect on the clinical outcome, as testified by increased mortality and lower risk of acute GVHD. Moreover, the higher risk of JC reactivation in RIC HSCT recipients may be viewed as deriving from the higher burden of immunosuppressive drugs required in these pts. Disclosure of Interest: None declared. Introduction: Infectious complications are frequent in patients after allogeneic hematopoietic stem cell transplantation (HSCT). Community-acquired respiratory virus (CARV) infections are caused by diverse viral agents circulating throughout the year, but some show seasonality. Respiratory tract infections (RTI) range from asymptomatic replication to significant disease with high morbidity and mortality especially after HSCT. Parainfluenza virus (PIV), Influenza, Respiratory syncycial (RSV) and human Metapneumovirus (hMPV) are some of the most frequently detected CARVs after allogeneic HSCT; therefore we analyzed our patients in a retrospective manner. Material (or patients) and methods: All patients after allogeneic HSCT with respiratory symptoms (outpatient unit or hospitalized due to any reason) were screened for CARVs in daily routine between 07/13-06/14. PIV, Influenza, RSV and hMPV were analyzed in a deep throat swab by singleplex PCR. All patients with a positive PCR were included in this retrospective, single-centre study to evaluate the incidence of symptomatic upper respiratory tract disease (URTID) or lower respiratory tract disease (LRTID: pathological sputum production, hypoxia or pulmonary infiltrates with a positive PCR in respiratory secretions). Additionally we were interested in the persistence of the virus as well as risk in factors such as IgG levels, CD4 counts, GvHD, immunosuppressive medication, patient age and time after HSCT. Results: 658 patients were seen in our outpatient unit or on the ward, 90 of them within the first 100 days after HSCT. We diagnosed 34 PIV infections, 16 hMPV and 2 Influenza infections. No differences were found for patient age (median: 53; range: 20-73), sex, conditioning regimen or IgG level. Most cases of PIV infection were diagnosed between d+100 and d +365 (n = 20), hMPV was also frequently detected after d+365 (n = 7; 44%). At the day of diagnosis, 30.8% of patients were treated with immunosuppressive medication (n = 16) (Steroids n = 7; CNI n = 8; mTOR-Inhibitor n = 1). 10/34 (29.4%) PIV patients and 5/16 (31%) hMPV-patients were suffering from an acute or chronic GvHD. Most patients complained about dry cough and rhinitis, 2 patients reported about fever. All 52 patients had a URTID, only 3 showed a LRTID and had to be treated on ward. In all 3 cases we found PIV with a bacterial super infection. None of the patients died. In most patients the virus persisted for a few weeks, even after clearing of symptoms (median: 28 days; range: 4-98 days Conclusion: Preliminary data of our aHSCT cohort study show that fever before engaftment is common. In spite of a fluoroquinolone prophylaxis, breakthrough bacteremias occur, with a low early mortality rate. Lymphoma patients have a significantly higher chance to develop both Gram-positive and Gram-negative bloodstream infections than individuals with myeloma. Stratified management is crucial to improve the outcome of early infections therefore further analysis of multiple risk factors is planned (14). BSI etiology was classified as either mucosal barrier injury (MBI) from neutropenia or GVHD, central-line associated (CLA) or secondary. We aimed to study the temporal distribution of BSI post-allo-HCT as well as its distribution by infection source and transplant indication. Results: A total of 29 patients experienced 37 episodes of BSI. 14 episodes occurred pre-engraftment (PE BSI), 11 occurred early (between engraftment and D +100), and 12 after D +100 (Late BSI). PE BSI: Most patients received prophylactic systemic antibiotics during neutropenia (levofloxacin or ceftazidime) per protocol. 13 patients had 14 BSI occurring at a median of 8 days post-transplant. 4 were caused by gram-negative (GN) bacteria, 5 by gram positive (GP) bacteria and 5 were polymicrobial. 6 patients had hematologic malignancy and 7 had immunodeficiency (5 CGD). Etiology was MBI for 8 and CLA for 6 episodes. Early BSI: 11 episodes among 10 patients occurred a median of 61 days post-transplant. 7 episodes were caused by GN and 4 by GP bacteria. Nine episodes were considered CLA and 2 MBI. Late BSI: 12 episodes occurred among 9 patients at a median of 180 days post-transplant. 11 occurred among patients with relapsed hematologic malignancy and only 1 in a patient with immunodeficiency (GATA2). 8 of 12 episodes occurred in the setting of retreatment neutropenia. 5 episodes were MBI, 5 CLA, 1 secondary to pneumonia (Citrobacter koseri identified in a BAL) and one primary S. pneumoniae bacteremia. 6 episodes were due to GN and 6 due to GP. Microbiology: A total of 43 microorganisms were isolated: 24 GP, 18 GN and 1 yeast. Staphylococcus aureus (6), Staphylococcus epidermidis (6) and viridans-group streptococci (5) were the most common GP. Enterobacteriaceae (11, including 7 Klebsiella pneumonia) were the most common GN. There was only one case of Pseudomonas aeruginosa bacteremia. Six of 21 patients with BSI died, compared with 8 patients of the 78 who never had BSI, but no deaths were attributable to BSI. Conclusion: BSI is common in all phases after allo-HCT, but its attributable mortality is low. Post-transplant BSI is primarily caused by either intravascular catheters or mucosal breaks. In our series late BSI occurred more frequently in the setting of relapsed malignancy and retreatment neutropenia than among those transplanted for immune deficiency. Disclosure of Interest: None declared. Thirty two out of 65 underwent a HSCT: 2(6.2%) autologous HSCT and 30 (93.7%) allogeneic HSCT. Of the remaining 33: 6 patients (18%) were affected by multiple myeloma, 19 (57.5%) by lymphoma, 5 (15%) by chronic lymphocytic leukemia and 3 (9%) by others including systemic mastocytosis and amyloidosis. LMV prophylaxis was administered in 45 (69%) subjects positive to anti-HBc +/-anti-HBs, in 17 (26%) isolated anti-HBs positive subjects (none of them vaccinated) and in 3 (5%) HSCT recipients from an HBV-positive donor. Results: Patients were evaluated every three months by monitoring the trend of HBV antibodies, the HBsAg antigen appearance, the HBV-DNA and transaminases levels. The median duration of LMV prophylaxis was 18 months (range, 1-125) and indication was to prolong it up to 2 years after the end of immunosuppressive therapy. Six out of 65 (9,2%) experienced HBV reactivation after a median of 24 months (range, 2-40), 2 (33%) completed LMV prophylaxis 3 months before HBV reactivation. Four patients (66%) reactivated HBV during LMV treatment. Three reactivated patients sero-reverted in HBsAg positivity with disappearance of HBsAb and HBV-DNA 42000 UI/mL without transaminases elevation, 3 other patients experienced virological reactivation with appearance of HBV-DNA o2000 UI in presence of HBsAb. HBV genotype D was detected in all cases, 2 (50%) isolates had immune escape HBsAg mutations and 1 LMV resistance mutation (L80I). Conclusion: HBV reactivation occurs in hematological patients under LMV prophylaxis. It is accompanied by seroconversion anti-HBs to HBsAg, high HBV-DNA, and/or mutations in RT and in HBsAg that affect treatment efficacy and immune-mediated control of the virus. Quarterly clinical and laboratoristic followup was helpful in detecting viral reactivations, occurring both during LMV prophylaxis and after the interruption, as recommended by recent guidelines. Disclosure of Interest: None declared. Fecal microbiota transplantation after allogeneic HSCT for curing recurrent Clostridium difficile infection: Why using the stem cell donor again? C. Robin 1,2,* , M. Paul 3 , B. Nebbad 4 , F. Beckerich 1,2 , R. Lepeule 5 , N. Ait Ammar 2,6 , C. Rodriguez 2,7 , C. Cordonnier 1,2 1 Hematology, Henri Mondor Hospital, 2 UPEC, 3 Pharmacy, 4 Bacteriology, 5 Antimicrobial stewardship team,Microbiology, 6 Parasitology, 7 Virology, Henri Mondor Hospital, CRETEIL, France Introduction: Clostridium difficile infection (CDI) is a common complication after allogeneic hematopoietic stem cell transplantation (HSCT). Fecal microbiota transplantation (FMT) is the treatment of choice for recurrent CDI. We report the first FMT performed in an allogeneic HSCT recipient, using the same stem cell and feces donor, and argue for this choice. Material (or patients) and methods: A 51-year old man without any prior history of CDI was transplanted from his HLA identical sister after conditioning with Fludarabin, Busulfan and ATG for chronic lymphocytic leukemia with del 17p. Three months after HSCT, he developed an obstructive syndrome, restricting the forced expiratory volume (FEV) of 61% of the theoric, and then grade II skin and liver graft-versus host disease (GVHD). He was successfully treated with steroids and cyclosporine. Six months after HSCT, the patient received cefotaxime for pulmonary bacterial superinfection. Abdominal pain and diarrhea occurred eight days later, and CDI was diagnosed. The patient was not neutropenic and had no severity criteria of CDI. He was successfully treated with oral metronidazole for 10 days. Seven months after HSCT, while the patient was still receiving cyclosporine and 20 mg/d of prednisone, FEV deteriorated (48% of the theoric), and extensive ocular GVHD was diagnosed, thus prednisone was increased to 30 mg/d with local treatment for ocular GVHD. Three, non-severe, episodes of CDI consecutively occurred (8, 9 and 10 months after HSCT) and were successfully treated with 10 days of vancomycine then a 10-day course of Fidaxomicin and then Fidaxomicin until the day before FMT. The patient at this time had normal blood counts, was full donor chimerism, receiving cyclosporine and hydrocortisone 20 mg. Due to these recurrent CDI episodes, we proposed FMT. However, few data are available on FMT in immunocompromised patients. The risk of bacterial translocation could be higher than in other patients. Additionally, any antigenic stimulation can in theory trigger a new episode of GVHD. In order to improve the tolerance of FMT, we sollicitated his sister -the HSCT donorfor feces donation. She was tested in the stools and blood according to French guidelines (Sokol H et al. Dig Liver Dis. S199 2015 Sep 7) for FMT and found eligible. Fresh feces were prepared and diluted in 200 ml sterile saline (0.9%) solution. The microbiota sample (12 g) was instilled to the recipient through a nasojejunal tube 12 months after HSCT. Results: No complication was observed after FMT, especially no fever, gut symptoms, or bacteremia. Five months after FMT, no recurrence of CDI has been observed. Moreover, we did not observe any exacerbation of GVHD, and the respiratory function tests and ocular symptoms improved, allowing a progressive decrease of the immunosuppressors without rebound of GVHD. Conclusion: This is the first report of FMT using the same donor for HSCT and FMT. The procedure was safe, and confirmed its expected efficacy for recurrent CDI. Moreover, it should be noticed that FMT did not trigger any gut GVHD. In allogeneic HSCT, we hypothesize that FMT from the same donor would be better tolerated that from any other one, and could minimize the risk of gut antigenic stimulation, and consequently the risk of triggering and/or worsening GVHD. Further studies are needed to confirm this hypothesis. Disclosure of Interest: None declared. Introduction: Vancomycin is a glycopeptide antibiotic with a narrow therapeutic interval. According to the current PK/PD (pharmacokinetic/pharmacodynamics) model an AUC-MICratio of4400 should be reached for the majority of bacteria, usually translating into target trough levels e between 10 and 15 mg/l. With comon dosing strategies (1000 mg q12h for adults), this target seems not to be reached successfully, especially at the beginning of the treatment. Therefore, we introduced a weight-based loading dose and a maintenance dose dependent on weight and on renal function as well as an early therapeutic drug monitoring. Material (or patients) and methods: We conducted a retrospective chart review including two groups of adult patients: Consecutive patients receiving the standard dosing (SD) strategy from May to September 2014 and patients receiving the loading dose (LD) strategy from June to October 2015. Only those patients with the first vancomycin serum level measured before the 7 th dose of vancomycin were included. Results: 29 patients in the SG and 26 patients in the LD group were evaluable. The groups were similar regarding age (53 vs. 59 yrs.), weight (78 vs. 75 kg) or renal function (serum creatinine 0.88 vs 0.89 mg/dl; GFR 97 vs 87 ml/min) at the start of vancomycin. Mean loading dose was 20.7 mg/kg. First vancomycin serum levels were determined after a median of 4 (SD) vs. 2 (LD) applications of vancomycin. Median first vancomycin levels were 7.3 (2.7-27) vs 10.5 mg/l (3.8-17.5) with a significant lower variance (Mann-Whitney-U-Test P = 0.02) in the LD group (standard deviation: 5.7 vs. 3.5). Conclusion: The use of a weight-based loading dose of vancomycin results in a more accurate target attainment of serum trough levels. Furthermore, the earlier achievement of steady state concentrations allows an earlier quantification and possibly necessary dose adjustment compared with the standard fixed dosing strategy. Therefore, the use of vancomycin loading dose seems to be useful to reach the aim of early initiation of antibiotic treatment with adequate PK/PD parameters. Disclosure of Interest: None declared. Introduction: Central venous catheters (CVCs) are extensively used in patients undergoing allogeneic hematopoietic cell transplantation (HCT). In these patients CVCs are placed routinely either via internal jugular (IJV) or subclavian veins (SCV). Several studies have compared rates of complications between both insertion sites, particularly in the intensive care unit setting. However, there are no data in the literature comparing the insertion sites in HCT patients. Thus, it was the purpose of this study to analyze systematically complications of CVCs at the different insertion sites in these patients. Material (or patients) and methods: All consecutive patients who underwent HCT at our institution from 01/2011 to 06/2013 were included into this retrospective analysis. Inclusion criterion was insertion of a CVC due to HCT. 3-lumen standard nontunneled CVCs (Arrow, Reading, PA) were placed after informed consent either via the IJV or the SCV. CVCs were inserted by experienced physicians after disinfection by 70% propanolol under full barrier precautions. The insertion site was chosen at the responsible physician´s own discretion. CVCs were covered by chlorhexidine gluconate impregnated dressings, either as transparent polyurethane (Tegaderm CHG) or as sponge dressing (Biopatch, Ethicon). No patient received a systemic antibiotic prophylaxis. Reasons for premature removal of CVCs were suspected or proven CVC-related blood stream infection (BSI), progression of local infections at the insertion site, or any other CVC related severe adverse event. Study end points were local infections, fever, BSI, duration of catheterization, CVC congestion, thrombosis, pneumothorax and suspected or proven catheter related deaths. Thrombosis occurred only once in SCV-CVC. Pneumothorax occurred twice after failure to place the CVC in the SCV. In both cases the CVC was placed thereafter in the IJV. Congestion of CVCs was observed in 28% (17/60) of IJV-and in 34% (14/41) of SCV-CVCs (ns), respectively. 7 patients died with an inserted CVC (4 x IJV, 3 x SCV), one of them potentially CVC-related. Conclusion: In contrast to earlier studies on patients without hematologic malignancies these data demonstrate clearly that CVCs placed in the SCV are not superior. Moreover, local infections and fever occurred earlier and more frequent in CVCs placed in the SCV. Disclosure of Interest: None declared. Introduction: Encephalitis is a prevalent and highly lethal post-transplant complication, still taking many difficulties in its diagnosis, with consequent early inadequate management. The causative role of herpes virus 6 (HHV-6) in post-transplant encephalitis has been recently well-described in cord blood HSCT, but is still underestimated in the haploidentical setting (haplo-HSCT). Material (or patients) and methods: We refer to the most accepted definition for HHV-6 encephalitis, consisting in the evidence of limbic encephalitis (with typical clinical and/or MRI signs) and demonstration of HHV-6 in cerebrospinal fluid; diagnosis is possible even without the evidence of HHV-6 in CS fluid, when limbic encephalitis is associated with plasmatic high viral load and other causes of CNS disease have been excluded. According to these criteria, our institution in 7 years (2009-2015) recorded 14 cases of HHV-6 encephalitis after haplo-HSCT: adult patients (M 9, F 5; median age 51y) who received PBSC grafts for high-risk disease (8 AML; 2 ALL; 2 NHL; 2 HD). The majority of patients underwent unmanipulated graft, followed by rapamycine-based GvHD prophylaxis; 5 of them received antithymocyte globulin, 6 post-transplant cyclophosphamide. Only three AML patients received T-cell depleted graft. HHV6 viral load was assessed by quantitative PCR. Results: HHV-6 encephalitis presented in all cases with acute onset of disorientation and confusion; common symptoms were important memory loss (4), convulsive seizures (3), hyponatremia (severe in 2 patients). One patient experienced hypothermia and intermittent axial myoclonus. Electroencephalography showing disrupted general organisation was performed in 12 cases; epileptiform abnormalities were also detected in 36% of cases. MRI revealed limbic encephalopathy in 64% of cases, often consisting in temporal and hyppocampal lesions (figure 1). All patients had HHV-6 plasmatic reactivation, with a median viral load of 16824 copies/ml. Lumbar puncture was performed in 8 patients, demonstrating HHV-6 positivity in all cases, with a median viral load of 89261 copies/ml in cerebrospinal fluid. Median time at the beginning of encephalitis was 25 days after HSCT. At the time of viral positivity all pts were receiving acyclovir as prophylaxis, except three. Eight patients experienced acute GVHD, in most cases requiring high dose steroid treatment. All patients received antiviral pharmacological treatment, using as first choice therapy foscarnet, even though the most four severe cases required a prolonged combination therapy with foscarnet and ganciclovir. Resolution occurred in 6 cases, with an overall mortality rate of 57% at 6 months. The patients who survived were mostly treated with foscarnet or foscarnet+ganciclovir (5), one was treated with ganciclovir alone. Conclusion: HHV-6 encephalitis carries a poor prognosis after haplo-HSCT. As the common post-transplant antiviral prophylaxis is not effective against HHV-6 disease, an early diagnosis and adequate treatment with combined antiviral therapy are needed to improve patients survival. Regular HHV-6 monitoring, and early CNS evaluation through lumbar puncture and brain MRI in case of presentation of the aforementioned highly conserved clinical pattern is recommended in post-transplant management. Disclosure of Interest: None declared. Microbiological surveillance in the BMT Unit: analysis of a single centre experience D. Vincenti 1,* , C. Annaloro 1 , G. Saporiti 1 , A. I. Gregorini 2 , M. Arghittu 3 , C. Marrazzo 1 , E. Tagliaferri 1 , A. Dodaro 3 , F. Onida 2 , A. Cortelezzi 2 1 BMT Center, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, 2 BMT Center, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico -University of Milan, 3 Microbiology Laboratory, Fondazione IRCCS Ca' Granda Ospedale Maggiore Policlinico, Milan, Italy Introduction: Microbiological surveillance in hematologic wards, including BMT units, is frequently considered a low output strategy with possible resource wasting. This report is aimed at summarizing a single Centre experience on this field. Material (or patients) and methods: Over a seven years period, 369 HSCT recipients (autologous 232, allogeneic 137, male 190, female 179, median age 52 years, range 19-71) underwent microbiological monitoring including weekly swabs of mucosae and cutaneous folds. Apart from considering the series as a whole, the observation time was also divided into two equal periods of 3.5 years/each in order to allow further univariate comparisons. All patients received fluoroquinolone prophylaxis and each febrile episode was treated with piperacillin/tazobactam (PTZ) and amikacin. The results of microbiological surveillance were compared with those of 158 non-coagulase negative staphylococci (CNS) bacteremias. Results: One microbiological positivity as a minimum was observed in 266/369 HSCT recipients. 703 positive samples (Gram+ 293, Gram-410) were disclosed at least once in one or more cutaneous sites/patient and 420 (258 Gram+, 162 Gram-) at least once in one or more mucosal swabs/patients. Over the 2 time periods, the increase in Gram-and the decrease in Gram+ were statistically significant ( p40.01), although Gram+ still built up the majority of mucosal isolates. Among Gram+, Enterococci (326) and Staphylococcus aureus [SA] (173) were the most common isolates. Vancomycin Resistant Enterococci (VRE) were 29/326 (8.9%): 26/217 (12%) in the former and 3/109 (2.7%) in the latter period ( p40.01). SA was methicillin resistant in 53/70 (76%) isolates from the former period and 57/103 (55%) in the latter ( p40.05). Among Gram-, Escherichia coli (283) and Klebsiella pneumoniae (92) were the most common isolates, with 60 strains being esbl+; no clear-cut trend towards an increase over time of esbl positivity was observed. Among the other isolates, the mostly remarkable were Acinetobacter baumannii (13), Enterobacter spp (53) and Pseudomonas aeruginosa (68). Carbapenem resistance was observed in 28 strains, of whom 15 Pseudomonas aeruginosa. Of 572 Gram-, 37 were PTZ sensitive/ceftazidime (CFZ) resistant, 29 were PTZ resistant/CFZ sensitive; 94 were simultaneously resistant to PTZ and CFZ. There was a nonsignificant trend over time toward an increase in PTZ resistance. In 56/158 non-CNS bacteremias, infection was heralded by previous mucosal colonization; SA (16) and Escherichia coli (15) were the most frequent. Conclusion: Cutaneous swabs, where the increase in Gramcolonization occurred earlier, bear the potential of predicting the evolution of mucosal findings. Severe bacteremias are frequently heralded by colonization by the same agent, so enabling to plan a tailored antibiotic regimen in case of antimicrobial resistance. Despite long-term use of the same prophylaxis and therapy, the increase in both Gram-colonization and bacterial resistance was rather slow. Unexpected decrease in VRE was likely due to factors other then local policies, such as the decrease of antibiotics use in livestock. Disclosure of Interest: None declared. Comparative results between micafungin and intravenous itraconazole as primary anti-fungal prophylaxis in allogeneic stem cell transplantation. Retrospective study in a single centre D. Serrano 1,* , M. Kwon 1 , J. Gayoso 1 , G. Rodiguez-Macias 1 , A. garcia 1 , C. munoz 1 , P. Balsalobre 1 , J. L. Diez-Martin 1 1 Introduction: Invasive fungal infections (IFIs) are an important cause of morbidity and mortality in allogeneic hematopoietic stem cell transplantation (HSCT). In patients with cancer a meta-analysis showed that primary antifungal prophylaxis (PAP) with mold-active antifungals agents reduced the documented fungal infections (Ethier BJC 2012). Micafungin, an echinocandin with activity against Candida and Aspergillus, has a good safety profile, even in patients with liver or kidney impairments, and it can be used at the same time with the conditioning treatment (CT). It has no significant drug-drug interactions. We have changed the PAP with intravenous itraconazole by micafungin, during the neutropenic period in HSCT. The aim of this retrospective study is evaluate the success with each PAP and incidence of IFI in every cohort of patients. Material (or patients) and methods: Between 2006 and 2010, 67 patients, cohort 1, received fluconazole together CT, and started intravenous itraconazole on the day +1 (200 mg/d, load dose 400 mg first day). In cohort 2 73 patients received micafungin (50 mg/d) since CT. Both PAP were used until resolved mucositis and oral fungal prophylaxis could be started. Intravenous PAP was changed by empirical or preemptive/targeted antifungal treatment when they were needed.We present the characteristic and comparative results of the patients treated with each of these drugs. Results: Both cohorts presented the same median age (42 vs. 43 ) and similar gender distribution. In cohort 2 there were more patients diagnosed with Hodgkin`s disease and chronic mieloproliferative syndrome and less with acute lymphoblastic leukemia and multiple myeloma, Besides, in this cohort, more patients showed advanced disease, and received more haploidentical progenitors stem cell. There were no differences in CT, and in neutrophils engraftment in both cohorts. Successful PAP was similar in both cohorts (54 vs 60%), Little differences in empirical (40 vs 36%) and pre-emptive (6 vs 4%) antifungal treatment were observed. In the neutropenic period our patients had a very low number of IFIs, 5 (7,5%) and 3 (4%) respectively, and IFI-associated mortality was the same, 1 patient, in every cohort. Table 1 . Conclusion: During the neutropenic period in allogeneic stem cell transplantation, micafungin was well tolerated with no S202 associated toxicity or drug-drug interactions, and was as effective as itraconazole in preventing invasive fungal infections. Disclosure of Interest: D. Serrano Funding from: This study was partially supported by a grant from Astellas Farma., M. Kwon: None declared, J. Gayoso: None declared, G. Rodiguez-Macias: None declared, A. garcia: None declared, C. munoz: None declared, P. Balsalobre: None declared, J. L. Diez-Martin: None declared. Introduction: Allogeneic hematopoietic cell transplant (allo-HCT) recipients are at risk of cytomegalovirus (CMV) reactivation due to compromised immune systems. The objectives of this study were to examine the clinical impact of CMV among allo-HCT patients, including frequency of co-infections, length of hospital stay (LOS), and mortality. We also examined the incremental economic burden from the hospital perspective. Material (or patients) and methods: Recipients of allo-HCT between January 2010 and December 2012 were identified from hospital discharge records of the National Inpatient Sample (NIS) database. The NIS database does not enable patient follow-up and thus our study focused on the initial allo-HCT hospitalization. The costs reported in the database represent the estimated costs to hospitals. Allo-HCT recipients were grouped into 2 cohorts: patients with CMV infection and those without, based on International Classification of Diseases, Ninth Revision (ICD-9) diagnosis codes. Demographics and hospitalization characteristics, including coinfection frequencies, inpatient mortality, and hospital LOS were evaluated. The cost to hospitals was also compared between cohorts. Results: We identified 4186 allo-HCT recipients (mean age: 41 years), of which 58% were female and 20% were aged less than 18 years. Of allo-HCT recipients, 8.6% (n = 358) were diagnosed with CMV infections during the initial HCT hospitalization. Patients diagnosed with CMV had a greater likelihood of co-infection with double stranded DNA viruses, including BK virus (5.9% vs. 1.8%, P40.0001) and adenovirus (3.1% vs. 0.8%, P40.0001) relative to those without a CMV diagnosis during the initial allo-HCT hospitalization. Mean hospital LOS was significantly longer for patients with a CMV infection versus those without (54 vs. 32 days, P40.0001). Inpatient mortality rate was also significantly higher among patients diagnosed with CMV compared with those who were not (19.6% vs. 5.6%, P40.0001). The estimated mean hospital cost was significantly higher for patients diagnosed with CMV versus those without ($250,759 vs. $139,948, P40.0001; median $193,566 vs. $108,087). Conclusion: Patients diagnosed with CMV during the initial allo-HCT hospitalization had significantly higher rates of coinfections and inpatient mortality and greater hospital costs than those without CMV diagnosis. The hospital LOS was over 50% longer among patients diagnosed with CMV, resulting in approximately twice as high hospital costs compared to those without reported CMV infection. Since the NIS database does not contain information about CMV viral load or subsequent CMV infections after HCT hospitalization, the estimate of CMV and other virus infection rates in this study are likely an underestimate of the overall rate of CMV infection in this population. Therapeutic strategies that can reduce the rate of CMV infection during the initial allo-HCT hospitalization may potentially impact on patient outcomes and the economic burden to hospitals. Introduction: Infection with BK virus (BKV), a doublestranded DNA virus, is an important cause of hemorrhagic cystitis after allogeneic hematopoietic transplant (HCT). This study assessed the real-world outcomes associated with BKV infection among allogeneic HCT recipients enrolled in a US health plan. Material (or patients) and methods: The MarketScan Research Databases were used to identify enrollees with an International Classification of Diseases, Ninth Revision (ICD-9) or current procedural terminology code for an allogeneic HCT between June 26, 2010 and June 30, 2014. Eligible patients had 360 days of plan enrollment prior to HCT. The first HCT procedure (transplant date) for each patient was defined as the index procedure. Patients with an ICD-9 diagnosis code for BKV during the first 365 days following the transplant were included in the BKV group. These patients were compared with those without any documented BKV diagnosis. Cox proportional hazards models were used to compare incidence rates, and Wilcoxon-Mann-Whitney tests were performed for continuous outcomes. Results: Among 3035 allogeneic HCT patients identified, 95 patients (3.1%) were included in the BKV group (mean age 38.8 years, 63.2% male) and 2940 patients (96.9%) were included in the no-BKV group (mean age 47.6 years, 56.7% male). In the BKV group, mean (median) time from transplant to first BKV diagnosis was 93.8 (71.0) days. A significantly higher proportion of patients in the BKV group was identified with documented cystitis compared to the no-BKV group (n = 41, 43.2% vs n = 131, 4.5%; incidence rate 0.07 vs 0.01 per patient-month; P4.0001). Patients in the BKV group also had significantly more hospitalization days during the 365 days post-transplant (mean [median] hospitalization days per patient-year: 109.2 [86.4] days vs 96.0 [57.0] days; P = .0085) and higher readmission rates (n = 78, 83.9% vs n = 1561, 55.1%; incidence rate 0.22 vs 0.11 per patient-month; P4.0001). The mortality rates were similar across the two groups, as measured by the number of hospital discharge records indicating death or transfer to hospice (n = 20, 21.1% vs n = 482, 16.4%; incidence rate 0.03 vs 0.02 per patient-year; P = .3951). Conclusion: Patients with BKV infection experienced a higher burden of cystitis and hospitalization than patients without BKV infection. The true incidence of BKV infection may have been underestimated because of a lack of screening for BKV, as well as under-reporting due to varying ICD-9 coding practices. Given the higher morbidity associated with BKV in allogeneic HCT recipients, strategies that minimize the incidence of BKV infection could potentially reduce the overall clinical and economic burden in this patient population. period. This study assessed the health outcomes among pediatric recipients of allogeneic HCT enrolled in a US health plan, who had a diagnosis for Adv infection. Databases were used to identify enrollees with an International Classification of Diseases, Ninth Revision (ICD-9) or current procedural terminology code for allogeneic HCT between June 2010 and June 2014. Eligible patients were less than 20 years old and had 360 days of plan enrollment prior to HCT. Patients who had an ICD-9 diagnosis code(s) for AdV during the 365 days following the transplant were included in the AdV group. These patients were compared to those without any documented AdV diagnosis codes. Cox proportional hazards models were used to compare incidence rates, and Wilcoxon-Mann-Whitney tests were performed for continuous outcomes. Results: Among 350 allogeneic HCT pediatric recipients identified, 34 patients (9.7%) were included in the AdV group (mean age 7.4 years, 73.5% male) and 316 patients (90.3%) were included in the no-AdV group (mean age 10.5 years, 57.0% male). In the AdV group, mean (median) time from transplant to first AdV diagnosis was 91.2 (62.0) days and 5 patients (14.7% of AdV patients) had their first AdV diagnosis during the transplant admission. All 34 patients in the AdV group had other documented opportunistic infections while 259 patients (82.0%) in the no-AdV group had such infections (incidence rate 0.68 vs 0.30 per patient-month; P = .0123). A higher proportion of patients in the AdV group had documented non-AdV dsDNA viral infections (n = 21, 61.8%) compared to the no-AdV group (n = 105, 33.2%; incidence rate 0.13 vs 0.05 per patient-month; P = .0017). Patients in the AdV group also had significantly more hospitalization days during the 365 days post-transplant (mean [median] hospitalization days: 151.8 [135.2] days vs 100.3 [63.8] days; P = .0003) and higher readmission rates (n = 26, 86.7% vs n = 178, 58.9%; incidence rate 0.20 vs 0.11 per patient-month; P = .0118). Furthermore, these patients also appeared to have a higher mortality rate as measured by hospital discharge records indicating death or transfer to hospice (n = 13, 38.2% vs n = 46, 14.6%; incidence rate 0.05 vs 0.02 per patient-month; P = .0024). Conclusion: Pediatric patients with AdV infection experienced a significantly higher burden of hospitalization, readmission and mortality, and were also more vulnerable to opportunistic infections. Mortality may have been under-or overestimated as transfer to hospice care was used as a proxy for death, and deaths occurring outside of the hospital would not appear in the insurance claims. The true incidence of AdV may be higher, as not all AdV infections were likely to have been diagnosed and non-specific coding practices may lead to under-reporting of identified AdV infections. Introduction: Respiratory syncytial virus (RSV) is the most common cause of respiratory viral infections in patients undergoing haematopoietic stem cell transplantation (HSCT) and is associated with increased morbidity and mortality. Little is known about the best management strategy in this high risk group and there are very little data on oral ribavirin treatment. The outcome of adult allogeneic HSCT recipients with RSV infection treated with oral ribavirin was analysed at a single UK centre. Material (or patients) and methods: We performed a retrospective analysis of 23 consecutive RSV cases treated with oral Ribavirin between December 2010 and February 2015. Patient characteristics: male 12, median age 52 years (range 20 to 69). Underlying diagnoses were acute leukaemia 11, myelodysplasia 4, aplastic anaemia 3, multiple myeloma 3, chronic leukaemia 2. Allogeneic HSCT characteristics: 16 reduced intensity conditioning (RIC) and 7 myeloablative conditioning; stem cell source: peripheral blood 20, bone marrow 2, cord blood 1; volunteer unrelated donor 16, sibling allograft 6 and 1 cord. Results: Diagnosis was established by polymerase chain reaction (PCR) assay as part of routine screening and, in addition, for patients presenting with respiratory symptoms. At diagnosis, 7 patients presented with lower respiratory tract infection (LRTI) as defined by new infiltrate on chest X-ray (CXR), signs on auscultation or hypoxia, whereas 16 experienced upper RTI. We initiated oral Ribavirin at a dose of 15 mg/kg divided into 3 daily doses with no subsequent dose escalations, as per local protocol. The median treatment duration was 10 days (range 5 to 47). Additional therapies were administered concomitantly as follows: 3 patients received immunoglobulin replacement due to low immunoglobulin levels, 17 patients received antibiotics and 3 also received antifungal treatment due to a strong suspicion of superimposed bacterial or fungal infection. Ribavirin was well tolerated. Two patients experienced adverse effects: nausea in 1 patient and haemolytic anaemia in 1 patient, although unproven to be ribavirin-related. In total, 19 patients completed an oral treatment, whereas 4 were escalated to aerosolised ribavirin. Sixteen remained outpatients throughout and 7 required hospitalisation with a median admission period of 7 days (range 1 to 16). None of the patients required intensive care admission. After a median follow-up of 17 months (range 4 to 48), 17 patients are alive and 6 died. One death was considered to be RSV-related in a patient who developed RSV pneumonitis on a background of relapsed disease. Conclusion: RSV in post allogeneic HSCT patients remains a challenge. Prompt initiation of treatment in immunocompromised patients is essential and may avoid unnecessary hospital admission. Our experience supports the use of oral ribavirin in adult HSCT recipients, nevertheless larger prospective studies are needed to determine the optimal therapy for this patient group. Introduction: Immunization is the only efficient way to prevent yellow fever (YF), a life threatening viral hemorrhagic fever. Yellow fever vaccine (YFV) is a life-attenuated vaccine classically contraindicated in immunocompromised patients (pts). Despite vaccine guidelines allow YFV in allogeneic SCT (allo-SCT) recipients free of GVHD and of immunosuppression after a period of 2 years, most of the physicians are reluctant to perform it and usually provide letters of contraindication to their pts in case of travel. To evaluate the safety and immunogenicity of YFV in allo-SCT recipients, we performed a retrospective analysis in 2 SCT units and 2 travel centers in France. Material (or patients) and methods: YFV was performed with the YF-17D, the only vaccine licensed in France. Serological analyses were performed using the plaque-reduction neutralization test (PRNT 80), the standard technique for assessing humoral neutralizing response to YF-17D immunization. A YFvirus neutralization test (YF-NT) of 10 U/L is accepted as a marker for clinical protection; 80 U/L is the highest value. All patients gave their inform consent for publication of the data. Results: 21 allo-SCT recipients (6 adults, 15 children) immunized with YFV after SCT were identified. None had active GVHD, and all but 1 were free of immunosuppression. The latter had 2 mg/d of prednisone. YFV was performed at a median time of 3.3 years (y) after SCT (from 1 to 22.5 y) and a median time of 2.4 y (0.4 to 20.5 y) after immunosuppression withdrawal. 8 of them had been immunized before SCT: 1 of the 5 pts evaluated for YF antibodies titer before YFV had a protective serum titer albeit at a low level (patient #16, YF-NT = 10 U/L). In the 16 evaluable pts, protective YF antibody titers were noted post-SCT immunization at a median time of 2.6 y after YFV (median YF-NT 80 U/L, IQR ). However, for pt #16, YF-NT did not increase after YFV. Disease, donor, stem cell source, conditioning and post SCT history had no impact on the level of protection. In 5 pts, persistence of a protective level after 5 years was documented. No adverse events were reported. In pt #16, as both donor and recipient had been immunized before SCT, it was not possible to identify the origin of the protective immunity after SCT. We identified 8 pts only vaccinated before SCT: 3 pts (donor/recipient pair immunized before SCT) retained protective titers after SCT (YF-NT: 20 to 80 UI/L), 4 pts (transplanted with related immunized or unrelated donors) lost their immunity after SCT, and 1 transplanted with an unrelated cord blood for refractory aplastic anemia retained protective immunity 2 y after SCT (YF-NT = 80 UI/L). Conclusion: YFV appears safe and efficient in allo-SCT recipients free of GVHD and immunosuppression. Although some pts were vaccinated earlier we recommend observing a period of 2 y after SCT. Immunization with YFV may be proposed in these pts in case of travel after evaluating their immune recovery. Donor or recipient YF immunity may persist after SCT. YF antibody titers should be evaluated before revaccination if there is a history of YFV to avoid unnecessary revaccination. Prospective evaluation of YF antibodies titers after SCT in recipients transplanted with immunized and non-immunized donors will help to understand the origin of the persistent protective immunity in some recipients. Disclosure of Interest: None declared. Introduction: The increase in BSI by Gram-with an adverse profile of antimicrobial susceptibility is regarded as a major threat to survival in hematologic pts, including HSCT recipients. Awareness of the local ecology is considered a prerequisite to plan possible interventions. This report aims at summarizing a single Centre experience on this field. Material (or patients) and methods: Over a seven years period, BSI data were collected in a series of 369 HSCT recipients (autologous 63%, allogeneic 37%, male 51%, median age 52 yrs). Coagulase negative Staphylococci (CNS) and Corynebacteria were considered as a cause of BSI only if the same strain was retrieved from two sets drawn from different veins. Since 2004, all pts received fluoroquinolone (FQ) prophylaxis and each febrile episode was treated with piperacillin/tazobactam (PTZ) and amikacin. Apart from considering the series as a whole, the observation time was also divided into two equal periods of 3.5 years/each in order to allow further univariate comparisons. Results: Among 369 HSCT pts, 256 BSI were recorded, 176 Gram+ and 80 Gram-. FQ resistance was observed in 127/176 (72%) Gram+ and 56/80 (70%) Gram-; only in 2012, Gramoutnumbered Gram+. Among Gram+, 98 were CNS, 21 Staphylococcus aureus and 15 Enterococci. 92/119 (77%) Staphylococci were methicillin-resistant and 1/15 Enterococci was vancomycin resistant. Among Gram-, Escherichia coli (29, esbl+ 7) was the most common isolate, followed by Stenotrophomonas maltophilia (11), Pseudomonas aeruginosa (10), Klebsiella pneumoniae (7, 0 esbl+), Acinetobacter baumanni (7). Of 80 Gram-, 9 were PTZ sensitive/ceftazidime (CFZ) resistant, 5 were PTZ resistant/CFZ sensitive; 13 were simultaneously resistant to PTZ and CFZ and 5 were carbapenem resistant (3 Acinetobacter baumanni and 2 Pseudomonas aeruginosa). Carbapenem resistant BSI were observed in pts colonized by the same strain before admission. More generally, 56/158 (35%) non-CNS BSI were preceded by colonization with the same strain. No microepidemic burst were observed. Over time, a significant increase ( p40.01) in Gram-BSI and a decrease in CNS ( p40.01) was observed; moreover, there was a non-significant trend toward an increase in PTZ resistance. Conclusion: These data summarize BSI etiologies over a prolonged period of time in which the same prophylaxis and empirical antibiotic therapy have been used, highlighting the following: 1. Despite a progressive increase in Gram-and a striking decrease in CNS etiologies, the majority of BSI are still due to Gram+; the high rate of FQ resistance among bacteria inducing BSI is not surprising, considering the prophylaxis failure. 2. Despite prolonged use, PTZ sensitive/CFZ resistant outnumber CFZ sensitive/PTZ resistant strains; the use of carbapenem in empirical therapy does not seem advisable. 3. All of carbapenem resistant strains and most of the esbl+ ones were acquired before admission in BMT Unit, with no spreading within the ward. 4. Despite a widespread rethinking of FQ prophylaxis, our data still favor this strategy and show the very long-term efficacy of a PTZ based empirical antibiotic therapy. Disclosure of Interest: None declared. Introduction: CMV reactivation is a frequent complication of allogeneic hematopoietic cell transplantation which can be negatively impactful to graft versus host disease and immunosuppression. Protective effects of CMV reactivation S205 on disease relapse have been reported for various diseases. However, it remains unclear what is the contribution of donor CMV serostatus. Material (or patients) and methods: Patients who received Allo-HCT between 2005-2012 at Moffitt Cancer Center were included in this retrospective analysis. Both recipient (R) and donor (D) CMV serostatus were ascertained prior to proceeding to Allo-HCT. Weekly CMV surveillance was performed by polymerase chain reaction (PCR) method and preemptive CMV therapy was implemented. CMV reactivation (CMV-ANY) was defined as any detectable CMV copies by PCR (40) detected in more than one instance. False positives were defined as single instance CMV reactivation of o250 copies with no subsequent CMV reactivation detected. CMV reactivation requiring therapy was defined as any instance where CMV reactivation exceeded 1000 copies by PCR (CMV-1000). Cord blood transplants were excluded from this cohort for subsequent analysis. Results: A total of 1066 patients were originally identified, but only 216 underwent an allogeneic HCT for lymphoid malignancies (table 1) Introduction: The purpose of this study is to explore the risk factors and outcome for Cytomegalovirus (CMV) reactivation and CMV refractory to antiviral chemotherapy after allogeneic haematopoietic stem cell transplantation (aHSCT). Material (or patients) and methods: CMV reactivation occurred in 282 of 685 (41.2%) patients treated with myeloablative conditioning regimen. Among the patients with CMV reactivation, 84 of 282 (29.8%) cases developed refractory CMV reactivation (RCR), and 37 of 282 (13.1%) cases progressed into CMV diseases. Results: Patients with RCR have a higher cumulative incidence of CMV diseases (26.2% versus 7.6%, P40.001). Seventy nine of 84 cases (94.0%) developed RCR before 100 days after aHSCT (5 cases with PP65 detection only). The copy number of CMV-DNA more than doubled compared to its initial baseline in 42 of 74 (56.8%) cases after 2 weeks of antiviral therapy, whereas 32 cases did not have a double increase above baseline. CMV disease developed in 15 of 42 (35.7%) and 3 of 32 (9.4%) (P = 0.011) cases in these two groups. The copy number of CMV-DNA increased at least 5 times above baseline in 24 of 74 (32.4%) patients, and among these 24 patients, 8 of 24 (33.3%) cases developed CMV disease. Among patients who do not have a 5-fold increase of CMV-DNA copy number, 10 of 50 (20.0%) patients developed CMV disease (P = 0.232). Both univariate and multivariate analysis demonstrated that the risk of CMV reactivation and RCR increased in patients from HLA-haploidentical donor and matched unrelated donor, or without using peripheral blood (PB) as stem cell source. The multivariate analysis revealed that patients who developed acute graft-versus-host disease (aGVHD) ≥ grade 2 have increased risk of developing CMV reactivation, whereas RCR occurred more frequently in those with total body irradiationcontaining regimens and a high dosage methylprednisolone (MP) for aGVHD treatment. The prevalence of RCR is three and six times higher in patients who received MP 1-2 mg/kg daily (P = 0.024) and ≥ 2 mg/kg daily(P = 0.001), with an odds ratio (OR) of 2.74 and 6.03 respectively. Conclusion: Hence, high dosage corticosteroids treatment is associated with incidence of RCR during the early phase after aHSCT. Disclosure of Interest: None declared. Usefulness of presepsin, procalcitonin and C-reactive protein as biomarkers in patients after hematopoietic stem cell transplantation I. Stoma 1,* , A. Uss 2 , I. Karpov 1 , N. Milanovich 2 , I. Iskrov 2 , I. Lendina 2 1 Belarusian state medical university, 2 City clinical hospital №9, Introduction: Bacterial bloodstream infections (BSI) remain a frequent complication in the preengrafment period after hematopoietic stem cell transplantation (HSCT) demonstrating high levels of mortality [1, 2] . Decision about the start and adequate choice of antimicrobial therapy in a patient with febrile neutropenia after HSCT is complicated because of the low frequency of culture isolation and reduced clinical manifestations of infection. Usefulness and choice of sepsis biomarker to distinguish BSI from other causes of febrile episode is still discussed in HSCT recipients in modern epidemiological situation. Objective of the study was to perform a comparative analysis of diagnostic values of presepsin, procalcitonin (PCT) and C-reactive protein (CRP) as sepsis biomarkers in adult patients after HSCT. Material (or patients) and methods: A prospective observational clinical study was performed at the center of hematology and bone marrow transplantation in Minsk, Republic of Belarus. Biomarkers (presepsin, PCT, CRP) were assessed on a first day of febrile neutropenia episode in adult patients after HSCT. Microbiologically confirmed BSI was set as a primary outcome. Results: Clinical and laboratory data was analyzed in 52 neutropenic patients after HSCT aged 18-79 years. Out of the biomarkers assessed best diagnostic value was shown in presepsin (AUC 0.889; 95% CI 0.644-0.987; P40.0001) with 75% sensitivity and 100% specificity, then in PCT (AUC 0.741; 95% CI 0.573-0.869; P = 0.0037) with 62% sensitivity and 88% specificity. Optimal cut-off value for C-reactive protein was set as 165 mg/l while it had an average diagnostic value (AUC 0.707; 95% CI 0.564-0.825; P = 0.0049) with low sensitivity (40%) and should not be routinely recommended as a biomarker in adult patients with suspected BSI after HSCT. Conclusion: Presepsin as a biomarker should be recommended in adult patients with suspected BSI after HSCT with a cut-off value of 218 pg/ml. PCT is inferior to presepsin in sensitivity and specificity, but still shows a good quality of diagnostic value with an optimal cut-off value of 1.5 ng/ml. CRP showed average diagnostic value with low sensitivity (40%) and should not be routinely recommended as a biomarker in adult patients with suspected BSI after HSCT. Introduction: Hemorrhagic cystitis (HC) is a well-known complication following allogeneic hematopoietic stem cell transplantation (allo-HSCT) that can be categorized as earlyonset or late-onset. Early-onset HC is usually caused by adenovirus or cytomegalovirus whereas late-onset HC mainly by polyomavirus BK. BU-containing myeloablative conditioning, unrelated donors and GVHD has been reported as risk factors increasing the chance of infection in bone marrowtransplant patients. Furthermore reactivation of human polyomavirus BK (BKV) may cause polyomavirus-associated nephropathy or polyoma virus-associated hemorrhagic cystitis.We present 15 patients with BK polyoma virus (BKV) ascociated hemorrhagic cystitis and 2 patients with BK polyoma virus associated hemorrhagic cystitis and nephritis. Material (or patients) and methods: Between 2013 and 2015, 90 patients received an allogeneic BMT at Acibadem Adana Hospital Pediatric Bone Marrow Transplantation Unit. 17 patients experienced BKV associated hemorrhagic cystitis and nephritis. BKV was detected in the urine analysis and blood by PCR (polymerase chain reaction) in all patients. Results: We presented 17 patients with BKV infection, age ranging from 3 to 20 with an average of 11.7 years. They underwent allo-HSCT due to thalassemia major (9 patients), aplastic anemia (3 patients) and leukemia (5 patients). The patients were treated with hydration, continuous bladder irrigation, ciprofloxacin, cidofovir and weekly intravesical hyaluronic acid instillation for four weeks. Ten patients showed complete resolution of hematuria. Three patients with refractory following therapy also received hyperbaric oxygen. Hemodialysis was performed in two patients who developed renal failure due to nephritis. Conclusion: Past exposures with the BK virus is widespread but significant consequences of infection are uncommon in the immunocompetent population. Reactivation of infection occurs under conditions of immunosuppression such as during GVHD treatment with patients who underwent HSCT. Early detection and treatment is crucial for successful management of BKV cystitis and nephritis. Neverthless even when treated with all the modalities, in some patients treatment faillure can be observed. Patterns of infection and infectious-related mortality after steroid-refractory acute graft versus host disease I. Garcia Cadenas 1,* , I. Rivera 2 , R. Martino 1 , A. Esquirol 1 , P. Barba 2 , S. Saavedra 1 , G. Orti 2 , J. Briones 1 , S. Brunet 1 , D. Valcarcel 2 , J. Sierra 1 1 Hematology, Hospital de Sant Pau. Universitat Autònoma de Barcelona, 2 Introduction: Studies focusing on the incidence, etiology and outcome of infectious episodes in patients with steroidrefractory acute GVHD (SR-GVHD) are scarce. Material (or patients) and methods: A total of 127 consecutive adult patients treated with inolimomab (77%) or etanercept (23%) due to 2-4 SR-GVHD were retrospectively analyzed. Results: ORR for salvage therapy was 42.5% on day +30 and 4year OS was 15.5% (95% CI 11.9-19.1%). Eighty-one severe bacterial infections were registered. Forty-four patients (34.6%) had more than one episode of bacteraemia. Forty-six of them (36.2%) were due to coagulase-negative staphylococci and 36 (28.3%) were associated with multiresistant enterobacteria. The rate of CMV reactivation was 58.1% and 44 patients (34.6%) had recurrent episodes. Death due to CMV disease occurred in 4 patients. Four out of 79 patients monitored for EBV had a reactivation (5.1%) but only one progressed to PTLD. Nineteen patients (15%) developed a concomitant BK virus-associated hemorrhagic cystitis. Seventeen patients (13.4%) were diagnosed with an IFI at a median of 28 days after SR-GVHD diagnosis. Eleven out of the 43 (26%) long term survivors had pneumonia by community respiratory virus and/ or encapsulated bacteria, in most of the cases in association with chronic GVHD. Infectious complications were a significant cause of NRM with a 4-year CI of 53.9% (95% CI 47.4-60.4) associated or not with other complications. Use of inolimomab (HR 1.7, p 0.03), primary GVHD prophylaxis other than Sirolimus-Tacrolimus (HR 1.9, p 0.04), use of Rituximab in the 6 months prior to SCT (HR 1.9, p 0.03), severe infection before SR-GVHD diagnosis (HR 1.8, p 0.01) and a C reactive protein greater than 15 mg/L before first salvage treatment (HR 1.6, p 0.01) were significantly associated with infectious related mortality in multivariate analysis. Conclusion: This study reflects that opportunistic infections have a significant impact in the dismal long-term outcomes of patients with SR-GVHD. An additional important observation is the high rate of bacteraemia by multiresistant GNB. An improved understanding of the risk of severe infections in this setting may lead to the implementation of more effective prophylactic, monitoring and therapeutic strategies, although only increasing the response of SR-GVHD and the concomitant improvement of innate and pan-specific immune responses will lead to long-term GVHD and severe infection-free survival. Disclosure of Interest: None declared. preceding infections and their treatment on incidence of EBV infection in pediatric allo-HSCT recipients over a period of 24 months. Material (or patients) and methods: During analyzed period a total number of 232 allo-HSCTs (including 67 ALL, 47 AML, 28 NHL/HD, 60 solid tumors, 39 PID, 38 BMF, 21 other) were performed in all Polish pediatric HSCT centers (iPhot-13 project). The episodes of following viral infections were reported: EBV, CMV, ADV, BKV, HHV-6 and VZV. Preemptive anti-EBV therapy with rituximab was initiated after significant EBV-DNA-emia had been diagnosed. Results: The total number of episodes of viral infections was 173 in 119 patients. The overall incidence of analyzed 6 types of viral infections in allo-HSCT patients was 51.3% (119/232), and in 51.2% of them (61/119) multiple infections were diagnosed. Cumulative incidence of viral infections in allo-HSCT was: 28.0% (65/232) for CMV, 18.5% (43/232) for BKV, 15.5% (36/232) for EBV, 9.5% (22/232) for ADV, 2.6% (6/232) for VZV, and 0.9% (2/232) for HHV6. Out of 36 EBV infections, in 12 (33%) cases, the infection was preceded by other viral infection, including CMV (n = 6), BKV (n = 2), VZV (n = 1) or multiple infections (ADV+BKV ± CMV, n = 3) within 2 months (median: 30 days, range 7-50 days) before the diagnosis of EBV-DNA-emia. These infections had been treated with antivirals: GCV (n = 6), CDV (± GCV ± FSV, n = 5), or ACV (n = 1). In 8/12 (67%) cases, development of significant EBV-DNA-emia occurred during GVHD (4 acute, 4 chronic). With respect to the number of episodes of viral infections, the rate of significant EBV-DNA-emia within 60 days after the beginning of antiviral treatment was following: 8.8% (8/91) for CMV, 11% (4/36) for BKV, 0/14 for ADV, 1/7 for VZV, and 0/3 for HHV-6 infection. With respect to antiviral drugs, significant EBV-DNA-emia has developed in 8/76 (10.5%) GCV therapies; 5/61 (8.1%) CDV therapies, 2/42 (4.9%) FSV therapies, and 1/9 (11%) of ACV therapies. Conclusion: The incidence of EBV infection requiring preemptive treatment with rituximab in allo-HSCT children was 15.5% (36/232). In 33% (12/36) cases of EBV infection, these episodes were preceded by other viral infection treated with antivirals, which did not prevent development of significant EBV-DNA-emia. In 67% (8/12) of these cases, GVHD was a factor enabling development of significant EBV-DNA-emia during antiviral therapy of other infection. Disclosure of Interest: None declared. The impact of gut colonization by antibiotic-resistant bacteria on the outcomes of allogeneic hematopoietic stem-cell transplantation: a retrospective, single-center study J. Biliński 1,* , K. Robak 1 , Z. Introduction: In patients who have undergone allogeneic stem-cell transplantation (alloSCT), systemic infections with multidrug-resistant (MDR) bacteria are associated with a mortality rate of 36-95% [1] [2] [3] . The incidence of infection by MDR bacteria is rising because of selection pressure caused by increasing use of broad-spectrum antibiotics [4] . MDR bacteria can colonize niches in the gastrointestinal tract, and in specific circumstances may cause life-threatening systemic infections [4] [5] . There is also a growing evidence that the gut microbiome exerts immunomodulatory effects during alloSCT and can trigger autoimmune response or promote bacterial infections [6] [7] . To our knowledge, no study has addressed the epidemiology of gut colonization by MDR bacteria and its impact on the outcomes of alloSCT. Therefore,we decided to answer this scientific question. Material (or patients) and methods: We retrospectively analyzed data on 115 alloSCTs performed at a single transplant center. Results: Pretransplant microbiology screening identified colonization in 30% of cases. Colonization had a negative impact on overall survival (OS) after alloSCT in univariate (34 versus 74% at 24 months, P = 0.0002) and multivariate (hazard ratio 3.08, 95% confidence interval 1.6-5.94; P = 0.0007) analyses (Fig 1) . Non-relapse mortality (NRM) was significantly higher in colonized patients than in noncolonized patients (38% versus 10% at 24 months, P = 0.001). Colonized patients more frequently experienced bacteremia (40% versus 21%, P = 0.049), and more deaths were attributable to infectious causes in the colonized group (65 versus 28%, P = 0.041). We observed a significantly higher incidence of grade 2-4 acute graft-versus-host disease (aGvHD) in colonized patients than in noncolonized patients (43 versus 22.5%, P = 0.049), especially involving the gastrointestinal system (31% versus 15%, P = 0.070). Conclusion: We determined that gut colonization by MDR bacteria decreases the OS of patients undergoing alloSCT by increasing NRM and the incidences of systemic infection and aGvHD. The gut colonization may serve as poor microbiome diversity marker, but bigger studies are needed to investigate this observations, which may have a great impact on alloSCT procedure. Introduction: Polyomavirus infection is highly prevalent in humans. The most known human polyomaviruses are BK virus (BKV) and JC virus (JCV). Primary infection occurs during childhood, remaining latent in the uroepithelium afterwards. In HSCT recipients, polyomaviruses shedding and its association with several disorders has been described while the possible impact of age has not been established. Material (or patients) and methods: We conducted a prospective monitorization of urinary BK and JC shedding by qPCR in 968 urine samples from 128 patients who underwent 135 allo-HSCT between January 2010 and December 2014. Samples were collected at admission for transplant and weekly until discharge. A comparison of polyomaviruses shedding between a pediatric and an adult cohort was established. Twenty out of 135 patients (14.8%) were children with a median age of 8 years (range: 1-18) and the remaining 115 (85.2%) were adults with a median age of 52 years (range: 19-68). In the pediatric group, there was a predominance of males (75%) compare with 53.9% in the adult group. The most prevalent baseline disease in both groups was acute leukemia (65% in pediatrics versus 34.8% in adults). The majority of pediatrics (70%) underwent matched unrelated-donor HSCT compared to 50.4% of adults. Results: BK viruria was demonstrated in 15 (75%) and 70 (60.9%), pediatric and adult patients respectively. No differences were seen between both groups regarding median time to onset (day+6, range: -1 to +13 in pediatrics vs day+5.5, range: -12 to +167 in adults). Median duration of BKV excretion was 66 days (range: 4-128) in children and 20 days (range: 1-554) in adults. No differences at first (log10 4.3 vs 4.4 copies/ mL) or maximum (log10 7.1 vs 6.9 copies/mL) BKV load detection were found. JC viruria was seen in 2 (10%) pediatric patients while 54 out of 115 (47%) adults presented urinary JC shedding. A significant difference between both groups was found (P = 0.002). Median time for first JCV detection was day +99.5, range: 17-182 vs -6.5, range -12 to +347 in pediatrics and adults, respectively (P = 0.032). Median duration of JCV excretion was 55.5 days (range: 1-111) in children compared to 27.5 days (range: 1-535) in adults. No differences at first (log10 3 vs 4.6 copies/mL) or maximum JCV (log10 5.6 copies/mL for both groups) load detection were found. Conclusion: 1) BK shedding in urine was frequent in our series of allo-HSCT recipients. 2) JC shedding in urine was frequent in adults and rare in pediatric population. 3) A clear difference in time of onset of JC shedding was found between both groups. 4) Both polyomaviruses excretion seems to last longer in children. 5) Further analyses need to be conducted to establish the cause and meaning of these differences and the possible role in the development of cystitis and other disorders in HSCT recipients. Disclosure of Interest: None declared. Introduction: Haematological malignancies (HM) continue to be a leading cause of death in HIV-positive patients (pts) in the HAART era. We describe our single institution experience in Autologous Stem Cell Transplantation (ASCT) in HIV-1 infected patients with high-risk or relapsed HM, focusing on 1st-year infectious complications and long-term outcome. Material (or patients) and methods: We retrospectively reviewed 28 ASCT procedures performed between 2000-2015. Patient characteristics are described in table 1. BEAM conditioning regimen was the most commonly used. All pts were on ART. Antimicrobial prophylaxis consisted of a quinolone (received by 63% of pts), antifungals (fluconazol or micafungin, 66%), acyclovir (100%) and cotrimoxazole (100%). All transplants were completed in a tertiary hospital JACIE-accredited SCT unit. Results: A median count of 3,3x10e6/kg PBPCs were infused (r: 1, 5) . Mucositis was the most common toxicity (stomatitis 71% of pts; GI 42%), but it was graded 3-4 in only 3 cases. G-CSF support was routinely used, from day+7 until engraftment. Median days (d) to neutrophil engraftment was 14 (r: 10-33). Median admission length was 32d (r: 22-90). Reconstitution: Median CD4+ Pre-T: 180 cells/uL (r: 8-702), day 100: 171 (34-372), 1 year: 276 (200-721); CD8+ Pre-T: 578 (r: 78-1960), day 100: 1141 (108-2106), 1 year: 1092 (504-2400). Febrile neutropenia was universal. The pre-engraftment (pr-e) period was characterized by bacterial infections (i.e: 2 Streptococcus sp bacteremias, several CNS and GN rods CVC-related bacteremias, 4 Clostridium difficile infections, 1 UTI due to Enterococcus sp and 2 fatal septic episodes, one possibly due to Stenotrophomonas maltophilia and the other by Rothia mucilaginosa). Fungal infections were uncommon and centered in the pr-e phase (thrush, 2 oesophageal candidiasis, 1 probable IFD [pulmonary Aspergillosis] and a non-viable filamentous mold was isolated from the post-mortem spleen tissue culture after a septic death). Broad-spectrum antifungal therapy was started in 54% of pts (empirical/preemptive). Viral infections were common and usually delayed (pr-e: 3 CMV reactivations. o100d: 9 CMV, 1 EBV and 2 VZV reactivations. 4100d: 3 CMV reactivations, 1 CMV disease-retinal necrosis and 3 episodes of shingles). ART was discontinued during the early post-transplant (pos-T) period in 25% of pts. HIV viral load bleeps were seen in 35% of pts (mainly at 1st month) and 1 pos-T virological failure was diagnosed. One Strongyloidiasis case was detected in the 4100d period. Median follow-up: 74 months (r: 23-183). Five pts died during the first year pos-T (2 infection-related deaths and 3 relapses). One-year overall survival (OS) was 81%. Three more pts died during follow-up: 1 due to a septic shock 2 years pos-T, one of a MDS diagnosed 4 years pos-T and the other patient died of lung cancer. 5-year OS was 77% (CI 95%: 62-93). Currently 15 pts of the cohort (55%) are still alive, 5 have relapsed (4 of whom died) and 5 have been lost to follow-up. Six pts have developed secondary tumors. Conclusion: ASCT has been proven as a feasible and effective therapy for HIV pts with HM. In our experience, infectious complications are not scarce, but long-term survival is achievable and considering the increasing life expectancy of HIV population, it should be a mandatory approach when otherwise indicated. Disclosure of Interest: None declared. Introduction: Neurological complications (NC) are commonly seen after allogeneic stem cell transplantation (SCT) but little is known about the impact of the stem cell source on the incidence, characteristics and outcome of NC in patients undergoing SCT. Material (or patients) and methods: All 709 consecutive adults receiving an HLA-matched sibling transplantation (MST) (343 patients) or umbilical cord blood transplantation (UCBT) (366 patients) between January 2000 and May 2014 were included. NC were defined as any neurological event that occurred after the start of the conditioning regimen and before relapse or progression. NC were classified in different categories according to the final diagnosis, including central and peripheral nervous system involvement. Results: Median age was 38 years (range, 13-65) and 44 years (range, 14-67) in patients receiving UCBT and MST, respectively (Po 0.001). Most patients had AML/SMD (49%) or ALL (23%). The conditioning regimen was myeloablative (MAC) in 494 patients (67%), and was more frequent in the UCBT cohort S210 (78% versus 62% in MST, P40.001). Median follow up of surviving patients was 60.6 months (range, 6.7-170.5). Overall, 166 NC events were documented in 132 patients and are described in Table 1 according to the type of transplant. The most frequent events were encephalopathy (24%), CNS infection (20.5%) and neuropathy (14%). The main neurological complication in the MST group was the encephalopathy (25%) and in the UBCT group it was the CNS infection (27%). The median time to NC was 51 days (range, -4-2270). Cumulative incidence risk of developing at least 1 episode of NC at 5 years was 18%, being 25% and 13% after UCBT and MST, respectively ( p40.001). The 5-y CI of CNS infection was 4.8%, 1.5% in MST and 8.3% in UCBT ( p40.001). 35% of all the infections were fungal. The 5-y CI of encephalopathy was 4.1% in MST and 7.3% in UCBT (P = 0.07). In multivariable analysis type of SCT (HR 2.3, 95% confidence interval 1.6-3.3, P40.001) and patient´s age (HR 1.9, 95% confidence interval 1.4-2.8, P40.001) were independently associated with the risk of developing NC. Overall survival at 5 years was 16% in patients who developed NC (16%) and 43% for those who did not ( p40.001). Despite the advantages in contemporrary standards for stem cell mobilization, quantification of autologous grafts by CD34+ enumeration and implementation of standards for improvement of posttransplant supportive care, the risk of early posttransplant relapse and/or death still remains with theis treatment option. Material (or patients) and methods: the initially designed as retrospective-prospective included 263 patients with hematological malignancies treated with autologous transplantation from 2000-2015. All patiets were assesed before transplant for comorbidities. The aim of the study was development and validation of prognostic model for treatment with autologous transplantation in patients with hematological malignancies in which both comorbidity assessment and factors affecting stem cell mobilization, cryopreservation and quality of autologous graft, will promote a valide algorythm in risk startification of mortality and early relapse 1 to 3 years aftertransplant. Results: analysis of stem cell mobilization showed ( p40.0001) for higher number of days with aplasia and apheresis, lower number of WBC in peripheral blood before the start of mobilization ( 3 Platelet refractoriness (PR) is the repeated failure to obtain satisfactory responses to platelet transfusions. Blood products prepared with standard techniques carry leukocytes along with red blood cells and platelets. As a result of contamination these leukocytes can produce a wide variety of adverse effects including PR, transmission of infections, immunosuppression, non-hemolytic transfusion reactions and Graft Versus Host Disease. Use of leukoreduced blood for the prevention of PR is a grey area in Pakistan due to its cost and absence of relevant studies AIMS: Determine the development of alloantibodies in SCD patients with multiple transfusions and analyze the alloimmunization rate after the transfusion of blood with and without leukoreduction. Material (or patients) and methods: 21 previously transfused (A) and 18 newly diagnosed(B) patients were recruited from outpatient department, Civil Hospital Karachi. Lyari (suburb area) has incidence higher comparatively due to their specific ethnicity. We provided the leukoreduced blood to the new patients. Blood samples were collected from patients after a year (425 transfusions each) and kept at -20 0 C until use. Detection of platelet specific IgG antibodies were performed with solid phase red cell adherence assay (SPRCA (Table I) . Of additional clinical importance, CRP was also independently associated with NRM and OS in patients with a low risk HCT-CI of 0-2. Table I . Multivariate Cox regression model of co-variates. The hazard ratio (HR) and 95% confidence intervals (CI) are presented. A CRP score was constructed, with each independent variable given a score of 1: CRP ≥10 mg/l, HCT-CI ≥ 3, late S213 EBMT disease stage, and recipient CMV IgG sero-positivity. The maximum score for NRM is 4, and that for OS is 3. Conclusion: We report for the first time that an elevated CRP level of ≥ 10 mg/l prior to MAC allogeneic stem cell transplant is independently associated with NRM at 100 days and OS at five years, even in patients identified as low risk by the HCT-CI and EBMT score. We confirm that CRP is a robust biomarker that should be integrated into existing risk assessment tools when considering patients for myeloablative conditioned allogeneic stem cell transplantation. Disclosure of Interest: None declared. Inability to return to work and need for work disability pension among long term survivors of HSCT A. Tichelli 1,* , S. Gerull 1 , A. Holbro 2 , A. Buser 2 , J. Halter 1 , J. Passweg 1 1 Hematology, 2 Transfusion center, University Hospital Basel, Basel, Switzerland Introduction: Return to work or equivalent belongs to one of the critical goals of HSCT. However, physical and mental late effects may impede return to normal activity after HSCT. In case of inability to work, patients may need work disability pension to assure a reasonable livelihood. The aim of this study is to evaluate inability to work and need for work disability pension among long-term survivors of HSCT and to analyze possible determinants for need of social support. Material (or patients) and methods: Retrospective, single center cross-sectional study including all HSCT patients surviving ≥ 5 years, and seen at the outpatient clinic between January 2013-August 2015. The prevalence of the occupational status (return to work, inability to work, work disability pension), and the reasons for need of disability pension were evaluated. Results: 250 long-term survivors were seen during the study period. Four were excluded because of a lack of data on occupational status, 43 because they were retired at studytime. Of the 203 evaluable patients median age at HSCT was 36y, and at time of the study control 50y; time interval between HSCT and study control was 12y; 116 patients were males, 177 had allo-HSCT, 187 a malignant disease. At time of the study, 156 (77%) had an occupational status, 47 (23%) were unemployed. From the 156 patients who returned to work, 78 had full-time job, 26 reduced-time work by personal choice, 44 reduced-time work for medical reasons, 8 were housewives. According to the national statistics, among all insured individuals aged between 18 and 65 years, 3.4 to 4.0% are annuitants of work disability pension. 76/203 (37%) longterm HSCT-survivors obtained work disability pension. None of these 76 survivors had full-time job: 33 had returned to parttime employment, 41 were not working, 2 did not find a job. Among the 127 patients without work disability pension, 107 (84%) were full-time working, 16 (13%) had part-time, and 4 (3%) no employment ( p40.001). Patients with disability pension were significantly older at time of HSCT and at study time, had a shorter interval since HSCT, were more often living alone (not married, no partner), had more often Karnofsky score o70% and chronic GVHD of any degree, had more physical and mental active late effects, and higher score of fatigue, compared to the patients without disability pension. In contrast, gender, type of HSCT, primary disease, having children or not, and smoking did not differ between both groups. Factors associated with need of work disability pension Conclusion: 37% of long-term survivors need social financial support to manage day-to-day life. Older age at HSCT and at study time, living alone, poor physical or mental health condition and fatigue are possible determinants predicting need of social and financial support even long time after HSCT. These findings underline once more the importance of screening for chronic health condition and undertaking preventive measures in long-term survivors after HSCT. Disclosure of Interest: None declared. Reduced intensity conditioning and male gender are associated with increased cardiac-related mortality in patients undergoing allogeneic stem cell transplantation A. Natale 1,* , S. Santarone 1 , P. Olioso 1 , G. Papalinetti 1 , S. Angelini 1 , P. Di Bartolomeo 1 1 Hematology, BMT Center, Ospedale Civile, Pescara, Italy Introduction: Transplant-related mortality (TRM) is still a major concern in patients receiving allogeneic SCT. Little is known about the impact of cardiac mortality on posttransplant outcome. The aim of this study was to assess the incidence and risk factors of cardiac-related mortality ( Introduction: Allogeneic hematopoietic stem cell transplantation (Allo-TPH) is a curative treatment in many hematologic diseases, but it is associated with late effects which can decrease the Quality of life (QoL). As the number of long-term survivors after Allo-TPH increases progressively, QoL becomes a major concern. Aims: To assess health-related QoL in long-term survivors (at least 5 years) after Allo-TPH and identify QoL related-factors. Material (or patients) and methods: In our centre 307 patients underwent Allo-TPH between the years 2000 and 2010: 158 (51,5%) were alive at the time of the study. We evaluated 102 patients in complete remission, and older than 18 years at enrollment, in a cross-sectional study; 56 were excluded (52 lost for contact, 4 relapsed). Patients' characteristics are shown in Table 1 Results: EQoL-5D-5 L analysis: Median VAS score was 80 (range 30-100). There were no statistical differences in QoL-INDEX*100 (all groups of age) between transplanted patients and the general population. FACT-BMT analysis: Total Score mean was 121 (44-147). Mean Score in specific areas was: PWB 23 (2-28), SWB 24 (11-28), EWB 20 (2-24), FWB 22 and BMTS 32 (16-40). Age at Allo-TPH (18-44 vs 45-80 years) was statistically associated with PWB score (26 vs 24 P = 0,002) and FWB score (25 vs 22 P = 0,046) but no differences were found in EWB, SWB, BMTS and total score. Regarding gender, women had better FWB score (25 vs 22 P = 0,044). The presence of chronic GvHD was related with lower PWB score (24 vs 26 P = 0,004). Use of chronic drugs was associated with lower QoL in every area (total score 135 vs 125 P = 0,002). We did not find statistical differences associated with factors like conditioning regimen, donor type, progenitor source or use of ATG. Regarding working activity, 53 patients (52%) were employed, 2 (2%) unemployed, 24 (23,5%) retired and 23 ( Introduction: Hepatic veno-occlusive disease (VOD) is a common and serious complication of hemotopoietic stem cell transplantation (HSCT) in children. We aimed to assess prospectively the use of prophylactic defibrotide in pediatric patients undergoing HSCT. Material (or patients) and methods: In this study, 76 patients who underwent HSCT were given defibrotide prophylaxis as 25 mg/kg per day in four divided intravenous infusions over 2 h, starting on the same day as the pretransplantation conditioning regimen. The mean duration of use of defibrotide is 20 days as a prophylaxis. Results: In this study, 76 patients were recruited, 53 male patients and 23 female patients, with the average of 9.3 years, range 1-20; 4% infants, 55% children and 41% adolescent. There were 33 patients with thalassemia major, 30 patients with leukemia, 7 patients with aplastic anemia, one patient with Diamond Blackfan anemia, two patients with congenitale dyserythropoetic anemia, one patient with osteopetrosis, one patient with famial hemophagocytic lymphohistiocytosis, and one patient with Kostman syndrome. All transplants were allogeneic. No serious side effects were seen. In eight patients developed clinical VOD (Seattle criteria). In these patients, defibrotide dose was increased to a treatment dose of 40-60 mg/kg per day. One infant patient with Kostman syndrome and one patient with aplastic anemia died due to hepatic and pulmonary venoocclusive disease. After 24 months of follow up, 6 patients who developed VOD are being well and no patient have transplant related complications. Conclusion: Hepatic veno-occlusive disease, which is caused by hepatocyte and sinusoidal vessel endothelium damage, can ocur early after HSCT, and in its severe form, may lead tol iver faillure, hepatorenal syndrom, portal hypertension, and eventually death from multiorgan faillure. In this prospective study, we demonstrated that the use of defibrotide is safe and effective in preventing and treating VOD in pediatric patients at high risk. Introduction: Recent studies have reported an improved outcome over the years for children, adolescents and young adults (CAYA) undergoing hematopoietic stem cell transplantation (HSCT) and then admitted to Pediatric Intensive care Unit (PICU). This is due to the advances in the transplant procedures and in the supports delivered in the PICU (e.g. non invasive ventilation). 1,2 Nevertheless, the rate of mortality is still high, especially when invasive ventilation (IV) and continuous renal replacement therapy (CRRT) are needed. 3 The main purpose of this retrospective study was to describe incidence, causes and outcome of PICU admissions of CAYA who underwent HSCT in HSCT Unit of Padova. Furthermore, we analyzed risk factors predisposing to PICU admission and factors associated with PICU mortality.. Material (or patients) and methods: We evaluated data of 512 CAYA (0-23 years) who received HSCT in our HSCT Unit from January 1998 to April 2015. Results: 3-year cumulative incidence of PICU admission was 13.5% (95% CI: 10.5%416.5%). The main causes of PICU admission were: respiratory failure (36%), failure of different organs and apparatus (16%) and sepsis (13%). 57.1% of patients died in the PICU. The overall 90-day probability of survival after PICU admission was 34.3% (95% CI 24. 8-47.4) . Introduction: Severe combined immunodeficiency (SCID), is a group of rare inherited disorders characterized by complete absence of T cell-mediated immunity and by impaired B cell function. The overall incidence is about 1:50,000-1:100,000 newborns. There is molecular heterogeneity among the patients, but the clinical presentation of patients are similar. Material (or patients) and methods: Here we plan to compare the HSCT characteristics of SCID patients with Artemis defect (T-B-NK+) (group 1, n = 8) and JAK3 defect (T-B-NK+) (group 2, n = 8) who had HLA full matched stem cell transplanted without conditioning regimen. The median age at the beginning of the symptoms is 1.5 in the first group and 4.5 months in the second group. Median age at diagnosis is 6 and 4 respectively in groups. Median age at transplantation is 5.5 in Group 1 and 7 months in Group 2. Parental consanguinity was found 100% in both groups. BCG vaccination was performed in 6 of 8 patients in both groups. BCG dissemination occurred in 1/6 (16.6%) and 0/6 (0%) of BCG vaccinated patients in group 1 and 2 respectively. HSCT source was bone marrow in 7/8 (87.5%) and 4/8 (50%) of patients in first and second group respectively. The source in the remaining patients (1/8 (12.5% ) and 4/8 (50%) respectively in group 1 and 2) was peripheral blood. Acute and chronic GVHD was seen respectively in 3/8 (37.5%) and 1(12.5%) patients in group 1, 5/8 (62.5%) and 3 (37.5%) patients in group 2. Death ratio was 1/8 (12.5%) and 2/8 (25%) in groups 1 and 2 respectively. IVIG replacement need is 2/7 (28.6%) of survived and 0/6 (0%) of survived patients in Group 1 and 2 respectively. Conclusion: Although there was no statistical difference between the groups, the GVHD ratio was found to be higher in group 2, and IVIG replacement need was more in group 1. Introduction: Severe combined immunodeficiency (SCID), is a group of rare inherited disorders characterized by complete absence of T cell-mediated immunity and by impaired B cell function. The overall incidence is about 1:50,000-1:100,000 newborns. There is molecular heterogeneity among the patients, but the clinical presentation of patients are similar. Material (or patients) and methods: Here we plan to compare the HSCT characteristics of SCID patients with RAG1/2 defect (T-B-NK+) (group 1, n = 8) and Artemis defect (T-B-NK+) (group 2, n = 8) who had HLA full matched stem cell transplanted without conditioning regimen. The median age at the beginning of the symptoms is 1 in the first group and 1.5 months in the second group. Median age at diagnosis is 7 and 6 respectively in groups. Median age at transplantation is 6 in Group 1 and 5.5 months in Group 2. Parental consanguinity was found 87.5% in first, 100% in second group. BCG vaccination was performed in 6 of 8 patients in both groups. BCG dissemination occurred in 2/6 (33.3%) and 1/6 (16.6%) of BCG vaccinated patients in group 1 and 2 respectively. HSCT source was bone marrow in 8/8 (100%) and 7/8 (87.5%) of patients in first and second group respectively. The source in the remaining patients (1/8 (12.5%) in group 1) was peripheral blood. Acute and chronic GVHD was seen respectively in 4/8 (50%) and 3/8 (37.5%) patients in group 1, 3/8 (37.5%) and 1(12.5%) patients in group 2. Survival is 100% in Group1 and 87.5% in Group 2. IVIG replacement need is of 1/8 (12.5%) and 2/7 (28.6%) patients in Group 1 and 2 respectively. Conclusion: Although there was no statistical difference between the groups, the GVHD ratio was found to be higher in group 1. We could not find any reason of this difference. It may derive from other unknown patient/donor characteristics Introduction: Severe combined immunodeficiency (SCID), is a group of rare inherited disorders characterized by complete absence of T cell-mediated immunity and by impaired B cell function. The overall incidence is about 1:50,000-1:100,000 newborns. There are many factors which have effect on development of GVHD, such as HLA incompatibility. In a few studies, ABO blood group incompatibility was shown to have no effect on the development of GVHD. Material (or patients) and methods: Here we plan to compare the GVHD characteristics of SCID patients according to ABO and Rh compatibility. For this purpose, in order to compare homogeneous groups, we have chosen from our SCID patient series the patients transplanted from HLA fullmatched donor by not using conditioning regimen. There was 29 patients fullfilling these criteria. Results: We first seperated patients into 3 groups; transplanted from ABO/Rh compatible donors (Group 1, n = 13), from ABO incompatible donors (Group 2, n = 13), from Rh incompatible donors (n = 3). The acute and chronic GVHD ratio was 9/13 (69.2%) and 5/12(41.6%) (1 lost) in Group 1; 6/13 (46.1%) and 2/11 (18.1%) (2 lost) in Group 2 respectively. Out of 13 patients in group 2, 7 of them had major (Group 2a), 6 had minor (Group 2b) ABO incompatibility with their donors. Acute and chronic GVHD ratio was 2/7 (28.6%) and 1/5 (20%) (5/7 survived) in Group 2a, 4/6 (66.6%) and 1/6 (16.6%) in Group 2b. We further analysed the group 1 and 2 according to the gender (Group1x, patients having donor with same gender and ABO compatible; Group1y, donor with different gender and ABO compatible; Group2x, donor with same gender and ABO incompatible; Group 2y, donor with different gender and ABO incompatible). Acute and chronic GVHD was 3/6 (50%) and 2/6 (33.3%) in Group 1x; 6/7 (85.7%) and 4/7 (57.1%) in Group1y; 1/5 (20%) and 0/5(0%) in Group 2x; 5/8 (62.5%) and 2/7 (28.6%) (1 lost) in Group 2y. Conclusion: Although the number of patients in groups were low and there are several factors affecting GVHD, according to the results, we think that the ABO compatibility, major or minor, has no effect on GVHD; but gender difference has effect on acute and chronic GVHD risk. (0) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) . Pts received either chemotherapy (72%) or total body irradiation (TBI, 28%) based MAC. Busulfan, fludarabine, and melphalan with rabbit ATG was the most commonly used conditioning regimen (66%). Donors were matched related (MRD), matched unrelated (MUD) and mismatched unrelated (MMUD) in 38%, 39%, and 23% of pts, respectively. At 1 yr, non-relapse mortality was 17% (Figure 1 ). The rate of grade 2-4 acute GVHD was 11.5%. The toxicities observed in the first yr are shown in There was no increase in grade 4/5 toxicities with administration of TBI based MAC compared to chemotherapy alone. For pts with≥9 toxicities in the first 100 days, the overall survival (OS) was 72% compared to 87% for pts witho 9 toxicities (P = 0.012 Figure 1 ). Results: The one year cumulative incidence of haemodialysis was estimated 18.3%. There was a significant difference in the median time to initiate hemodialysis in patients over 55 years old and younger ones: 121 versus 63 days (P = 0.05). In univariate analysis, we found that exposure to polymyxin B (p 0.001) and amikacin (p 0.026) were associated with haemodialysis. The multivariate analysis revealed that only septic shock (HR: 3.5 P o0.001) and exposure to cidofovir (HR:4.0 P = 0.011) significantly increased the risk for renal replacement therapy. Also, the presence of acute GVHD tended (HR:1.5 P = 0.067) to be a risk for this outcome. The need of renal replacement therapy had significantly increased the mortality risk in 6.1-fold (95% CI: 3.6-11; p 0.0001). Conclusion: Judicious use of cidofovir and interventions to optimize immune reconstitution are measures that can reduce the incidence of dialysis. The peak of kidney toxicity in the elderly occurs later, probably due to the lower intensity of conditioning regimens. The occurrence of acute GVHD increases vulnerability to infections and prolongs immunossupressive drugs exposure, possibly raising the risks of sepsis and dialysis. The need for hemodialysis has a strong impact on allo-HSCT mortality. We could not assess with confidence the impact of chronic kidney disease pre allo-HSCT due to a low proportion this condition in our patients at the baseline. Introduction: Gastrointestinal mucositis (GIM) is one of the most frequent post-transplant complications seen in 75-99% auto-and allo-HSCT recipients (Vagliano L et al, 2011) . It is associated with pain of varying intensity depending on GIM grade. The GIM extension also varies, most often it involves mouth isthmus, lips cheeks and oesophagus mucosa. Patient-controlled analgesia (PCA) provides flexible analgesics dosing schedule and sense of involvement for the patient. It is a standard in adult settings and, according to European Pain Federation (EFIC) guidelines, it is a feasible option in pediatric patients. Topical analgesics use in children is difficult due to the lack of compliance, while although opiates provide good analgesia there safety profile in children is often unsatisfactory. Tramadol is a centrally acting synthetic analgesic. Its active metabolite (mono-O-dimethiltramadol) has relatively prolonged effect (half-life about 6 hours) and demonstrates both opioid and nonopioid mechanism of action. Although it is currently not a standard in pediatric population due to the lack of solid evidence, according to WHO recommendations it may be an additional option for patients without response to nonsteroidal anti-inflammatory drugs. While tramadol-based PCA is widely used in post-surgery patients, there are currently no publications on its use in pediatric population. Tramadol is a possible option in patients with moderate pain. It also safer in patients with GI tract involvement due to less expressed bowel movement suppression. It also may be a feasible option in pediatric population. Material (or patients) and methods: A total of 44 pediatric HSCT recipients (median age 5.6, range 3-10 years) with stage 2-3 GIM. All patients received myeloablative conditioning regimens (23 auto-HSCTs, 21 allo-HSCT). A median neutropenia period duration was 12 (range 7-20) days. Patients required systemic analgesics use for a median of 5 (range 3-10) days. Pain severity was assessed using Facescale and Visual Analogue Scale. All patients received tramadol-based PCA as initial systemic analgesic. The load dose was 0.5 mg/kg, baseline 0.25 mg/kg, bolus 0.25 mg/kg, lockout 15 min. The dose was aimed at achieving target pain score values (pain intensity to cope with). While older patients determined the target vales themselves, in younger children the default values of o 4 was used. Results: The median initial pain score value was 6 (range 4-10) and a median of pain-associated nightly wakes was 2 (range 1-3). The median target pain score value was 4 (range 2-6). Tramadol-based PCA allowed decreasing pain score value to a median of 3 (range 0-4), most of the patients slept well during the night. Only 1 (2.3%) patient required switching to morphine-based analgesia (next step of WHO "analgesic ladder") due to inadequate pain control on mucositis progression. There were few therapy-associated complications. All complications were short-term: drowsiness during the first day in 4 (9%) patients, dizziness in 2 (4.5%) patients. Introduction: The role of allogeneic hematopoietic stem cell transplantation (allo-HSCT) in MM is controversial due to the high morbidity and mortality and the lack of evidence of benefits in the overall survival. Material (or patients) and methods: We conducted a retrospective study in 35 consecutive patients who received allo-HSCT in our hospital from June 1995 to January 2015. They were classified into two groups according to the type of conditioning myeloablative (MAC) vs reduced-intensity (RIC). Results: aGVHD grade II / IV incidence: 50% vs 64% (P = 0.2) in the MAC vs RIC group, respectively, the latter showing a higher incidence of aGVHD III-IV (P = 0.02). The aGVHD incidence grades II-IV and III-IV has decreased in RIC following the amendment of the GVHD prophylaxis regimen (Cyclosporine A plus micofenolate mofetil was replaced by sirolimus (SR) plus tacrolimus(TC): 70% vs 40%, (P = 0.002), and 40% vs 20%, (P = 0.04), respectively. In 2 patients SR and TC were discontinued because of developing of transplantationassociated microangiopathy (TAM). cGVHD incidence: MAC: 57% of patients suffered cGVHD (63% severe form). RIC: 67% developed cGVHD (50% severe form). Likewise, a lower incidence was observed after the change of prophylactic regimen: Global cGVHD is reduced from 88% to 25% (P = 0.03) and moderate to severe cGVHD from 75% to 25% (P = 0.1). No patient developed severe GVHD in the SR+TC group. RESPONSE AFTER SCT: MAC: there was an improved level of response after the TPH in 69% of patients (8 CR, 1 VGPR), while 31% (n = 4) did not improve the response (RP) or progressed. RIC: it was observed an improvement of response after TPH in 67% of patients (5 CR, 2 VGPR). HSCT does not modify the disease in patients trasplanted in progression. Chemosensitive disease at the time of conditioning was associated with a better post-transplant response (P = 0.002). MORTALITY: MAC: TRM-100 is 25%, mainly by SOS (60%). The overall TRM is 45% due to GVHD (45%), infections (22%) and SOS (33%). RIC: Both TRM-100 and overall TRM are 35%, due to: aGVHD (83%) and TAM (17%).PFS and OS: With a median follow-up of 56 months , actuarial OS at 5 years is 39 ±8%, with no differences according to the type of conditioning (MAC vs RIC: 29 ± 10% vs 50 ± 12%, P = 0.2) nor number of previous lines of treatment (P = 0.7). In RIC is associated with higher OS: achieving CR after allo-HSCT (P = 0.04), chemo-sensitivity before HSCT: 37 ± 16% vs 0%. 5% vs 14,1%) , myeloablative conditioning (67,5% vs 52,1%), high dose cyclophosphamide (67,5% vs 50,7% and antithymocite globulin (47,5% vs 32,4%, but these differences were not statistically significant (p40,05) In univariate analysis, were associated with HC: presence of acute GVHD (70% vs 43,7%, P = 0,006) and CMV reactivation (57,5% vs 28,2%, P = 0,002), and the latter is maintained in the multivariate analysis. None relationship was found between sex, age, graft source (peripheral blood/ umbilical cord) or underlying disease. Only 1/40 (2,5%) patient died as direct consequence of HC, and there were no differences in overall survival between HC and no-HC. But, there were statistically significant differences in: days of hospitalization (64 vs 44, P = 0,001), transfusion of packed red blood cells (15,5 vs 6, P = 0,001) and transfusion of platelet concentrates (12,5 vs 4, P = 0,000). Conclusion: In our case serie the presence of GVHD and CMV reactivation were found as factors associated with the development of HC. While we did not find a higher mortality rate assosiated to HC, we observed an association to longer hospital stay and blood product transfusions, suggesting a higher morbidity assosiated to HC development. Introduction: Allogeneic hematopoietic stem cell transplantation (HSCT), a well-established therapy for a variety of hematological malignancies, is unfortunately associated with significant morbidity and mortality related to cancer relapse and transplant complications, including graft versus host disease (GvHD). GvHD-free, relapse-free survival (GRFS) is a recently reported composite end point which allows estimating risk of death, relapse and GvHD simultaneously [1] . T-cell depletion (TCD) is a well established strategy for GvHD prevention. In the present work we investigated the effect of partial TCD (pTCD) on GRFS in order to evaluate its impact on patients' morbidity-free survival. Material (or patients) and methods: We performed a retrospective study on 333 patients who underwent allogeneic HSCT for hematologic malignancies at our center from 2004 to 2014 with grafts from HLA identical siblings or HLA 10/10 matched unrelated donors. 171 patients received pTCD grafts, obtained by incubation with alemtuzumab in vitro washed before infusion followed on day +1 by an add-back of donor T cells. 162 patients received T cell repleted (non-TCD) grafts. Donor lymphocyte infusions were given at three months to all patients receiving pTCD grafts with RIC without GvHD and when needed to all patients with mixed chimerism. Kaplan-Meier estimates and Log-rank test were employed to determine the probability of 5-year overall survival (OS), progression free survival (PFS) and GRFS. Events determining GRFS included grade 3-4 acute GvHD, systemic therapyrequiring chronic GvHD, relapse, or death. Cox regression was used to examine the independent effect on OS, PFS and GRFS of clinical factors including age, underling disease, disease status at transplant, disease risk index, conditioning, donor type, stem cell source, year of transplantation and T-cell depletion. Cumulative incidence estimates of relapse and nonrelapse mortality (NRM) were calculated with relapse or death from other causes defined as competitive events. Results: According to institutional practices, the group receiving pTCD grafts comprised more patients transplanted in complete remission (67%) than the group receiving non-TCD grafts (41%, P o0.0001). Similarly, the pTCD group comprised fewer patients with a high/very high disease risk index (17%) than the non-TCD group (51%, P o0.0001). pTCD was associated with improved OS and PFS in univariate analysis, but this association failed to reach significance in multivariate analysis. pTCD was associated with significantly improved GRFS (40%, 95%CI 33%448%) compared to non-TCD HSCT (24%, 95%CI 17%432%, P40.0001). The effect of pTCD on GRFS remained highly significant in multivariate analysis (HR 0.63, P = 0.004) . No effect of pTCD was observed on relapse cumulative incidence (P = 0.396), although this result may be the consequence of aforementioned differences in patient groups studied. Conversely, 5-year NRM cumulative incidence was significantly decreased in patients receiving pTCD (10%, 95%CI 6%415%) compared to patients receiving non-TCD allogeneic HSCT (15% 95%CI 11%422%, P = 0.045). Conclusion: pTCD appears to improve GRFS in allogeneic HSCT recipients, suggesting that pTCD could improve patients' quality of life without impairing the curative potential of allogeneic HSCT. 8 h (i.v.) . First-line therapy of NF was piperacillin-tazobactam 4.5 g/6 h (i.v.) or refrigerated meropenem 1 g/8 h (i.v.) using a portable intermittent infusion pump for both drugs. Teicoplanin was added if oral mucositis of NCI-CTC-score grade ≥ 2, signs of infection at the catheter insertion, or Gram-positive infection. Amikacin was started if fever persisted more than 3 days or in case of Gram-negative infection. Fever without hemodynamic instability responsive to first-line antibiotics was treated at-home. Only patients with NF and focal infection, persistent fever or signs of severe sepsis were readmitted. Results: Median age at transplantation was 48.6 (range, 17 to 69) years. All patients included had a good performance status (ECOG ≤ 2). The quantity of peripheral blood CD34+ cell dose (x10 6 /kg) infused was 3.9 (range, 1.5 to 21). Recovery (days) of granulocyte count above 0.5x10 9 /l was 11 (range, 7 to 26). The global incidence of grades ≥ 2 oral mucositis and gastrointestinal toxicity (NCI-CTC-score) was 21%. Of the 211 patients, 129 (61%) developed NF. Median day of appearance of fever was day +4 (range, day +2 to +13). Patients with fever received intravenous antibiotics for a median of 8 (range, 5 to 16) days, and teicoplanin and/or amikacin was added in 62 (48%) of the 129 febrile episodes. Bacterial infection was documented in 23 (18%) of febrile episodes. Coagulase-negative Staphylococci was the more frequent bacteria isolated (n = 17; 74%). The other infectious agents were Pseudomonas aeruginosa (n = 4), Streptococcus viridans (n = 1) and Enterococcus faecalis (n = 1). Of the 129 patients developing NF only 22 (17%) fulfilled criteria for readmission: pneumonia (n = 4), persistent fever (n = 15) and severe sepsis with hemodynamic instability (n = 3) requiring admission to Intensive Care Unit. Therefore, in 107 (83%) patients the NF was managed fully at-home and none of them died, while one (4.5%) of the patients readmitted died, which was due to respiratory syncytial virus pneumonia (P = 0.17 Full time employment post BMT decreased from 64% to 32.5% principally as a consequence of ill health with those in the lowest income strata increasing from 21% to 36%. Seventy-two percent and 8% of survivors were in married or defacto relationships, respectively, at the time of survey. Conclusion: This study provides the largest and most comprehensive account of the experience of survival following BMT in Australia. Survivors experience a high incidence and a broad range of physiological and psycho-social late effects of BMT, many of which are not currently addressed. Importantly, BMT also had a major financial impact on survivors with many no longer engaged in full-time employment or experiencing a decline in socioeconomic status. Our results reaffirm the need for continuing education and counselling (both pre-and post-BMT) and for ongoing multidisciplinary long-term follow-up of BMT survivors. Disclosure of Interest: None declared. Abt. Hämatologie und internistische Onkologie, 2 Klinik und Poliklinik für Gastroenterologie und Rheumatologie, Universitätsklinikum Leipzig, Leipzig, Germany Introduction: Allogeneic hematopoetic stem cell transplantation (HCT) as the only curative option for a multitude of hematologic diseases is increasingly used, even in elderly and unfit patients. HCT is associated with a considerable morbidity and mortality. Besides infections and Graft-versus-Host-disease, liver-related complications like sinusoidal obstruction syndrome (SOS) are major obstacles in the clinical routine. An important risk factor is a preexisting chronic liver disease. Liver elastography (Transient elastography, TE) combined with the controlled attenuation parameter (CAP) may be helpful to identify patients at high risk and thus allow risk stratification and preventive actions prior to conditioning and HCT. TE is an ultrasound-based technique using shear-waves to analyze liver tissue stiffness, which correlates with the extend of hepatic inflammation and fibrosis. CAP is a measure of ultrasound attenuation and correlates with the degree of liver steatosis. Both methods have been extensively evaluated in patients with chronic liver diseases of different etiology and represent accurate alternatives to liver biopsy. Material (or patients) and methods: Patients (pts) with malignant or non-malignant hematological diseases, who undergo HCT are prospectively evaluated before and post conditioning prior to HCT by ultrasound, liver elastography and non-invasive liver fat quantification by CAP. These parameters are correlated with the clinical course of the patients with an observation time post-HCT of 100 days. Event was defined as SOS, GvHD grad II -IV, sepsis from any cause or death. Results: 49 pts (median age 58 y, range 18-74, sex f/m: 19/30, median body mass index 25.0 kg/m 2 ) were included in the trial and transplanted until September 2015. Twelve pts (24%) had an event (2 female, 10 male) during the follow-up period (SOS n = 3, GvHD n = 7, sepsis n = 1, toxic liver damage n = 1). Five pts (12%) had increased liver stiffness indicating hepatic inflammation and/or risk of fibrosis prior to the conditioning therapy (TE47.1 kPa: 7.5-26.3 kPa). This was correlated to increased spleen size (181 vs 108 mm). Three of these five patient (60%) had an event during the follow up period. Follow up examinations (n = 19) after conditioning and prior to HCT showed considerable dynamics in some pts for liver stiffness (TE: -56 to +37%), size of the spleen (longitudinal diameter -21 to +12%). Furthermore, pts with an event had a significantly lower liver fat content at baseline measured by CAP (median CAP 222 dB/m vs. 243 dB/m, P = 0.0479). Conclusion: Our preliminary results show that ultrasound and elastography detect a relevant number of pts with risk factors for liver related complications after HCT. Low CAP-values at baseline were associated with higher risk of HCT related events. Thus, transient elastography combined with CAP could be used for risk stratification prior HCT and help to establish preventive actions. Disclosure of Interest: None declared. After allogeneic hematopoietic transplantation frequency of marrow clonogenic cells at engraftment is associated with immune activation, predicts treatment related mortality and is an independent factor for overall survival V. Zammit 1 Introduction: We have previously demonstrated that a reduced frequency of clonogenic cells in marrow, early during engraftment, preced the development of acute GVHD (Milone G, Exp Hematol 2015; 43:430) . Based on these results we hypothesized that evaluation of marrow function may also reflect alloreactivity and that may predict transplant related mortality. Material (or patients) and methods: In 88 consecutive patients (mean age 46 y) who underwent allogeneic HSC transplantation, we have prospectively studied, at day +18 after transplantation, frequency of marrow CFU-GM and BFU-E. At the same time point we collected, in a subset of 26 patients, serum samples to determine levels of soluble ICAM-1, IL-6 and soluble receptor for IL2 (rec IL2). All patients received T-replete allogeneic stem cell transplantation from HLA Identical sibling (n 44) of from an UD (n 44), Underlying diagnosis were: AML 47; ALL 14; MM 13; Lymphoma 14. Fifthy-two patients were in advanced phase and 36 in early phase. Results were related to clinical outcome in terms of Treatment Related Mortality (TRM), Relapse Rate (RR) and Overall Survival (OS). Results: Median Marrow clonogenic cell frequencies at day +18 were: CFU-GM = 126 (x10e5 plated cells); BFU-E = 60 (x10e5 plated cells); marrow clonogenic frequency was not different in transplant done from different donor type (CFU-GM was 122 and 146, in Mud and HLA-Id, P = 0.9) nor according to different HSC source (CFU-GM were 133 after transplant from PBSC versus 118 in transplant from BM; P = 0.9). No differences were found according to underlying diagnosis (P = 0.6) or phase of disease (P = 0.5). Marrow CFU-GM and BFU-E at day +18 was, however, lower in patients suffering GVHD and requiring corticosteroid treatment in respect to patients with no GVHD (CFU-GM: P = 0.03; BFU-E: P = 0.005). In the group of patients having at day +18 a frequency of marrow CFU-GM over the median, cumulative Incidence of TRM at 2 y was 7% while it was 37% in the group having CFU-GM below the median (Gray test: P = 0.0006). Cumulative incidence of Relapse was not significantly different in the two groups (35% versus 31%, Gray test P = NS). Overall Survival was significantly lower in group having frequency of marrow CFU-GM below median in respect of patients having frequency of marrow CFU-GM over median: 35% versus 62% (log-rank: P = 0.006). In multivariate Cox analysis, after adjusting for disease status (early versus advanced), recipient age and donor type (unrelated versus HLA-Id sibling), the frequency of marrow CFU-GM below median was retained as an important factor for an increased risk of death (HR = 2.830; P = 0.002). The group having a growth of CFU-GM below median, had, at day +18, a significantly higher level of rec IL-2 in serum (13,600 pg/ml versus 7,300 pg/ml, P = 0.04) while no differences were detected in IL6 levels and in soluble ICAM-1. Conclusion: Frequency of marrow clonogenic cells studied early at day +18 is associated to immune activation, predicts TRM and is an independent factor for OS. It can be used as sensitive biomarker for transplant outcome. Disclosure of Interest: None declared. Introduction: Hematopoietic stem cell transplantation (HSCT) affects patients' nutrition and metabolism and therefore, appropriate nutritional support seems to be essential for optimal regeneration. Currently there are no official recommendations regarding nutritional practices after HSCT and the approach to patient's feeding is based on local protocols and customs, which most likely vary among centers. The aim of the survey was to get insight into current nutritional practices in EBMT transplant centers. Material (or patients) and methods: The survey was submitted to all EBMT transplant centers (N = 469). The response rate was 27%. Results: About 80% of respondents gave a very high score of importance (47/10 degree scale) to appropriate feeding in peritransplant period. Dietician support is available in 65% of centers, but the final choice of nutritional support is made by physician (in 95% of centers). 47% of centers use standard lowmicrobial diet during peritransplant period in alloHSCT both after standard and RIC conditioning and 53% of centers use it after autoHSCT. In 34% of centers it is prohibited to use food brought from outside of hospital, while in 40% this practice is allowed. Besides situations of lactose intolerance, lactose-free diet is used in 22% of centers in case of gastointestinal (GI) GvHD. Nutritional status is formally assessed before HSCT in 57% of centers, usually by treating physicians (96%), but also by dieticians (in 34% of centers). During in-patient stay, it is performed as routine practice in 56% of centers while in 40% only if necessary. After discharge, nutritional counselling is performed routinely periodically in 21% of centers, while in 33% only if necessary. The method of nutritional support most commonly used as first-line intervention is using oral nutritional supplements (in 39% of centers). Oral immunonutrients are used in 25% of centers and in most (83%) it is glutamine, while in 35% omega-3 acids. Intravenous nutrition is used in 67% of centers, mostly consisting of glucose solutions (in 90% of centers), but also aminoacid solutions (in 76%) and lipid solutions (in 64%). Parenteral nutrition (PN) is used in 94% of centers. 17% of centers use it always after fulldose TBI-based conditioning, 23% always after chemotherapybased standard conditioning and 9% also after RIC. In 58% of centers only standardized formulae of PN are used, while 37% use individualized formulae. In 43% of centers, PN preparations are routinely enriched with glutamine. Tube feeding (TF) is used in 47% of centers and small proportion of centers use it routinely as prophylaxis of malnutrition. To perform TF, 95% of centers use naso-gastric tube, while 22% use naso-jejunal tube. In 85% of centers standard commercially-available formulae are used, in 32% of centers commercially-available, but disease-specific, while only 2.5% of centers use natural mixed food. In initial phase of GI GvHD, 22% of centers stop oral feeding already in stage 1 disease extending up to 85% of centers in stage 4 disease. Comparably, at stage 1, 26% of centers implement already PN extending to 89% of centers at stage 4. Over 80% of respondents agree, that diet in GI GvHD should be high-calory, high-protein and low-fiber. Conclusion: Altogether, the results of this multicentre-survey revealed heterogenous nutritional practices in EBMT transplant centers. Disclosure of Interest: None declared. Introduction: Ex vivo CD 34+ selection reduces GVHD after allo-HCT without increasing relapse, but requires a myeloablative conditioning regimen, and assessment of toxicities is important in older patients (pts) who are usually offered reduced intensity conditioning. Material (or patients) and methods: Grade 3-5 toxicities by CTCAE 4.0 were retrospectively collected on 200 pts who received CD34+ selection using the ClinicMACSs system between 2006-2012 and compared between pts ≥ 60 or o60 yrs using Cox regression. Results: 80 pts ≥ 60 (56% male) with acute leukemia/ myelodysplastic syndrome (MDS, 79%), myeloma (11%), and other hematologic malignancies (10%) were identified, with a median age of 64 (range 60-73) and median HCT-CI of 2 (0-10). Busulfan, fludarabine, and melphalan conditioning with rabbit ATG was used for 90%. The donor was matched related (MRD) in 38%, matched unrelated (MUD) in 45%, and mismatched unrelated (MMUD) in 18%. The o 60yr pts (49% male) with acute leukemia/MDS (53%), myeloma (18%), and other hematologic malignancies (15%) had a median age of 49 (19-59) and median HCT-CI 2 (0-9). Allografts were from a MRD (38%), MUD (35%), and MMUD in 26%. Pts o60 were more likely to receive total body irradiation (45 vs 2.5%, P40.001), but demographics were otherwise similar. Grade 2-4 acute GVHD by day 100 was 12.5% in pts ≥ 60 vs 11% in o60 (P = 0.82). Overall survival (OS) was not different between ≥ 60 and o60, with a 3-yr OS 49% in ≥ 60 and 59% in o60 (P = 0.11). At 3yrs, NRM was also similar with 30% in ≥ 60 and 25% o60 (Fig 1 A, P = 0.39) . The most common toxicities in pts ≥ 60 were cytopenias after day 30 (30% of total toxicities collected) & electrolyte abnormalities (15%). Infections included c. difficile (5% of infections) and viral organ disease (12%). In pts ≥ 60, HCT-CI ≥ 3 correlated with cardiovascular (HR 2.76, P = 0.006) and neurologic (HR 2.16, P = 0.045) and male gender with oral/GI (HR 0.38, P = 0.002). Disease other than leukemia correlated with hematologic (myeloma HR 3.04, other HR 6.02, P40.001). Median toxicities per pt ≥ 60 was 17 (1-52) and not numerically different than pts o60 (16, 2-63), but ≥ 60 had more neurologic (HR 2.75, P = 0.001) and less oral/GI (HR 0.62, P = 0.011) (Table) . The median number of toxicities at Day 100 was 9. Pts ≥ 60 who experienced ≥ median toxicities had a 3-yr OS of 61% vs 43% for pts witho median (P = 0.3). However, when compared to o60 yr pts with omedian toxicities, the 3-yr OS were significantly different ( Fig 1B) . headache, seizures, visual disturbance, and altered mental function associated with reversible white matter edema affecting the posterior parietal and occipital lobes of brain. PRES has been reported to occur in association with numerous co-morbidities, including hypertensive encephalopathy, preeclampsia, eclampsia, infections, electrolyte imbalance, immunosuppressive drugs, allogenic bone marrow transplantation (BMT), solid organ transplantation, autoimmune diseases, and high-dose cancer chemotherapy.We report ten patients who developed PRES during the prophylaxis and treatment with Cyclosporine A (CsA), tacrolimus and methylprednisolone for graft versus host disease (GVHD) after allogenic BMT. Material (or patients) and methods: Between 2013 and 2015, 90 patients received an allogeneic BMT at Acibadem Adana Hospital Pediatric Bone Marrow Transplantation Unit. Posterior reversible encephalopathy syndrome was observed in ten patients. Medical records and magnetic resonance images (MRIs) were evaluated. Results: We presented ten patients with PRES, age ranging from 3 to 19 with an average of 11.4 years. Of the ten patients, nine were diagnosed with thalassemia major and one with acute myeloid leukemia. All patients were treated with CsA or tacrolimus and methylprednisolone for prophylaxis and treatment of GVHD. PRES occurred at a median of 121 days (range 5-625 days) . Clinical findings at onset of leukoencephalopathy were hypertension, headache, seizures, visual disturbance, and altered mental function. One patient with thalassemia major died of PRES, the other 9 patients are alive with normal neurological status. MRI showed abnormalities in all patients including patchy bilateral cortical and subcortical lesions, especially in parieto-occipital lobes. Conclusion: Bone marrow transplantation is associated with several neurological complications that may be triggered by underlying diseases, BMT procedure, and severe immunosuppression. PRES is an uncommon but serious complication after BMT. We have reported nine cases of PRES who received allogeneic BMT to emphasize the importance of early recognition, treatment and appropriate management of PRES after the transplantation (Table 1) . [P219] S228 R-CHOP was the initial treatment in 2 patients whereas 3 patients had R-CODOX/M/R-IVAC and one patient received dose adjusted EPOCH chemotherapy. AlloSCT was performed after the initial response (all were in complete remission) with BEAM Alemtuzumab conditioning with cyclosporine as GvHD prophylaxis (4 unrelated and 2 sibling donors) with a median follow up of 30 months (range: (Figure 1) . In our small series there was no progression after 18 months from Allo SCT. One patient progressed 6 weeks and died 8 weeks post AlloSCT whereas another patient progressed 12 weeks and died 14 weeks post AlloSCT. Conclusion: Our single centre experience of limited number of patients with DHL suggests AlloSCT as a consolidative treatment in first complete remission, can induce prolonged survival with potential cure. This needs to be studied in a larger cohort to confirm our preliminary findings and lead to consensus in DHL management Introduction: The diagnosis of a malignant disease is always a severe threat to a person and adolescent as well as young adults (AYA) are in a damageable process of growing-up. Nearly all components of meaningful life are in a phase of reorganization. Otherwise, risk behavior in this age group is common and may affect the outcome of stem cell transplantation (SCT). Aiming to understand the special needs of a young minority of patients after SCT and to improve the psychosocial care we sought to evaluate their consulting requirements in a single center survey. Material (or patients) and methods: 35 AYA alive after SCT between 2010 and 2015 and a control group of 32 senior patients transplanted in 2014 at the Jena University Hospital were invited to participate. A questionnaire addressing the individual psychosocial and medical consulting requirements was sent to all patients. Results: 41 of 65 questionnaires (63%; 18/33 AYA and 23/32 non-AYAs) were completed and returned. Median age of participating AYA was 32 (21-39) years and 53 (40-63) years in the control group. Both groups consisted of 39% female and 61% male patients. On a one to six graded Likert scale issues regarding body shape as well as sexuality 50% of AYA compared to 26% of controls (P = 0.041) reported on high needs (value 5 -6) of counselling. Although not significantly different AYA had higher counselling requirements regarding medication (38%), family/relationship (44%), physical activity (50%) and rehabilitation (56%) than controls (26% / 22% / 33% / 33%), respectively. Other aspects, including nutrition, followup, vaccination, alternative medicine, skin alterations/alopecia, recreation techniques as well as level of care/premature pension were of high concern in most patients, but comparable between both groups. 81% of all participating patients would prefer individual consultation instead of grouped information sessions (13%). Conclusion: Conclusions: Both patient cohorts indicated need for advice after SCT by individual consultation. However, AYA reported on increased need for counseling in medical and psycho-oncological issues. They seem to benefit from special themes, i.e. body shape and sexuality. A prospective multicenter study in this minority of patients is ongoing in order to evaluate the special needs of AYA in the setting of SCT. Disclosure of Interest: None declared. The best 10 years' survival is seen in Artemis SCID (87.5%), ADA and IL2RG SCID (84%). Survival was significantly better in those diagnosed early (less than 4 weeks old) (90.7% vs 64.1%, p 0.003), those transplanted before the age of three months old (86.9% vs 65.1%, p 0.01), those without chronic GVHD (95.2% vs 58.3%, P40.0001), whom are free from infection prior to transplant (90.0% vs 68.1%, p 0.004) and those not requiring PICU admission (77.6% vs 52.5%, p 0.02). Conclusion: These data emphasise the importance of early diagnosis and transplant in SCID, which contribute significantly towards superior survival outcome following transplantation. Disclosure of Interest: None declared. Introduction: Hematopoietic stem cell transplantation (HSCT) is proven to be curative for severe combined immunodeficiency (SCID). Here, we discuss the outcome of donor chimerism and thymic output for IL2RG/JAK3 SCID posttransplantation at our centre. Material (or patients) and methods: A cross-sectional study of IL2RG/JAK3 SCID patients who underwent HSCT from 1987-2012. The conditioning regimens used were reduced intensity conditioning (RIC) (Fludarabine/Melphalan), low toxicity myelo-ablative conditioning (Treosulphan/Fludarabine or Treosulphan/Cyclophosphamide) and myelo-ablative conditioning (MAC) (Busulphan/Cyclophosphamide). CD4+ naïve cells (CD3+CD4+45RA+) measurement was used as an indicator of the thymic output post-transplantation. Results: 31 of 43 patients with IL2RG/JAK3 SCID survived. The 10-year survival was 71.9%. Median age at last follow up was 10 years old (range 2-25).495% donor T cell chimerism was observed for all patients irrespective of donor and conditioning regimen received. However, B cell and myeloid donor chimerism at last follow up was best following low toxicity or myelo-ablative conditioning regimen with matched related (MRD) or matched unrelated donor (MURD) compared to unconditioned and matched sibling donor (MSD) recipients. Most haplo-identical donor recipients have poor myeloid chimerism ( o5% donor cells), despite receiving MAC conditioning. 16/29 are off IVIG (10 B and myeloid chimer-ism450% donor, 6 B and myeloid chimerism o50% donor). 13/29 with B and myeloid chimerism o50% donor continue IVIG. The mean CD3+ count at last follow up is highest for those who received MSD and unconditioned transplants. Recipients of low toxicity conditioning have higher CD3+ counts compared to RIC and MAC conditioning, across all types of donor. MSD with unconditioned transplants have the best thymic output at last follow up, followed by MRD and MURD with low toxicity myelo-ablative conditioning. Conclusion: T cell donor chimerism is not influenced by conditioning or donor types. However, conditioning with MRD or MURD demonstrated better myeloid cell chimerism compared to unconditioned MSD or haplo-identical donor recipients with MAC. Sustained CD3+ output are seen after 20 years post-HSCT for of IL2RG/JAK3 SCID and even though unconditioned MSD recipients have poor myeloid chimerism; they have highest thymic output at last follow up. Disclosure of Interest: None declared. Introduction: Improvement in hematopoietic stem cell transplantation (HSCT) techniques and medical care has improved post-transplant survival for patients with severe combined immunodeficiency (SCID). Many studies have demonstrated evolution of immune-reconstitution, but there are scarce data considering long-term quality of life (QoL), with one study suggesting poor function compared to healthy controls(1). We objectively assessed QoL of SCID survivors at our centre according to their genetic diagnosis. Material (or patients) and methods: All SCID patients more than 2 years post-HSCT attending the HSCT Clinic follow up were invited to answer the Pediatric Quality of Life Inventory (PedsQL), Generic Score Scale v4.0 questionnaires as part of the routine psychology assessment. The PedsQL questionnaires consist of parent and patient reports, both have 6 domains (physical, emotional, social, school, psychosocial and total). Results: 59 of 88 (67%) patients responded; comprising 45/73 (62%) patients aged 45 years and 49/77 (64%) families. Fourteen children aged o5 years old were excluded from the child questionnaires. Twenty-eight patients were not contactable and three refused. Most older patients answered the questionnaires, whereas parents answered for most of the younger patients. The median interval post-HSCT was 11 years (range, 2-27). The proportions of responder according to SCID genotypes were IL2RG/JAK3 SCID (20 of 31, 65%), IL-7Ra Deficiency (10/14, 71%), Adenosine Deaminase (ADA) (12/16, 75%), Artemis (5/7, 71%) . QoL for patients with IL2RG/JAK3, IL-7Ra Deficiency and Artemis SCID were not significantly different to published UK norm (Patient Report) in all domains, in contrast to a previous study (1) . However, ADA SCID survivors had significantly lower QoL, across all except the emotional components. Parents of IL2RG/JAK3 SCID reported significantly lower QoL in school, psychosocial and total domains compared to UK normals, whilst parents of ADA SCID reported significantly lower QoL across all except the emotional component domains. The QoL scores were not significantly different between IL-7Ra and Artemis deficient SCID (all domains) ( Table 1) . Parents and children of RAG1/2 SCID reported significantly lower QoL in school component compared to UK norm. Parents of IL2RG/JAK3 SCID with on-going intravenous immunoglobulin therapy reported lower QoL compared to those without the therapy. Conclusion: In contrast to previously(1), a number of SCID genotypes were associated with normal Qol. Particular risk factors include ADA SCID and need for ongoing medication. Factors postulated as causing reduced QoL include prolonged hospitalization, parental consanguinity and on-going medical issues or medication, but most parameters we examined were normal. Larger qualitative studies are needed to clarify QoL differences in SCID survivors. Introduction: Several models to predict outcome of allo-HSCT have been used. There is a continued need to define and develop risk-assessment measures to help physicians throught the decision making process. Aim: we wanted to identify pretransplant features associated with worse survival in our series of patients. Material (or patients) and methods: We retrospectively analysed 229 patients who underwent allo-HSCT in our centre during the last years. Association between survival and pretransplant HCT comorbidity index (HCT-CI), EBMT risk score (EBMT-RS), serum ferritin (PTSF), PRBC transfusions (tPRBC), patient/donor ABO blood groups, patient/donor age and patient/donor gender was analysed. Results: Median age of the series was 47 years . Median follow-up was 14.3 months (0.3-164.9). Overall survival (OS) at days +100 and +365 was 89.6% and 66.7%, respectively. PTSF (median: 1497 [17.6-8286.4] ), tPRCT (median: 24 ) and EBMT-RS (median: 3 [1] [2] [3] [4] [5] [6] ) were associated with shorter OS (log-rank test). A higher risk of death was observed in patients with: PTSF ≥ 1000 ng/mL (HR: 1.98, CI: 1.05-3.71, P = 0.03), tPRCT≥40 (HR: 1.71, CI: 1.13-2.60, P = 0.01), tPRBC≥20 (HR: 1.49, CI: 1.01-2.20, P = 0.04) and with the increase of each point of the EBMT-RS (HR: 1.34, CI: 1.01-1.78, P = 0.04). No significant impact on OS was observed for HCT-CI or for patient/donor gender, age and ABO group. Conclusion: In our experience, PTSF, tPRBC and EBMT-RS are associated with OS of allo-HSCT patients. Disclosure of Interest: None declared. A cohort pilot study on the use of a preventive treatment of oral mucositis using low-level laser therapy on patients undergoing a haematopoietic stem cell transplantation I. Jeandet 1,* , J. A. Martignoles 1 , S. Tronchon 1 , C. Brouillat 1 , S. Filiol 1 , C. Villemagne 1 , S. Benderbal 1 , E. Chalayer 1 , J. Cornillon 1 , D. Guyotat 1 , E. Tardy 1 1 Hematology, Institut de Cancérologie de la Loire, Saint Priest en Jarez, France Introduction: Oral mucositis (OM) is a serious and frequent side effect due to the conditioning regiments (CRs). The data of nearly 200 patients undergoing an autologous hematopoietic stem cell transplantation (HSCT) have been prospectively analysed by the EBMT Mucositis Advisory group (1). This study revealed that 45% of patients suffered from grade 3 or 4 OM and with the CRs using 12 Gy TBI the risk of severe OM increased up to 75%. Low level laser therapy (LLLT) can be used to treat and prevent OM due to its anti-inflammatory, antalgic and healing properties. The American guidelines published in 2014 recommends the use of the LLLT to prevent OM (2) . However, this procedure is not as developed among the French haematological centres. Thereby we report the results of a pilot study showing the feasibility of this process and the outcome. Material (or patients) and methods: The design of our study is a prospective and monocentric study involving 47 patients who underwent an autologous or allogeneic HSCT from May 2014 to June 2015. The chosen protocol for the preventive use of the LLLT consisted of sweeping the entire oral cavity for 2 minutes at a power setting of 250mW. The first session was performed on day 2 of the conditioning and carried out on a 48 h rotation until day 10. The treatment of a proven OM was performed every day until signs of healing appeared. The treatment was carried out by a trained nurse or nurses' aide and consisted of a 10 s per cm 2 treatment at a power setting of 500mW on the OM's lesion. Results: The feasibility of a systematic preventive treatment of OM has been shown. The training of the caregivers (nurses and nurses' aide) with the use of the LLLT was straightforward. The process was not time consuming. The patient's acceptance of treatment was easily gained which was proven by the fact that no patients refused treatment. The tolerance was excellent due to the absence of any sides effects. Concerning the characteristics of the 47 patients: 25 men, 22 women, median age of 51years old (20-60), 35 of which received an autologous HSCT (18 BEAM, 15 Melphalan based CR and 2 BuCy). 12 of which received an allogeneic HSCT with the use of a myeloablative conditioning regimen (3 Cy-TBI, 7 FB4 or FBC and 2 BuCy). The median number of viable CD34+ cells reinfused was 4.76x10 6 /kg and the median duration of aplasia was 11 days (6-37). 33% of the patients did not suffer from an OM grade41. Among the OM grade 3 patients, the symptoms duration lasted for 6 days. No patients suffered from a grade 4 OM. There was no link observed between OM and the type of CR, the aplasia duration nor the HSV reactivation. Introduction: Patients undergoing allogeneic transplantation (SCT) are at increased risk for thromboembolic events. When evaluating the risk of recurrent events and need for secondary prophylaxis, genetic testing for inherited thrombophilia is widely used. Material (or patients) and methods: We describe exemplarily two patients with thrombophilia testing post SCT. Patient A was tested homozygous for FV Leiden (FVL) at the age of 49. Consistent with this finding the APC test showed pathological resistance. He never experienced a thrombosis before. Ten years later the patient was diagnosed with secondary acute myeloid leukemia (AML) making an allo-SCT the appropriate cure for the disease. During the initial chemotherapy for AML but prior to SCT he experienced a thrombosis of his jugular vein triggered by a central venous catheter. One month later he underwent allo-SCT and experienced another thrombosis of his right leg during hospitalization. From that time on he received heparin therapy continuously. With the intention to stop anticoagulation, genetic testing for FVL was performed again and showed no mutated status but wild type of both alleles. The functional APC test, though, showed persistent pathological resistance indicating a permanent increased risk for a recurrent thromboembolic event. Patient B received allo-SCT for childhood precursor acute lymphoblastic leukemia (ALL). During prior therapy according to the ALL-BFM protocol he developed recurrent thrombosis, at least one provoked during asparaginase application. During treatment before SCT a decreased protein C level was noted, which improved to normal shortly after SCT, but deteriorated again. After counselling to determine his risk of further thrombotic events a molecular analysis of non-hematopoietic cells (buccal swab) was performed which could rule out a mutated protein C gene. Results: In patient A, genetic analysis of peripheral blood for the FVL mutation produced a false negative result due to the patient's prior allogeneic SCT. Using APC as a functional test for the circulating F V which is synthesized by the recipient liver however revealed the homozygous state of the FVL mutation. Together with the clinical presentation with multiple thrombotic events an indefinite anticoagulation should be considered. In a situation where a functional test is not available, a molecular analysis of non-hematopoietic cells might be necessary (patient B). In an era of increasing use of transplantation special care must be taken when considering genetic tests in transplant recipients. Both after allogeneic stem cell transplantation and liver transplantation misleading genetic test results are possible if peripheral blood leucocytes are used (1, 2) . Further, knowing the site of synthesis of clotting factors is crucial: In particular testing for FVL might elucidate discrepancy in genetic and functional testing using APC resistance as circulating FV is synthesized primarily by the patient's hepatocytes with a smaller pool originating from donor megakaryocytes. In contrast, in patients after liver transplantation care must also be taken in genetic analysis of clotting factors produced by the new liver. Introduction: Unrelated mismatched volunteers (MMUD), cord blood or haploidentical(haplo) related donors are alternative stem cell source for patients with hematological malignancies, candidates to allogenic transplantation(HSCT), lacking a matched family donor. In these patients, due to sensitization for previous pregnancies and/or transfusions, antibodies against HLA loci can be observed. The presence of donorspecific antibodies(DSA), seems to be predictive of graft failure. However, so far, neither a clear approach to manage the presence of DSA, nor a standard desensitization protocol are available. The aim of this study was to evaluate the incidence of DSA in patients with a donor search for HSCT, and the efficacy of our desensitization protocol. Material (or patients) and methods: From August 2014 to October 2015, we prospectively screened for DSA, 37 consecutive patients candidates to allotransplant,using multianalyte bead assays, performed on the Luminex platform, including Lifecode Screen and LSA I/II (Immucor). The results were expressed as mean fluorescence intensity(MFI);MFI ≥1000 was considered positive.According to our policy,if patient had DSA and more than one available donor,we selected the donor to whom patient had no DSA.If only one donor was available,we employed a desensitization protocol, before starting,on day -7, pretransplant conditioning regimen. The aim of this schedule was to avoid interference with chemotherapy and anti T cell globulins,infused during condition regimen.Anti-CD20 monoclonal Ab(rituximab)on day -15,2 single volume plasmapheresis(PP)on day -9 and -8, endovenous Ig on day -7 have been administered.HLA selected platelets for DSA absorption were infused in case of the persistence of antibodies before transplantation. Results: Ten out of 37 patients(27%) showed anti HLA antibodies: 5 for I class HLA antigens,3 for class II and 2 patients for both I and II class, respectively.Three out of 10 patients showed DSA (8%). One patient, having an alternative MMUD, could change the donor. Two patients were treated with the desensitization protocol;the first patient, with an haplo donor, obtained a fast decrease of DSA level; the second patient, due to the persistence of high DSA level against MMUD, received 2 transfusion of HLA DSA selected platelets units. Both patients obtained engraftment on day +21 and +22 after transplantation, respectively. Neither DSA rebound nor other complications were observed during the follow-up. Conclusion: Our prospective unicentric analysis suggests that anti HLA antibodies are frequent(27%) in hematological patients, therefore DSA should be evaluated routinely in HSCT with HLA mismatched donors.In our experience the combination of PP, rituximab, IVIG and platelet absorption, showed successful DSA reduction in sensitized patients. Prospective multicentric studies are required to define the role of DSA against each HLA locus and the MFI cutoff level, associated with higher risk of graft rejection.Moreover,transplant and transfusion specialists should joint to design a standard desensitization protocol. According to the level of serum ferritin (SF) before transplantation, we divided the patients into two groups: the effective treatment group (19 cases, SF o1000ng/ml) and iron overload group (38 cases, SF ≥ 1000ng/ml). Results: (1) 30/57 cases were received iron chelating treatment, 19/30 (63%) cases obtained obvious curative effect, the SF level before transplantation waso1000ng/ml, the median was 561(223-846) ng/ml. The effort of other 11/30 (37%) cases was not obvious, with the SF level before transplantation was ≥ 1000ng/ml,the median was 1262(1100-2352) ng/ml. 27/57 patients didn't received iron chelating therapy before transplantation, the SF level before transplantation was ≥ 1000ng/ml,the median was 1540(1320-3112) ng/ ml. (2) The rate of fully-engraftment in the effective treatment group and iron overload group was 19/19 (100%) and 34/38 (89%). And the latter group had 4/38 (11%) cases failed the hematopoietic reconstitution.The medium time of myeloid and platelet reconstitution of the 19 cases of effective treatment group was (12 ± 2) and (16 ± 6) days respectively, while those of 34 cases of iron overload group was (13 ± 3) and (18 ± 6) days. The hematopoietic reconstruction was shorten in the former group, however the difference between the two groups had no statistical significance (P = 0.441, P = 0.579). (3) The infection rate of the effective treatment group and iron overload group was 7/19 (37%) patients and 28/34 (82%) patients respectively. The risk of infection of the effective treatment group was decreased significantly (P = 0.002). (4) The incidence of aGVHD of the effective treatment group was 5/19 (26%) patients, all of which were I-II degree. The incidence of aGVHD of iron load group was 22/34 (65%) cases, with 16 cases of I-II degree, 6 cases of III-IV degree. The risk of aGVHD in the effective treatment group was significantly decreased (P = 0.01). (5) The median disease-free survival (DFS) in effective treatment group was 28.9 (0.3-89.5) and 21.2 (0.1-81.0) months in iron load group during a median follow-up period of 22.0 (0.1-89.0) months.The DFS of effective treatment group was prolonged (P = 0.053). Conclusion: Effective iron chelating therapy before transplantation was helpful to hematopoietic reconstitution, and significantly reduced the incidence and degree of infection and aGVHD, decreased transplantation related mortality(TRM) and prolonged DFS, thereby improved the success rate of transplantation in MDS. Disclosure of Interest: None declared. Optimizing busulfan exposure in pediatric hematopoietic cell transplantation using a weight-based dosing nomogram and therapeutic drug monitoring E. Van Maarseveen 1,* , L. van Reij 1 , T. Egberts 1 , K. Rademaker 1 , J. J. Boelens 2 1 Clinical Pharmacy, 2 Laboratory of Translational Immunology, University Medical Center Utrecht, The Netherlands, Utrecht, Netherlands Introduction: Busulfan (Bu) is an alkylating drug used in conditioning regimens for allogeneic hematopoietic cell transplantation (allo-HCT). Its narrow therapeutic range in combination with large inter-individual variability in exposure, even after intravenous administration, necessitates dose individualization. Recently, the optimal Bu exposure in paediatric allo-SCT was identified in a large multi-center study (1) . Striving for optimal Bu exposure a weight-based dosing nomogram was introduced at our paediatric HCT unit to reduce between-patient exposure varation (2) . In this study we prospectively assessed target attainment of Bu exposure using this model-based dosing nomogram supported with therapeutic drug monitoring (TDM). Material (or patients) and methods: All paediatric patients who underwent allo-HCT receiving Bu-based conditioning were included. Bu was administered once daily in a 3-hour infusion on four consecutive days. TDM-based dose correction was standard protocol. Drug levels were measured on day 1 and 4. Bu exposure was expressed as cumulative area under the concentration-time curve (cAUC) and estimated using a PK model and bayesian-based PK software (MwPharm 3.60). Bu target exposure was set at 80-100 mg*h/L. For each patient a 'hypothetical' cAUC without TDM was determined by extrapolating the AUC on day 1. Secondly, the 'true' cAUC was estimated based on PK data obtained on day 1-4 with incorporation of TDM-based dose adjustments. Medians and ranges were compared between cAUCs determined with and without TDM-based dose corrections. Finally, target attainment rates were compared between 'hypothetical' nomogram-based dosing simulation and the 'true' nomogram with TDM-based dosing situation. Results: Fifty patients were included with a median age of 9 years (3 months -18 years). Without TDM the median cAUC was 85,3 mg*h/L versus 96,2 mg*h/L with TDM. The range in individual cAUCs was significantly larger without TDM (45-156 mg*h/L) than with TDM (74-117 mg*h/L) (P = 0.001). Without TDM 34% of patients reached target cAUC 80-100 mg*h/L and with TDM this significantly increased to 70% of patients (P = 0.001). Conclusion: The weight-based dosing nomogram overall led to a mean busulfan exposure within the target range of 80-100 mg*h/L in paediatric patients, yet the inter-individual range was substantial. Therefore, TDM of intravenous busulfan remains recommended and is of utmost importance to reach optimal target exposure in paediatric patients to improve HCT outcomes. Introduction: Reduced-intensity conditioning (RIC) regimens cause lower non-relapse mortality (NRM), but result in higher relapse in higher-risk myelodysplastic syndrome (MDS). However, relapse is not the main problem after hematopoietic cell transplantation (HCT) in lower-risk MDS and post-transplant outcomes might be better with less intense non-myeloablative conditioning (NMC). We report here the results of a singlecenter feasibility study for NMC with cyclophosphamidefludarabine-antithymocyte globulin (CyFluATG) in MDS patients with bone marrow blasts less than 5 percent. We compared post-transplant outcomes between NMC with CyFluATG and RIC with busulfan-fludarabine-antithymocyte globulin (BuFluATG) that was used at the same center. Material (or patients) and methods: Fifteen MDS patients with bone marrow blasts o5% received allogeneic HCT after CyFluATG conditioning; cyclophosphamide (50 mg/kg x 2 days), fludarabine (30 mg/m 2 x 5 days), and rabbit antithymocyte globulin (ATG, 1.5 or 3.0 mg/kg x 3 days). Cyclosporine and short-course methotrexate were given for prophylaxis of graft-versus-host disease (GVHD). As a historical control, we selected 30 MDS patients who underwent HCT after BuFluATG; busulfan (3.2 mg/kg x 2 days), fludarabine (30 mg/m 2 x 6 days), and ATG. Patient and transplant characteristics were not significantly different between CyFluATG and BuFluATG except time period from diagnosis to HCT. The major transplantation outcome was similar in both groups (Table 1) . During the median follow-up duration of survivors of 45.8 and 74.0 months for CyFluATG and BuFluATG, three patients died and 2 relapsed in CyFluATG, and 8 died and 2 relapsed in BuFluATG. The 4-year overall survival (OS), event-free survival, NRM, and relapse rates were 80.0%, 63.0%, 20.0%, and 17.0% in CyFluATG and 73.3%, 73.3%, 20.0%, and 6.7% in BuFluATG, respectively. The incidences of grade 2-4 acute and chronic GVHD were 13.3% and 26.7% in CyFluATG, and 20.0% and 47.1% in BuFluATG. The neutrophil and platelet engraftment was significantly faster in CyFluATG compared to BuFluATG (median 12 vs. 14 days, P = 0.005 for neutrophil; median 15 vs. 21 days, P = .032 for platelet). CyFluATG showed faster immune reconstitutions of T-cells at 1 month after HCT than BuFluATG. 3 of 4 women of childbearing age in CyFluATG resumed their menstruation cycles and one successfully delivered a baby, while no woman in BuFluATG recovered menstruation again. Conclusion: In conclusion, NMC with CyFluATG regimen was feasible in MDS patients with low bone marrow blasts percentage in terms of engraftment and NRM. Posttransplant outcomes were comparable between CyFluATG and BuFluATG, and faster neutrophil engraftment and early T-cell reconstitution were observed with CyFluATG. Fertility of women seems to be maintained after CyFluATG. Disclosure of Interest: None declared. Conditioning with total body irradiation without antithymocyte globulin is assoicated with better outcome of cord blood transplantation in children with acute leukemia H. Introduction: While cord blood transplantation (CBT) is widely used for the cure of various diseases including acute leukemias, the optimal conditioning regimen needs to be determined. This study was aimed to identify the impact of conditioning regimen on the outcome of CBT in children with acute leukemia. Material (or patients) and methods: Medical records of patients with acute leukemia who received CBT at our institution between Jan 2002 and Jun 2015 were reviewed. Patients who had received prior transplantation were excluded. Patients were allocated into 3 groups (TBI+ATG-, TBI-ATG+, and TBI+ATG+) according to the presence or absence of total body irradiation (TBI) and antithymocyte globulin (ATG) in the conditioning regimen. Results: Seventy-one patients were identified and their median age at CBT was 7.3 y (range, 0.6-18.5). TNC and CD34+ cell doses were similar between three groups (P = 0.38 for TNC; P = 0.31 for CD34+ cells). All patients in TBI+ATGgroup (n = 20) were successfully engrafted, while 8 of 36 patients (22.2%) in TBI-ATG+ group and 3 of 15 patients (20.0%) in TBI+ATG+ group failed to achieve donor-origin hematopoietic recovery (TBI+ATG-vs. the others, P = 0.027). The incidence of cytomegalovirus (CMV) disease was 1 in TBI +ATG-group (5%), 11 in TBI-ATG+ group (30.6%), and 5 in TBI +ATG+ group (33.3%) (TBI+ATG-vs the others, P = 0.028). Early blood stream bacterial infection within 30 days after CBT was more frequently observed in TBI+ATG+ group (n = 7; 46.7%) as compared with that of TBI+ATG-group (n = 7; 35.0%) and TBI-ATG+ group (n = 11; 30.6%). The 5-y overall survival rates of TBI +ATG-, TBI-ATG+, and TBI+ATG+ groups were 81.1%, 53.8%, and 33.3%, respectively (P = 0.015). Conclusion: Patients conditioned with TBI without ATG achieved better engraftment and survival rates less suffering from CMV disease. Our data suggest that TBI-based conditioning without ATG should be a preferred option for pediatric patients with acute leukemia undergoing CBT. Disclosure of Interest: None declared. Results: Engraftment was prompt in all pts: day +10 (8-11) for neutrophils and +12 (9-37) for platelets, with 6 median days of neutropenia and 74% of pts achieved 499% donor chimerism by day +30. No grade III-IV organ toxicities were noted: mucositis gr I-II (26), mild hyperbilirubinaemia (5) and renal toxicity (3). During neutropenia 12 pts were treated for infections (pneumonia 1, gram negative perianal 1) and post-discharge 13 pts (2 gram negative and 1 candidaemia). Ten pts experienced CMV reactivation (4 infections), 13 EBV (1 PTLPD) and 1 tuberculosis (TBC). Until last follow up 9 developed grade ≥ II aGVHD (5 grade III-IV) and 11 pts cGVHD. Twenty pts are alive (19 in CR), while 5 died: 2 aGVHD grade III/ IV, 1 cGVHD overlap, 1 TBC (TRM 15%) and 1 disease related. The historic group consisted of 26 consecutive pts treated with FluBuATG (2002 FluBuATG ( -2012 for AML (20), MDS (5), MDS related AML (1) . The groups were similar in terms of age, disease risk, lines of treatment or HCT-CI, except for more UD in FluTreo (x 2 test). With 14 (2-136) for FluBuATG and 13 (2-27) months follow-up for FluTreo, 1-year cumulative survival probability was 57% vs 83% (P = 0.015), 1-year DFS 38% vs 89% (P = 0.001) and relapse rate 61% vs 8% respectively. In multivariate analysis regimen, age, disease phase, lines of treatment, HCT-CI score and GVHD were incorporated. The type of donor was not associated with the transplant outcome in univariate analysis and was not included. In terms for OS and DFS, the only significant favorable factor was FluTreo (P = 0.03 and P = 0.003). Conclusion: With the limitation of short follow-up, it seems that the Treosulfan-based regimen provides favorable OS and DFS compared with busulfan-based reduced intensity conditioning. The significant cytotoxicity of high dose Treosulfan along with its favorable extramedulary toxicity seems to underline the importance of myeloablation in order to achieve long term disease control, while maintaining low TRM in these medically infirm pts. Introduction: Total body irradiation (TBI) has been the backbone of hematopoietic stem cell transplantation (HSCT) for pediatric acute lymphoblastic leukemia (ALL). But TBI results late complication such as growth failure, endocrine dysfunctions, secondary malignancy and cataract in children. Many high dose chemotherapy based conditioning regimen has been developed. Furthermore, chemotherapy regimens may avoid or diminish late complications from irradiation. Material (or patients) and methods: We retrospectively evaluate the result of HSCT for patients diagnosed with ALL. From January 2003 to August 2015, twenty-four patients, aged less than 18 years, received myeloablative HSCT in pediatric hemat-oncologic department of Yeungnam University Hospital. The median age at HSCT was 8.4 years and the median period from diagnosis to HSCT was 1.5 years. Ten patients received HSCT at CR1 and 14 patients received HSCT at CR2. Stem cell sources were bone marrow (n = 6), peripheral blood (n = 9), and cord blood (n = 8). One patient received bone marrow and cord blood simultaneously. Twenty patients (83%) received melphalan based conditioning regimen. Eleven patients received stem cell of HLA-matched donor and 9 patients received stem cell of HLA-mismatched donor. Four patients received autologous HSCT. Seven patients were treated with TBI based conditioning regimen and 17 patients were treated with high dose chemotherapy based conditioning regimen without TBI. Results: Overall survival (OS) of non-TBI group was not inferior to TBI group. OS at 5 years after HSCT of TBI and non-TBI group was 71% and 74%, respectively (P = 0.712). Event free survival (EFS) at 5 years after HSCT of TBI and non-TBI group was 71% and 68%, respectively (P = 0.289). Relapse incidence (RI) at 5 years after HSCT was 16% in TBI group and 0% in non-TBI group (P = 0.052). Also, nonrelapse mortality (NRM) was 71% and 80% respectively (P = 0.747). HLA match donor group had better outcomes than those of HLA mismatch donor group. Stem cell source, age, status at HSCT, immunophenotype and sex were not associated with survival outcome. Patients with melphalan based chemotherapy had better EFS than nonmelphalan group (P = 0.01). S236 (4 CR1; 6 CR2; 2 CR3+), 6 had MDS (5 RAEB2; 1 RCMD), 4 had CML (3 blast crisis), 2 had CLL (1 Richter transformation), and 3 had MPD. We used a risk-adapted adjusting conditioning intensity of busulfan at 130 mg/m2/day intravenously for either 2, 3 or 4 days, with a fixed dose of fludarabine (F) at 30 mg/m2/day for 5 days, and of thymoglobulin (ATG) at 2.5 mg/kg/day for 2 days. Our algorithm was based on age, co-morbidity and disease risk. Two (3.8%) patients had FB2-, 14 (26.4%) had FB3-, and 37 (69.8%) had FB4-based conditioning. Five patients with high risk ALL received TBI 4 Gy in addition. As GVHD prophylaxis, all patients received cyclosporin A (3 mg/kg/day) initiated three days prior to stem cell infusion. Post-transplant therapy was initiated between days 30 and 60 after allo-SCT. Twenty six patients (49%) were planned for therapy post allo-SCT, 1 of them relapsed before starting it and 25 received it (10 vidaza (32 mg/m2/day for 5 days), 5 sorafenib (400mgx2/day), 4 dasatinib (50 mg/day), 1 dasatinib/ imatinib (400 mg/day) and 5 intrathecal cytarabine) whereas 27 patients (51%) could not receive it because of lack of approval by their third party payers. Results: All 53 patients, median age, 37 (16-65), received peripheral blood stem cells from HLA identical siblings. After a median follow up of 13 (range, 2-57 months), one-year nonrelapse mortality (NRM) was 2% and the cumulative incidences of grade 2 to 4 acute graft-versus-host disease (GVHD) and all grades chronic GVHD were 23% and 9%, respectively. The two groups (post transplant therapy YES or NO) were comparable for age, gender, disease subtype and distribution of CIBMTR risk groups, with a slight preponderance of higher risk receiving post-transplant therapy. Interestingly, on an intentto-treat analysis, 2-year overall survival (95% vs 61%; P = 0.04; Figure 1 A) and progression-free survival (81% vs 53%; P = 0.05; Figure Introduction: The publication of larger scale studies examining the role of Bu-Flu in AML and MDS has shown the clinical utility of this regimen in patients not appropriate for a full intensity allograft. Thymoglobulin (T) added to allograft preparative regimens has decreased the incidence and severity of acute (aGVHD) and chronic GVHD (cGVHD), but there remains controversy over optimal dosing. Material (or patients) and methods: To address the concerns that adding T to fludarabine in this regimen could result in greater t-cell depletive effects in our practice, we undertook a retrospective analysis of 74 patients (age range 38-73 years) with AML and MDS transplanted from 2005-2015 using the Bu-Flu preparative regimen with and without T. 55 had AML and 29 with MDS including 21 of these with residual disease; 28 patients received allografts from MRD and 46 from MUD donors (1 MMRD, 7 MMUD) . Results: The cumulative incidence of grade 2-4 aGVHD, cGVHD, relapse and NRM were 27%, 41%, 35% and 19% respectively. After a median follow-up of 10.2 months (0.92-121.9), PFS and OS were 46% and 57%. Comparing cohorts who did not receive T (13) vs those who did (61), patients receiving T had higher rates of CMV reactivation (P = o0.01) and relapse (P = 0.02), while the no T cohort was more likely to have aGVHD (P = 0.04), cGVHD (P = o0.01) and a higher NRM (P = 0.03) without a significant difference in PFS or OS. To determine if donor source played a role, we analyzed the impact of T use in MUD and MRD patients separately. MUD patients were more likely to show CMV reactivation (P = 0.03) while MRD patients were more likely to develop cGVHD (P = o 0.01). This was due to the higher rate of no T use (46%) within the MRD group, with more grade 2-4 aGVHD (P = 0.06) and more moderate to extensive cGVHD (P = o0.01). In addition, patients within the MRD group that received T had significantly more relapse (P = 0.04), less NRM (P = 0.02), and a non-significant difference in CMV reactivation, PFS and OS. While all MUD patients received T, there was a notable dose effect on aGVHD and cGVHD incidence and severity. Even though there was a higher incidence of aGVHD in patients receiving less than 4 mg/kg compared with 4-5 mg/kg and greater than 5 mg/kg, cGVHD and in particular, moderateextensive cGVHD was more likely in patients receiving 45 mg/ kg T (P = o0.01). CMV reactivation, relapse, NRM, PFS and OS were not impacted by ATG dose in the MUD group. Multivariate analysis controlling for confounding variables showed that ATG use had a significant HR 14.5 for CMV reactivation (P = 0.02), and a strong trend to significance for cGVHD (HR 0.2, P = 0.08) and for relapse (HR 14.2, P = 0.06). Patients developing aGVHD and cGVHD had significantly less associated relapse and CMV reactivation accounting for the insignificant impact on survival. Conclusion: While these results compare favorably with recent publications evaluating Bu-Flu-T regimens in the US and EU, this report demonstrates the differential dosing effects in MUD allografts and that T in MRD may affect incidence and severity of cGVHD albeit at lower dosing than used in MUD transplants. Our future efforts are focused on incorporating a newly developed exome sequencing-based alloreactivity potential analysis of donor-recipient pairs in guiding optimal use and dosing of T in the Bu-Flu regimen. Disclosure of Interest: None declared. Introduction: Allogeneic hematopoietic cell transplantation (alloHCT) is considered the best option for adults with high risk lymphoblastic leukemia (ALL), being cyclophosphamide plus total body irradiation (TBI) the conditioning regimen most frequently used. Other schemes based only on chemotherapy, principally high-dose combined alkylating therapy, are employed to avoid the toxicity and long-term side effects of TBI. Moreover, the facility to reduce the chemotherapy dose in these regimens makes them attractive, especially for patients with comorbidities, elderly patients and in second transplants. In this retrospective study, we analyzed the non-TBI conditioning regimens for alloHCT in adult patients with ALL in our center. Material (or patients) and methods: Between June 2005 and August 2015, twenty seven adults aged 16-68 years (median 38 y) with ALL underwent first or subsequent alloHCT with alkylating based therapy (Thiotepa-Busulfan-Fludarabine 59%, Busulfan-Fludarabine 22%, others 19%). A total of 12 patients (44%) received reduced-intensity conditioning regimen. Cyclophosphamide was part of graft versus host disease (GVHD) prophylaxis in 57% of cases. The diagnoses were B-ALL (78%) and T-ALL (22%), with 37% of Ph-positive patients. Disease status at alloHCT was first remission (CR1) in 52% and second remission (CR2) or more advanced disease (CR2+) in 48%. DR index was high or very high in 13 patients (48%) and a previous alloHCT was performed in 15% of the cases. Regarding donor type, there were 37% of HLA matched siblings, 52% of haploidentical relatives and 15% of matched or mismatched unrelated donors. The stem cell source was bone marrow graft in 56% of the transplants, while peripheral blood stem cell was used in 44% of the patients. Cord blood transplantations were excluded. Results: Neutrophil engraftment (defined as 40.5 x 10e9/L) was 100% with a median day of 17 (range, 10-28 days). Platelet engraftment (defined as 420 x 10e9/L) was 92% with a median day of 19 (range, . Full chimerism of donor cells was reached by the 84% of survival patients at day +30. With a median follow up of 480 days (range, 21-3499), the overall survival at 16 months was 50% (65% for patients in CR1 vs 32% in CR2+, P = 0.05) and leukemia free survival was 46% (62% for patients in CR1 vs 32% in CR2+, P = 0.05). The cumulative incidence (CI) of relapse was 35% (no differences between CR1 and CR2+). The CI rate of non-relapse mortality (NRM) was 22% (18% at +100 d), 7% for patients in CR1 vs 45% in CR2+. The incidence of acute graft versus host disease (GVHD) grade II-IV was 18.5% as well as 15.8% of patients developed chronic GVHD. At last follow up, fourteen patients (52%) remained alive and free of relapse. Causes of death were relapse (n = 6), GVHD (n = 2), pulmonary bleeding (n = 1), idiopathic encephalopathy (n = 1), idiopathic pneumonitis (n = 1) and bacterial sepsis (n = 1). Conclusion: In our experience, these results suggest that conditioning regimens based on combined alkylating agents for alloHCT in adults with ALL are feasibly and effective, with acceptable rates of NRM and outcomes, being comparable to already published data of TBI-containing regimens. Prospective randomized trials are needed to ensure this affirmation. Disclosure of Interest: None declared. Table 1 . Neutrophil recovery was faster in patients receiving CBV (median 11 days, range 9-17 days) days vs. 14 days, range 10-25 days, P o0.001) as was platelet recovery (median 16 days, range 8-48 days vs. 22 days, range 7-204 days, P = 0.013). Febrile neutropenia rates were similar (83% vs. 77%, P = 0.61). Incidence of grade III/IV mucositis was more in VP/CY/TT group (45% vs. 19%, P = 0.04). No secondary lung or cardiac toxicities were documented in either group. Median follow-up for all patients was 2.5 years. The 3-yr progression free survival (PFS) in the VP/CY/TT and CBV groups was 55% and 45% respectively, P = 0.19. The corresponding 3-yr overall survival (OS) was 81% and 71% respectively, P = 0.59. Secondary malignancies occurred in 9.5% (n = 4) of patients in the VP/CY/TT group and 2.3% (n = 1) of patients in the CBV group, P = 0.17. Introduction: BEAM regimen (BICNU/Carmustine, Etoposide, ARAC, Melphalan) has been used for decades before ASCT for the treatment of various lymphoid malignancies. After worldwide Carmustine shortage, new HDT regimen has to be proposed for these patients, introducing Bendamustine or Thiotepa instead of BICNU, or Busulfan (Bu)-based regimen. Use of another alkylating agent belonging to the nitrosourea family, Lomustine, instead of Carmustine was recently described as LEAM protocol. Material (or patients) and methods: From 06 to 12/15, 18 patients (pts) received ASCT after LEAM in our institution. Median age was 58 y (24-67). Indication for ASCT was Hodgkin (n = 5: 3 primary refractory, 2 in CR2 after early relapse); DLCBCL (n = 6: 3 in CR2; 3 in VGPR/PR after ≥ 1 line treatment (tt); MCL (n = 4: CR1-2) or other lymphoid malignancy (≥ CR2 Median time for neutrophil recovery4500 was 10 d (9-13); 12 d for platelets420 000 (9-55; only 16/17 patients evaluable), 18 d for platelets450 000 for (12-32, 15/17 evaluable). Compliance of Lomustine (an oral drug) is excellent with appropriate antiemetic prophylaxis. We observed 1 toxic death (66 y woman, WHO status = 2 before ASCT; gastric DLCBCL in PR after 3 lines tt): multi-organ failure after grade 4 mucositis, cryptococcus pneumonia and severe, biopsy proven, sinusoidal obstructive syndrome (SOS). Otherwise toxicity was mild: median grade 2 mucositis (4/17 grade ≥ 3), median grade 2 sepsis; no other renal, cardiac, lung or renal toxicity. Follow-up is short (median 3 months (m): no death or unexpected toxicity was observed after 2 m; 2 early relapses occurred at 2 m (primary refractory Hodgkin, blastoid MCL CR2). Conclusion: As BEAM was the most widely used HDT regimen for ASCT in lymphoid malignancies until BICNU was no more easily available, alternative regimen have to be proposed. Bendamustine may be an option as it is active in a lot of lymphoid malignancies and seems to have no cross-reactive toxicity with other drugs used in BEAM regimen: however, French experience of previously described safe BeEAM HDT revealed unexpected toxicities such as acute renal failure and SOS (Damaj G, personal communication SFGM-TC 2015). Thiotepa was mostly used in CNS lymphomas as it can pass the blood-brain barrier: TEAM was recently described retrospectively as a safe alternative to BEAM. Bu-based HDT may raise concern about acute hepatic and long term pulmonary or hematopoietic toxicity; it also may be an impediment to subsequent alloSCT. Although limited, our experience with LEAM regimen in this rather old and heavily treated pts exhibits promising short term safety. Assessment of efficacy requires more patients and longer follow-up. Further recruitment in this pilot study will be presented at the meeting. Introduction: the use of ATG during the conditioning therapy (Cx) has proved to reduce the incidence and severity of chronic GVHD and improve long term quality of life in pts undergoing allo-HSCT, particularly in the unrelated donor PBSC transplant setting. Since ATG is an animal-derived product, infusion reactions (particularly fever) are very frequent in spite of pre-medication, whereas practical management of them is not uniform among the different centers. The aim of this study was to describe the incidence and timing of febrile reactions and analyze the results of blood cultures (BC) performed during the episodes in a series of pts. Material (or patients) and methods: 73 pts consecutively admitted into the hospital for allo-HSCT during a 5-year period (2011-2015) were included. 42 pts were male, and 31 female. Median age was 51 years (17-71). The baseline diseases were: AML (28), MPD (16), NHL/HL (13), MDS (8), BMF/AA (5), MM (2), CLL (1). All but one were allo-HSCT from unrelated donors (98.6%). The stem cell source was PB in 67 cases (91.8%), and BM in 6 (8.2%). HLA matching was 8/10 to 10/10. The pts received Thymoglobulin as a part of the Cx from day -4 or -3. The median daily dose of ATG was 2 mg/kg (1.25-2.75) , and the median number of days with ATG was 3 (1-4). All pts received premedication with steroids, acetaminophen, plus/ less anti-histamines. Fever was considered when pts had ≥ 38°C. Results: 57 pts (78.1%) developed fever after the ATG infusion. During those episodes, 108 BC were performed (1.9 per pt with infusion-related fever). There was one case of false gram +bacteriemia (Coagulasa-negative Staphylococcus), and one case of true bacteriemia (0.92%). The true bacteriemia, due to Escherichia coli, happened in a severely neutropenic patient (ANC: 10/mcL). The bacteria was simultaneously isolated in urine, was multi-sensitive to beta-lactam antibiotics and was succesfully treated. Conclusion: the infusion-related fever after ATG administration was a very frequent event. In our series, all the blood cultures performed during the febrile episodes were negative, with the exception of the case of a severely neutropenic patient. As a conclusion, we suggest that the practice of blood cultures in previously asymptomatic pts who develop the fever after ATG infusion should be decided in an individual basis, rather than being systematic. Disclosure of Interest: None declared. Thymic activity and T-cell repertoire recovery after autoHSCT preceded by myeloablative radiotherapy or chemotherapy M. Głowala-Kosińska 1,* on behalf of Agata Chwieduk, Andrzej Smagur, Wojciech Fidyk, Jacek Najda, Iwona Mitrus, Sebastian Giebel 1 Dept. of Bone Marrow Transplantation and Oncohematology, Cancer Center and Institute of Oncology, Gliwice Branch, Gliwice, Poland Introduction: It was previously postulated, that pretransplant myeloablative treatment may impair thymopoiesis contributing in this way to delayed reconstitution of T cells after hematopoietic stem cell transplantation (HSCT). On the other hand, de novo generation of T cells after HSCT requires competent thymus. Various myeloablative conditioning regimens (TBI or high dose chemotherapy) routinely used in clinical practice may have potentially different impact on the thymus. However, no comparative study on thymic output and T cell repertoire in autoHSCT model has been presented so far. Material (or patients) and methods: Here we evaluated thymic output and TCR diversity in 45 lymphoma patients submitted to autoHSCT differing in respect to conditioning regimen: high-dose chemotherapy as monotherapy (BEAM, n = 22), or combination of total body irradiation with chemotherapy Cy/TBI (n = 23). Thymic output was assessed before and on +100, +180, +365 days after autoHSCT by flow cytometric counts of recent thymic emigrant RTE cells (CD31+ CD62L+ CD45RA+ CD4+) and quantification of signal joint TCR receptor excision circles (sjTRECs) by Q-PCR. T cell repertoire diversity was analyzed on day +365 after autoHSCT by spectratyping of CDR3 region in TCRVb chain. Results: BEAM group, in contrast to Cy/TBI group, manifested significantly higher proportions of RTE cells and sjTREC copy numbers on day +100 and on day +180. Analysis of TCRVb spectratypes on day +365 revealed more restricted (monoclonal or oligoclonal) T cell repertoires in Cy/TBI vs BEAM group (48,8%vs 18,2%, P = .0002). Conclusion: Summing up, the conditioning scheme based on BEAM chemotherapy may be performed with lower risk of thymic destruction and T cell repertoire distortion than Cy/TBI scheme. This finding may help to potentially improve conditioning schemes in order to efficiently perform myeloablation and maintain active thymopoiesis. Disclosure of Interest: None declared. shown wide interindividual variations in the PK parameters. For children with renal failure, high melphalan exposure has been previously (2) related with increased toxicity. We studied the possibility of using a test-dose of melphalan to determine the individual clearance and subsequently, the optimal therapeutic dose for each child with renal failure undergoing high-dose chemotherapy (HDCT) with autologous stem cell transplantation (ASCT). Material (or patients) and methods: A test-dose of melphalan was performed in 13 patients with renal failure undergoing melphalan-based HDCT and ASCT since 2011 at Gustave Roussy. Renal failure was defined by an EDTA clearance o75% or nephrectomy. The test-dose, consisting of 10% of the full scheduled dose of melphalan over a 10 minutes infusion, was administered several days before the HDCT. The sampling time points were: before dosing and 5, 23, 90, 120 and 240 minutes after the end of the infusion. Plasma melphalan concentrations were determined by a validated high performance liquid chromatography method over the range 12.5 to 2000 ng/mL. Melphalan PK parameters were obtained using a population PK model and the target AUC was set at 423 mg/L.min (3) . The patient received the lower dose between the calculated and the theoretical dose. After the full dose, PK was performed to evaluate real melphalan exposure. Toxicity and efficacy were evaluated after the completion of treatment. Results: Eleven patients with neuroblastoma were treated with busulfan-melphalan combination and 2 with nephroblastoma received etoposide-melphalan combination. Median age at ASCT was 3.9 years (0.5-7.1). Before HDCT, 5 patients presented a toxic renal failure, a nephrectomy was performed for 7 patients and 1 was 6 months-old. PK estimations led to reduce doses in 9 patients since the median predicted AUC was 572 mg/L.min . A correlation test between EDTA and melphalan clearances during the test-dose showed that renal function and melphalan clearance were linked (P = 0.0279). After a median time of 10 days (6-56) after a testdose, melphalan full dose was administered according to the theoretical full dose in 4 patients and to the reduced dose calculated according to PK estimations in 9 patients. The median obtained AUC after full dose was 396 mg/L.min (224-956) with a target AUC after full dose (+/-50%) reached for 9 patients. There was no increased toxicity of HDCT in these patients. Conclusion: This PK study allowed us to administer melphalan in patients with renal failure avoiding increased toxicity. These results confirm the useful of a test-dose in these patients to determine the optimal full dose to achieve the targeted exposure based on AUC calculation. Introduction: In vivo T cell depletion with anti-thymocyte globulin (ATG) reduces the incidence of graft-versus-host disease (GVHD) without increasing the risk of relapse in allogeneic stem cell transplantation (allo-SCT) performed with peripheral blood stem cells (PBSC) from matched related (MRD) or unrelated donors after conventional cyclophosphamidebased myeloablative conditioning (MAC) regimens for acute myeloid leukemia (AML). The reduced toxicity Fludarabine and 4 days of intravenous (iv) Busulfan (Flu-ivBu4) MAC regimen is increasingly used for allo-SCT in AML. The impact of the use of ATG in the Flu-ivBu4 MAC is unknown, in particular with MRD. Material (or patients) and methods: To address this question, we conducted a retrospective registry-based study in a population of 566 adult patients who underwent a first HLA identical allo-SCT with a Flu-ivBu4 regimen for AML in CR1 between 2006 and 2013. Among those patients, we compared the outcomes of 145 (26%) who received ATG (ATG group) to 421 (74%) who did not (no-ATG group). Results: Patients in the ATG group were older (median age 49 vs. 44 years, P = 0.002), less frequently seropositive for CMV (73% vs. 87%, P40.0001), had more frequently received PBSC (93% vs. 80%, P = P40.0001), were transplanted more recently (median year 2012 vs. 2011, P40.0001) and had more secondary AML (12% vs. 7%, P = 0.05). The two groups were comparable in terms of time time from diagnosis to transplantation (median 156 days in both groups), gender of recipient and donor, frequency of female donor to male recipient (31% vs. 23%, P = 0.06), and cytogenetic risk (intermediate in 59 vs 54% of patients with available data). While, the cumulative incidence of grade II to IV acute GVHD at day 100 was comparable in the two groups (15% in the ATG group vs. 22% in the no-ATG group, P = 0.10), the use of ATG was associated with reduced 2-year cumulative incidence of chronic GVHD (cGVHD) (31 vs. 52%, respectively, P = 0.0003) including that of extensive cGVHD (8% vs. 26%, respectively, P40.0001). With a median follow up of 23 months in the ATG group and 16 months in the no-ATG group, at 2 years post-transplantation, univariate analysis showed in the ATG vs. no-ATG groups comparable risk of relapse incidence (RI) (22% vs. 27%, P = 0.23), and of non-relapse mortality (NRM) (11% vs 17%, P = 0.15), but improved GVHD and relapse free-survival (GRFS) (57% vs. 33%, P40.0001), leukemia free survival (LFS) (67% vs 55%, P = 0.04) and overall survival (OS) (72% vs 59%, P = 0.04). In multivariate analysis, the use of ATG was associated with reduced incidence of cGVHD (HR = 0.46, P = 0.0001), improved GRFS (HR = 0.48, P = 0.00001), a trend for reduced NRM (HR = 0.59, P = 0.09), improved LFS (HR = 0.67, P = 0.027) and OS (HR = 0.65, P = 0.027) without affecting RI (HR = 0.72, P = 0.15). Age450 years had an unfavorable impact on NRM (P = 0.012), LFS (P = 0.004) and OS (P = 0.026) while female donor to male recipient increased the risk of cGVHD (P = 0.001) and reduced GRFS (P = 0.002). Conclusion: In conclusion, our data suggest that the use of ATG in association to the Flu-ivBu4 MAC regimen for AML patients allografted in CR1 with a sibling donor favorably impacts the final outcome of allo-SCT by reducing the risk of severe cGVHD without affecting the risk of relapse. These findings should be confirmed in well design 2 arm randomized studies. Disclosure of Interest: None declared. Introduction: Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare form of acute leukemia associated with an overall bad prognosis. Only very few cases have been reported to reach durable remissions thanks to chemotherapy alone. Allogeneic hematopoietic stem cell transplantation (HSCT) using a myelo-ablative conditioning regimen (MAC) has been reported to be the gold standard treatment for BPDCN. However, little is known about the place of reduced-intensity/ non-myelo-ablative conditioning regimens (RIC/NMA) in this setting. Material (or patients) and methods: We retrospectively collected from the database of the French Society of Bone Marrow Transplantation and Cell Therapy (SFGM-TC) all cases of BPDCN treated with allogeneic HSCT. Immunophenotypes at diagnosis were centrally reviewed in order to confirm diagnosis according to the Garnache-Ottou diagnostic criteria. Twenty-eight patients had a diagnostic score of 2 or more. The remaining 15 patients all had CD4+ CD56+ disease, but as they were mostly diagnosed before publication of this score, other markers (such as CD123, BDCA-2 and BDCA-4) were not performed routinely at that time, precluding calculation of a score at least equal to 2. Results: From February 2003 to January 2014, 43 patients with BPDCN received an allogeneic HSCT in 21 French centers. Median age was 57. About 80% of the patients were transplanted in CR1. Donors were matched siblings in 42% of cases. Peripheral blood was the main hematopoietic stem cell source used in this series (70% of cases). Conditioning regimens were MAC in 18 cases and RIC/NMA in 25 cases. Four patients (9%) had engraftment failure or secondary graft rejection, 3 of whom having received cord blood units. All these 4 patients were transplanted again 2 to 17 weeks after the first transplant. After a mean follow-up of 668 days for the entire cohort (1050 days for alive patients), 22 patients (51.2%) were alive, 19 of whom being disease-free (44.2%). Eleven patients had relapsed, at a median of 225 days post-HSCT (range: 74-821 days). Two-year cumulative incidences of relapse (CIR) and non-relapse mortality (NRM) were 25.5% (95% CI = [0.13-0.40] ) and 32.8% (95% CI = [0.186-0.479]) respectively. At 2 years post-transplant, disease-free survival (DFS) and overall survival (OS) were 44.9% (95% CI = [0.291-0.595]) and 52.2% (95% CI = [0.357-0.664]), respectively. Even though not statistically significant, patients receiving a MAC (n = 18) were less likely to relapse than patients receiving RIC/ NMA (2-year CIR = 7.1% and 36% respectively, P = 0.137), but had a higher NRM rate (43.9% versus 26% at 2 years, P = 0.419), resulting in similar 2-year DFS and OS (57.1% versus 38%, P = 0.511 and 57.1% versus 49.7%, P = 0.91). There was a trend for a lower incidence of NRM at 2 years in patients transplanted from a sibling donor versus others (16.7% and 39.9% respectively, P = 0.0505), but donor source had no effect on CIR (P = 0.826), DFS (P = 0.194) and OS (P = 0.188). Conclusion: In this series of 43 patients with BPDCN, allogeneic HSCT was associated with a good disease control, but NRM was high. In this regard, transplantation from a sibling donor appears to be the best option. RIC/NMA are feasible and may also reduce the incidence of NRM, but at the expense of a higher incidence of relapse. Disclosure of Interest: None declared. Introduction: Allogeneic stem cell transplantation (SCT) is a curative option for adult patients with high-risk AML and MDS. However, the optimal conditioning regimen especially for elderly patients has not yet been defined. We performed a retrospective single center analysis for three RIC regimens according to the FLAMSA-protocol (Schmid et al. JCO 2005) with either (1) Busilvex (FLAMSA-Bu) or (2) 4 Gy total body irradiation (FLAMSA-TBI) or (3) the BFM protocol (Marks et al. Blood 2008) . Material (or patients) and methods: 183 patients with high risk AML (n = 156) or MDS (n = 27) transplanted between 1/2007 and 12/2014 were included in this study. AML-patients were not (n = 52), beyond (n = 53) or in 1. CR (n = 51). Patients were stratified as high-risk based on clinical course, cytogenetics, molecular genetics, and/or IPSS score, respectively. Cohorts were analyzed for comorbidities according to the HCT-CI (Sorror et al. Blood 2007) and for the EBMT score (Gratwohl, BMT 2012). With a median follow-up of 1.3 (0-8.4) years overall survival (OS), disease-free survival (DFS), relapse rate (RR), treatment related mortality (TRM) as well as incidence of aGvHD and cGvHD were analyzed using Kaplan-Meier statistics. Results: 79, 37 and 67 patients were treated in groups (1), (2), and (3), respectively. There were more AML (81%, 78%, 94%) and fewer MDS patients (19%, 22%, 6%) in group (3) (P = 0.04). 59%, 79%, and 70% of AML patients were not or beyond 1. CR (P = 0.14). Median age was 58 (19-71), 57 (22-67), and 63 (24-74) (P = 0.51), and 41%, 35%, and 46% were female (P = 0.53). Stem cell source were PBSC in 95%, 95%, and 88% from MRD (16%, 19%, 21%), MUD (66%, 41%, 54%), and MMUD (18%, 41%, 25%) (P = 0.07). Median CD34+ cell numbers/kg for 169 PBSC grafts were 6.7x10 6 , 5.7x10 6 , and 6.5x10 6 (P = 0.34). All patients received CSA and MMF as post-transplant GvHD-prophylaxis, and all but three were treated with ATG/Thymoglobulin. The mean HCT-CI score was higher in group (3) (2.0, 1.6, 3.2) (P = 0.02), whereas the mean EBMT-score was balanced (4.2, 4.5, 4.5) (P = 0.90). OS was 71%, 60% and 59% at 12 months, and 59%, 47% and 50% at 24 months (P = 0.27). DFS was 66%, 47%, and 57% at 12 months, and 57%, 44% and 46% at 24 months (P = 0.16). RR was 12.4%, 31.5% and 12.4% at 12 months, and 18.1%, 35.3% and 25.2% at 24 months (P = 0.17). TRM was 23.5%, 28.9% and 30.6% at 12 months, and 28.5%, 28.9% and 36.8% at 24 months (P = 0.58). aGvHD III-IV°occurred in 24%, 24% and 11% of patients (P = 0.15), and extensive cGvHD in 9%, 15% and 14% (P = 0.29). Conclusion: Our data suggest a trend toward a better OS and DFS for cohort (1) as compared to cohorts (2) and (3). RR was highest in group (2) whereas TRM was similar in all three cohorts. The incidence of III-IV°aGvHD and extensive cGvHD was balanced. The different OS and DFS between cohorts (1) and (3) may be due to different risk profiles with a significant higher ratio of AML patients and a higher mean HCT-CI in group (3). In contrast, since patients risk profiles were similar in both cohorts, FLAMSA-Bu seems superior to FLAMSA-TBI. In conclusion, different RIC regimens are available for high-risk AML and MDS patients but FLAMSA-TBI seems to be less effective than FLAMSA-Bu. Disclosure of Interest: None declared. Introduction: Early immune reconstitution has been associated with improved survival in full intensity T replete allogeneic stem cell transplants.Reduced intensity Alemtuzumab conditioned allogeneic stem cell transplants have enabled older patients to undergo a potentially curative procedure for the treatment of high risk myeloid malignancies. Material (or patients) and methods: We have retrospectively analysed 50 patients with high risk MDS (n = 14) and high risk AML (n = 36) who underwent a reduced intensity Alemtuzumab based conditioning in our transplant centre between 2006 and 2014.The median age was 62.03 years (range 51-72) and there were 33 male and 17 female recipients. All donors were fully matched at HLA-A,B,C,DR,DQ loci.Eleven of them had a fully matched sibling donor while 39 received a graft from a fully matched unrelated donor. 29 patients were conditioned with fludarabine, melphalan and alemtuzumab while 21 patients were conditioned with fludarabine, busulphan and alemtuzumab. Sibling donor recipients received 30 mg of alemtuzumab on D-1 while unrelated donor recipients received 25 mg of Alemtuzumab on D-2 and D-1. Median follow up of the patients was 18.2 months (range 15-90).Immune reconstitution was considered when the recipients reached a CD4 count of above 200. Results: The median overall survival was 27.4 months while the median progression free survival was 33 months.CD4 count of4200 was achieved in 26 patients by 12 months (52%) while the rest of the patients have not achieved full immune reconstitution.Those patients who had reconstituted their immune system were 70% likely to be alive at 2 years compared to 25% for those patients without immune reconstitution.14 patients (28%) developed grade 1-2 GvHD while only 1 (2%) grade 3-4 GvHD. 8 patients received DLI for mixed T cell chimerism (median 2 doses). Conclusion: Immune reconstitution at 12 months is a critical determinant of survival in patients undergoing reduced intensity Alemtuzumab based conditioning stem cell transplants for high risk MDS and AML.Larger prospective studies are needed to validate the results of our single centre experience and also review the impact of strategies to improve immune reconstitution. Disclosure of Interest: None declared. Introduction: Reduced intensity Alemtuzumab conditioned allogeneic stem cell transplants have enabled the successful delivery of a potent graft versus leukaemia effect while GvHD rates remained low contributing to an acceptable non relapse mortality for those patients. Material (or patients) and methods: Fifty patients with high risk MDS (n = 14) and high risk AML (n = 36) who underwent a reduced intensity Alemtuzumab based conditioning between 2006 and 2014 in our transplant centre were analysed.The median age was 62.03 years (range 51-72) and there were 33 male and 17 female recipients.Eleven of them had a fully matched sibling donor while 39 received a graft from a fully matched unrelated donor. Twenty nine patients were conditioned with fludarabine, melphalan and alemtuzumab while 21 patients were conditioned with fludarabine, busulphan and alemtuzumab. Median follow up of the patients was 18.2 months (range 15-90). Results: The median OS was 27.4 months while the median PFS was 33 months.At 3 months there were 36 patients with full donor (FD) whole blood chimerism and 14 with mixed chimerism (MD) while at 12 months there were 37 patients with FD chimerism and 13 with MD chimerism. Fourteen patients (28%) developed grade 1-2 GvHD while only 1 (2%) grade 3-4 GvHD. Eight patients received escalated DLI for mixed T cell chimerism (median 2 doses). Patients with FD chimerism in whole blood at 12 months had a 90% 2 year progression free survival compared to 60% at 2 years for those patients with mixed chimerism. T cell chimerism at 3 months and 12 months post transplant did not influence OS and PFS. Conclusion: Whole blood chimerism in bone marrow at 3 months and 12 months post transplant can safely predict S242 progression free survival in reduced intensity allogeneic stem cell transplant recipients with alemtuzumab based conditioning.Patients with high risk MDS and AML achieving full donor chimerism have a significantly better progression free survival. Larger prospective studies are needed to validate the results of our single centre experience. Disclosure of Interest: None declared. Introduction: 5-azacitidine (5-AZA) is commonly used in case of relapse of myeloid disease after allo-SCT. Data suggest that this drug may increase graft versus leukemia (GVL) effect; however, few clinical data are available on feasibility of 5-AZA as prevention of relapse. Material (or patients) and methods: We retrospectively analyzed 13 patients transplanted between 2012 and 2015, for AML (n = 11), MDS (n = 1), MPD (n = 1). All patients had high risk disease (second allo-SCT for previous relapse: n = 3, secondary disease: n = 8, complex or monosomal karyotype: n = 4, FLT3/ITD: n = 2). Disease status at transplant was: CR1: n = 6, CR2: n = 3, refractory or upfront: n = 4. Four patients had a high HTC-CI score (≥ 3) at transplant. Type of donor were: HLAidentical: n = 4, haplo-identical: n = 2, UD: n = 7. Two patients received RIC regimen, 3 patients had a MAC regimen and 8 patients received a reduced toxicity regimen (5 patients had a sequential regimen). Median age at transplant was 55 years (range 31-68) and median follow up is 14 months (range 8-31). Results: All patients achieved neutrophil engraftment in a median time of 16 day. Five patients had aGVHD (grade 1: n = 2, grade 2: n = 3), before 5-AZA, requiring systemic treatment for 2 of them. Median time for 5-AZA administration was: 138 (range 103-269) days. 5-AZA was administered as prophylactic treatment in 9 patients (cytological and molecular remission) (2 of them had a mixed chimerism at day+100) or a preemptive therapy in 4 cases (cytological remission but progressive increase of minimal residual disease). The 2 patients with a previous history of aGVHD treated with systemic therapy, received 5-AZA as preemptive treatment and had before starting 5-AZA a cGVHD that did not worsen during 5-AZA. Median number of 5-AZA courses was 3 (range 1 to 9 cycles). Patients received 5-AZA either at 32 mg/m 2 during 5 days (n = 12) or 75 mg/m 2 during 5 days (n = 1). 5-AZA was scheduled every 4 weeks, except in 6 patients requiring 6 weeks of interval due to prolonged neutropenia (n = 4), GVHD (n = 1) or exploration of hypereosinophilia (n = 1). No febrile neutropenia or cytopenia requiring transfusion were reported during 5-AZA administration. Six patients received DLI as an additional strategy towards decreasing the risk of relapse. Five patients developed cGVHD during or after 5-AZA therapy, with a median delay after starting 5-AZA of 119 days (range 32-210). Extensive cGVHD occurred for 3 patients who all received prior DLI. All patients receiving 5-AZA as prophylactic therapy are still in molecular CR. Of the 4 patients starting 5-AZA as preemptive, 2 relapsed after 1 and 3 courses of 5-AZA, one obtained a molecular remission after 2 courses of 5-AZA and one has not been yet evaluated. The 2 patients, who had a mixed chimerism, achieved full donor chimerism after 6 and 7 courses of 5-AZA. At last follow up, 1 patient died of disease recurrence, 6 patients are still under 5-AZA treatment and 7 stopped the treatment (relapse (n = 2), GVHD (n = 2), prolonged neutropenia (n = 1), patient choice (n = 1), efficacy with persistent full donor chimerism (n = 1)). Conclusion: Early administration of 5-AZA after allo-SCT for relapse prevention is feasible and safe, including when combined with DLI. A treatment schedule of 32 mg/m 2 during 5 days is associated to increase of immunomodulating effect with limited toxicity. Disclosure of Interest: None declared. Introduction: Haplo-SCT is an attractive option in which virtually 100% of patients (pts) would have a suitable donor. In the first years T cell depletion was employed to manage acceptable rates of GVHD and rejection. New GVHD prophylaxis strategies as PTCy have made possible unmanipulated Haplo-SCT with excellent clinical results. Tregs play an important role in GVHD physiopathology. In murine models PTCy seems to preserve Tregs in the recipient. This fact may be part of its protective effect against GVHD.This study prospectively analyse T regs reconstitution post T cell repleted Haplo-SCT with PTCy comparing with HLA identical sibling donor SCT (HLAid SCT),and its relationship with GVHD. Material (or patients) and methods: Between Nov.2012-Sep.2014,30 pts with Haplo SCT and 8 pts with HLAid SCT in Gregorio Marañon Hospital were included. Peripheral blood graft was used. Conditioning regimen comprised fludarabine, cyclophosphamide,busulfan for Haplo and Flu/Bux or Flu/Mel for HLAid SCT.GVHD prophylaxis was high dose PTCy(+3+4), cyclosporine A and mycophenolatemofetil for Haplo-SCT and cyclosporine A/Methotrexate for HLAid-SCT. Table1 shows pts characteristics,with no differences between the two groups except for longer engraftment and lower acute GVHD incidence in Haplo-SCT group. Tregs(CD127-/w, CD25++, CD4+) were analysed by flow cytometry on FC500/Navios(Beckman Coulters) on +15, +30, +60, +90, +180, +360 days postransplant. For comparison Mann-Whitney U-test was used. Results: Haplo pts had lower T CD4+ cells on +30,+90,+180 and +360 than HLAid pts, with statistically significant differences on +30 and +90(+30:median 40(17-73) vs 514(327-732) P 0,00; +90:181(91-212) vs 258 (205-332) P 0,02). Tregs were also lower in Haplo group in the first month(7 (4-12) vs 29 (17-38) P0,00), but showed a fast reconstitution from this time on. At 3 months(+90) both groups already had similar Tregs counts (20(12-34) in Haplo vs 19(14-22) in HLAid p0,7), although total CD4+ cells were lower in the Haplo group at the same time point. T regs remained similar in both groups on+180 (27 (12-36) in Haplo vs 23,00 in HLAid p0,4) and +360 (23(19-30) in Haplo vs 23(19-51) in HLAid P0,9). In Haplo group, comparing pts with acute GVHD(21) and without acute GVHD(9), Tregs were better in pts who didn´t develop acute GVHD on +30(16(12-29) vs 6(4-13) p0,04), with no differences in total CD4+T cells at that time point (30(19-82) . According to our protocol, immunosupression was reduced promptly and only one patient relapsed. Interestingly, in this cohort, subsequently six patients experienced low level mixed chimerism (LLMC), reflecting the increased sensitivity of both methods, ii) 15 patients in 32 samples 4+3 months post-alloHCT, experienced mixed chimerism with negative FC-MRD(FC-/STR+) and no intervention was applied due to chimerism 4 = 97.5%. Five out of 15 patients had previously been FC+/STR-, as described before. None of the remaining 10 FC-/STR+ patients relapsed. Progression Free Survival (PFS) and Overall Survival (OS) did not differ in the two groups (FC+/ STR-,FC-/STR+) compared to the entire cohort. Similarly, no differences were found in PFS and OS when each method was assessed separately. However, patients experiencing concurrently FC and STR-PCR positive MRD showed a trend for shorter PFS and OS [PFS: mean 43 (95% CI: 21-65) vs 53(21-65) months, P = ns and OS: mean 50 (95% CI: 27-74) vs 60(54-65) months P = 0.08, medians not reached for FC+/STR+ vs other, respectively]. Conclusion: To conclude:i) positive MRD by FC is frequently followed by LLMC by STR-PCR, reflecting the high sensitivity of both methods and leading to prompt therapeutic interventions ii) concurrent detection of MRD by FC and STR-PCR indicates a worse prognosis. In this cohort of patients, 6-color FC was not superior to STR-PCR in terms of DFS and OS. This could be explained by the low incidence of relapse in our population following our immunotherapeutic interventions. Larger scale studies are warranted to evaluate the prognostic value of each method and answer the arising question of the benefit from the concurrent diagnostic approach. References: 1. Minimal residual disease after allogeneic stem cell transplant: a comparison among multiparametric flow cytometry, Wilms tumor 1 expression and chimerism status in acute leukemia. Rossi G, et al. Leuk Lymphoma. 2013. 2 . Impact of minimal residual disease, detected by flow cytometry, on outcome of myeloablative hematopoietic cell transplantation for acute lymphoblastic leukemia. Bar M, et al. Leuk Res Treatment. 2014 . Disclosure of Interest: None declared. The value of monitoring minimal residual disease in children with MLL-rearranged acute leukemia after hematopoietic stem cell transplantation B. Gruhn 1,* , S. Wittig 1 , C. Woehlecke 1 1 Department of Pediatrics, Jena University Hospital, Jena, Germany Introduction: Children with MLL-rearranged acute leukemia have a bad prognosis and are thus often eligible for hematopoietic stem cell transplantation (HSCT). We monitored minimal residual disease (MRD) by use of the MLL rearrangement after transplantation and adjusted our therapy depending on the results. Material (or patients) and methods: DNA was isolated from patient's peripheral blood (PB) or bone marrow (BM) mononuclear cells at initial diagnosis and patient specific MLL fusion gene sequences were established. Leukemic cell specific primers/probes were created which enabled us to monitor MRD with a quantitative real-time PCR. We investigated six patients. Each of them exhibited a different kind of MLL fusion gene. Results: The first patient with MLL-MLLT10/AF10-rearranged AML turned MRD negative throughout chemotherapy and could be transplanted in complete hematological and molecular remission. Complete chimerism (CC) of the donor was achieved and she stayed MRD negative in all controls. The second patient was transplanted because of the evidence of a MLL-SEPT6 fusion gene. He did not reach molecular remission before transplantation. However, he turned MRD negative right after transplantation and showed a CC posttransplant. The third patient with MLL-ELL-rearranged AML suffered from a third relapse after a second HSCT. We started a therapy with sorafenib and gave additional donor lymphocyte infusions (DLI). MRD monitored by MLL-ELL fusion gene decreased continually until it reached the limit of detection after two months. CC was already reached one month earlier. Two years later sorafenib was discontinued and MRD is still not detectable two years after withdrawal of sorafenib. Patient 4 developed a secondary AML with MLL-MLLT3/AF9 rearrangement after being treated for relapsed pre-B-ALL. Even though she did not reach hematological remission before transplantation, MRD decreased continuously and became negative after HSCT. Despite additional DLI, MRD levels began to increase again. We started a therapy with sorafenib and MRD decreased until it reached the limit of detection as shown in the Figure. Interestingly, donor chimerism was 100% at every time point. Patient 5 did not reach MRD negativity before HSCT due to relapsed MLL-MLLT1/ENL-rearranged AML. Unfortunately he relapsed and developed additional myeloid sarcoma. After a second HSCT, MRD could be reduced by DLI, but he finally died of myeloid sarcoma. Patient 6 was transplanted due to infant MLL-AFF1/AF4rearranged ALL and presented with decreasing donor chimerism eight years after transplantation. However, MRD measured by MLL-AFF1/AF4 was not detectable and the patient stayed in molecular remission without immunological intervention. Conclusion: In patients with high risk leukemia monitoring of MRD after HSCT is essential for the detection of impending molecular and hematological relapse. As MLL fusion genes are highly specific for the leukemic cell, they represent an ideal S244 target for it. With a quantitative real-time PCR one malignant cell can be detected in 10 5 normal cells. Both patient's BM and PB are convenient and should be examined in regular intervals to evaluate remission status or therapy success. Thus MRD can be decision guidance on whether additional immunotherapy should be administered or can be omitted. Concerning sensitivity and specificity MLL showed to be superior to chimerism analysis. Disclosure of Interest: None declared. Introduction: With the introduction of pediatric-inspired induction regimen, prognosis of adult acute lymphoblastic leukemia (ALL) has substantially improved, leading to an increased assessment of minimal residual disease (MRD) during all induction course and a progressive use of MRDdriven therapeutic approach. In most modern pediatric protocols, persistence of MRD during the first courses of induction therapy is an adverse prognostic factor and constitutes a strong indication for early allogeneic bone marrow transplantation (BMT). However, it remains unclear if an effort towards MRD eradication should be attempted before BMT, as the indication for intensification chemotherapy differs between different protocols and relatively few data are available on outcome of MRD positive adult ALL patients. The aim of this study was to analyze the role of pre-BMT MRD assessment as predictor for the post-transplant relapse risk. Material (or patients) and methods: We retrospectively analyzed the outcome of 53 consecutive ALL patients receiving allo-BMT, with available pre-BMT multicolour flow cytometry (MFC) MRD assessment. Median age at transplant was 30 years. Disease phase was CR1 in 20 (38%), CR2 in 17 (32%), and CR3 in 14 patients (30%). Thirty-five patients (66%) had B lineage ALL, whereas 18 (34%) had T-ALL. Relapse-free survival (RFS) was calculated from the time of transplantation until last follow-up or documented leukemic relapse. A positive MFC MRD was defined by the presence of no less than 25 clustered leukemic cells/10 5 total events (threshold of 2.5x10 -4 residual leukemic cells) at four-color flow-cytometry. Results: Relapse occurred in 30 patients (57%). Three-year RFS was 42.2% (median 10 months). The probability of relapse was significantly affected by disease status at BMT (higher in patients transplanted in CR2 or CR3, P40.001), karyotype (higher in high risk patients, Po0.03) and MFC MRD ( Introduction: Allogeneic bone marrow transplantation (BMT) offers the greatest chance of cure for patients with high-risk acute myeloid leukemia (AML). Persistence of disease or high levels of pre BMT minimal residual disease (MRD) have been reported to predict relapse risk after BMT. WT1 expression levels and multicolor flow cytometry (MFC) are the most common tools to evaluate MRD. We reported that combining WT1 expression and MFC for MRD detection after induction therapy affects relapse risk in AML. The aim of this study was to analyze the role of combined MFC/WT1 pre-BMT MRD assessment as predictor for the post-transplant relapse risk. Material (or patients) and methods: We retrospectively analyzed the outcome of 224 consecutive AML patients receiving allo-BMT in 1st or 2nd complete remission (CR). Pre-BMT marrow samples were analysed for WT1 expression and MFC as MRD evaluation. Median age at transplant was 45 years. Disease phase was CR1 in 161 and CR2 in 63 patients; 163 and 61 patients received mieloablative and reduced intensity conditioning, respectively. Median follow-up was 64 months. Relapse-free survival (RFS) was calculated from the time of transplantation until last follow-up or documented leukemic relapse. A positive MFC MRD was defined by the presence of no less than 25 clustered leukemic cells/10 5 total events at four-color flow-cytometry. Real-time PCR for WT1 was performed on DNA Engine 2 (Opticons, MJ Researchs). WT1 copy number/Abl copy number 500x10 4 was used as cutoff value for abnormal WT1 expression. Results: Relapse occurred in 63 patients (28%). Three-year RFS was 68.8% (median not reached). The probability of relapse was significantly affected by the occurrence of acute GVHD (grade 4 = 2 versus 0-1, Po 0.03) and MRD status before transplantation, measured with any method (Po0.001 for WT1-based MRD, P40.05 for MFC based MRD, P40.0001 for combined MRD). Multivariate RFS analysis revealed that the combined MRD evaluation was the only independent predictor of RFS (P o0.001). MFC-MRD was the strongest predictor of longer RFS since only two relapses occurred in the 24 MFC-MRD negative patients and 3-years RFS was 89.9%. Among MFC-MRD positive patients, WT1 MRD status further stratified the risk of relapse as the 3-years RFS was 73.3% in MFC MRD pos/WT1 MRD neg patients and 44.4% in MFC MRD pos/WT1 MRD pos patients (median not reached and 61 months, respectively, P o0.01, Fig. 1 ). Multivariate overall survival analysis confirmed that combined MRD evaluation was the only independent predictor of OS (P o0.001). Conclusion: Pre transplant MRD evaluation by WT1 and MFC on bone marrow samples is a widely applicable and reliable predictor of relapse risk. Patients with both negative pre-BMT MRD markers have a significantly longer RFS, while patients with both positive MRD markers display an higher risk of relapse. Identifying patients who have an higher risk of relapse could open the way to apply pre-emptive therapeutic strategies to prevent AML relapse, from donor lymphocyte infusion to other innovative approaches. Introduction: Natural Killer (NK) cells are important components of the innate immune system, and play a relevant role in defence against tumors and viruses. In the setting of haploidentical hematopoietic stem cell transplantation (Haplo-HSCT), donor NK cells may be "alloreactive" by means of KIR/HLA mismatch in graft versus host (GvH) direction. In T-depleted Haplo-HSCT, NK alloreactivity has been associated with a higher disease free survival (DFS) rates for transplanted patients, contributing to graft-versus leukemia (GvL) effect. Moreover, donors having KIR B haplotypes (characterized by the presence of more activating KIR) or expressing KIR2DS1 correlated with a better clinical outcome of transplantation. Material (or patients) and methods: We analyzed NKalloreactivity in the setting of unmanipulated Haplo-HSCT with high dose post-transplant cyclophosphamide (PT-CY) for patients affected by acute myeloid leukemia or myelodisplastic syndromes. 101 consecutive patients transplanted from September, 2010 to October, 2014 were enrolled. Donors and patients were studied for HLA-class I alleles and related KIRligands, and donors for KIR gene profile. NK cell phenotype of transplanted patients was analyzed. Results: Disease status at HSCT was the most relevant factor affecting patients' outcome (P o0,0001), with 3 y 78% OS and 76% DFS rates for "early" patients (CR1+CR2, n = 61) versus 33% OS and 28% EFS rates for "advanced" patients (CR3 or active disease, n = 40). Neither NK alloreactivity nor the presence of donor KIR B/x genotype nor the presence of donor KIR2DS1 seemed to play a role in preventing leukemia relapse. NK alloreactive patients had DFS rate similar to non-NK alloreactive group (62% vs 59%, P = 0,47) with a better, still S246 non-significant trend in overall survival (72% vs 60%, P = 0,14) for NK-alloreactive patients. Similarly, Haplo-HSCT from donors with KIR B/x genotype (with B content score ≥2) or who had KIR2DS1 was not associated with better outcome (P = 0,67 and P = 0,89, respectively). We observed a high expression of CD56 and inhibitory CD94/NKG2A, while low KIRs, on NK cells from transplanted patients, indicating an immature phenotype even at several months after transplant. Conclusion: The different immunosuppressive approach of unmanipulated Haplo-HSCT compared to T-depleted Haplo-HSCT can inhibit NK cell function. Indeed, cyclosporine from early transplant days has been shown to interact and possibly inhibit NK cells in vivo. Post-transplant cyclophosphamide is known to selectively kill activated T-cell and inducing longterm tolerance, but it can also affect NK cells. Further studies are needed to better understand the complexity of this intriguing issue, leading to a more complete definition of NK cell functions in this Haplo-HSCT setting. Disclosure of Interest: None declared. Introduction: Despite allogeneic hematopoietic stem cell transplantation (HSCT), prognosis of fms-like tyrosine kinase 3 (FLT3) mutated acute myeloid leukemia (AML) remains poor due to the high risk of relapse. Sorafenib, a multikinase inhibitor active on FLT3, has shown encouraging results in these patients. Material (or patients) and methods: Here, we report the use of sorafenib after HSCT in 19 adults with FLT3 positive AML treated in two hematologic departments. Results: Patients (pts) underwent their first HSCT between 2012-2015. Median age at HSCT was 48 years (range 32-60). At time of HSCT, all but 2 pts were in complete remission (CR; CR1, n = 15; CR2 = 2). Median interval from CR to HSCT was 82 days (d). The majority of the pts (n = 17) received peripheral blood as stem cell source, 10 from a matched sibling, 3 from a matched unrelated, 3 from a haploidentical donor. Two other patients had a haploidentical bone marrow and one had a cord blood unit. Conditioning regimen was myeloablative in 10 pts, while reduced intensity regimen was used in 6 pts, with 3 receiving a sequential regimen. All pts achieved neutrophil engraftment in a median of 15 d, with all but 1 pt having full donor chimerism at day 30. Sorafenib as a single agent was introduced at a median time after HSCT of 104 d (range 29-406). Sixteen patients were out of immunosuppressive treatment (IST) without graft-versus host diseases (GVHD) at time of sorafenib start. Sorafenib was started at a dose of 400 mg twice a d (n = 9), 200 mg twice a d (n = 8) or 200 mg once a d (n = 2). Reason for starting sorafenib was relapse in one pt at d 62 (associated with chemotherapy), detection of minimal residual disease positivity (n = 5) or prophylaxis (n = 13, including 1 pt who received sorafenib for relapse before HSCT in combination with the induction regimen, obtaining CR before HSCT). Median duration of treatment was 124 d (range 10-1007). Dose reduction or withdrawal due to toxicities were needed in 4 and 4 pts, respectively. These included gastrointestinal (GI, n = 3), cardiac (n = 1), skin (n = 1, with GI), biochemical (n = 1) and hematological toxicities (n = 4, 1 with GI). Despite dose reduction, persistence of toxicities prompted to treatment withdrawal in 3 pts. One pt experienced relapse 1 month after sorafenib withdrawal. Seven pts experienced GVHD (grade I, n = 1 grade II, n = 6) (biopsy proven in 3 pts), resulting in dose reduction in 3 pts, and withdrawal in 1 pt. Six pts required systemic IST. Dose reduction was made in 1 pt due to financial reasons. One pt discontinued sorafenib due to disease relapse 14 days after introduction. With a median follow-up of 14 months (range 5-37), all but 1 pt are alive in CR with full donor chimerism. Salvage regimen prompted CR in 1 out of the 2 relapsing pts after sorafenib introduction. Four pts received donor lymphocyte infusion (DLI). Sorafenib treatment is ongoing in 9 pts (with 4 at reduced doses) with a median of 566 days. Conclusion: Our findings indicate that sorafenib might be used in post-HSCT with dose individualization according to patient tolerability. Sorafenib is effective after HSCT for FLT3+ AML, with 9 out of 19 patients still on therapy. Further analysis is needed to evaluate the immunomodulating role of sorafenib post HSCT. Our data support prospective controlled trials of sorafenib after HSCT. Disclosure of Interest: None declared. Introduction: Allogeneic hematopoietic stem cell transplantation (HSCT) is the standard therapy for patients with various malignant or non-malignant hematologic disorders. Several factors may affect transplantation success and chimerism monitoring could improve the management of transplant procedure. The aim of this study was to analyze chimerism, evaluate engraftment status and predict the outcome of patients with allogeneic HSCT by using real time qPCR targeting insertion/deletion polimorphisms (indel-qPCR). Material (or patients) and methods: Pre and post-transplant peripheral blood samples were collected from 22 patients who underwent allogeneic-HSCT with myeloablative or RIC regimen. Pre-transplant samples both for donors and patients were evaluated by using the STR (Mentype DIPscreen; Biotype, Dresden, Germany) for identification of specific alleles. Minimum 2 different alleles were selected for chimerism monitoring on post-transplant months 1, 2, 3 and 6. After transplantation, qPCR (Mentype, DIPquant; Biotype, Dresden, Germany) were performed by using selected alleles and quantification was done by using a specific software (Mentype, Chimera; chimerism software; Germany) for engraftment monitoring and predict the outcomes, following manufacturers' recommendations. Results: Sixteen patients with full donor chimerism (DC) had stable engraftment and they still survive free from leukemia or lymphoma. Six patients had unstable mixed donor chimerism (DC: 11% -77%), and 3 of them relapsed after allo-HSCT and 3 patients died. Conclusion: Decrease of donor chimerism appeared prior to graft rejection and disease relapse. The incidence of GVHD was higher in full donor chimerism group. This novel qPCR is a simple and accurate technique that shows very good concordance and provides higher rates of informative loci per patient. We conclude that the qPCR chimerism is a reliable and sensitive assay for engraftment monitoring and predicting the outcome. Prognostic impact of minimal residual disease measured by multiparametric flow cytometry in 80 allogeneic haematopoietic stem cell-transplantated children with acute lymphoblastic leukaemia I. Elorza 1,* , C. Palacio 2 , L. Gallur 2 , S. Gallego 1 , J. Sánchez de Toledo 1 , C. Díaz de Heredia 1 1 Pediatric Oncology and Hematology, 2 Hematology, Clinical Laboratories, Hospital Vall d´Hebron, Barcelona, Spain Introduction: Outcomes with allogeneic haematopoietic stem cell transplantation (allo-HST) are better than with chemotherapy in children with high-risk acute lymphoblastic leukaemia (ALL). The main drawback to successful transplant is relapse. The major prognostic factor for long-term relapse-free survival (RFS) is complete morphological remission prior to transplant. Multiparametric flow cytometry (MFC) is widely used to detect anomalous immunophenotypes in the diagnostic work-up of ALL and its monitoring throughout treatment. This study aimed to ascertain whether a relationship exists between MRD prior to allo-HST in children with ALL measured by MFC and outcome, assessed as RFS and overall survival (OS). Furthermore, other pre-and post-transplant factors associated with survival were studied. Material (or patients) and methods: MRD was quantified by MFC prior to allo-HST in 80 children with ALL (age range: 6 months-19 years). According to the MRD level detected, patients were divided into two groups: MRD-positive (n = 25) with blast cells ≥ 0.01% compared with total cell population, and MRD-negative (n = 55) with blast cell o0.01%. Results: RFS at 3 years post-transplant was 72%, with OS 51%. RFS in the MRD-positive group was 50% versus 80% in the MRD-negative group (Log Rank 9.5; P = 0.002). OS in the MRDpositive group was 30% versus 59% in the MRD-negative group (Log Rank 6.5; P = 0.01). Bivariate analysis showed the use of radiotherapy during conditioning and the presence of acute graft-versus-host disease (aGvHD) post-transplant to be protective factors against relapse and, in the case of aGvHD, also mortality. EFS at 3 years post-transplant was 36% in patients without aGvHD, 79% in those with grades I to II and 81% in those with III to IV. OS at 3 years post-transplant was 23% in patients without aGvHD, 56% in those with grades I to II and 57% in those with III to IV. When patients were stratified by MRD, aGvHD favored more those with positive MRD pre-transplant. Regarding post-transplant follow-up studies, patients with positive MRD measured by MFC relapsed less than those who remained negative: 87% versus 17. Conclusion: The presence of MRD pre-transplant measured by MFC identified a group of patients with a 5.5-fold greater risk of relapse and 3.4-fold of death, which confirmed the importance of its presence prior to transplant and the validity of the test for its identification. Introduction: Epstein-Barr virus (EBV) reactivation and/or its related disease is a serious clinical complication in patients who have undergone haploidentical hematopoietic stem cell transplantation (haploHSCT). T-cell depletion with antithymocyte globulin (ATG) is a common strategy for preventing GVHD in the allogeneic HSCT context. Previous studies reported that administration of ATG negatively affected T lymphocyte recovery and was a significant risk factor for EBV reactivation after haploHSCT. It remains unclear whether, in the setting of haploHSCT, different doses of ATG differentially affect the recoveries of T-cell subpopulations. Double negative, CD4 -CD8 -T cells are found to be involved in autoimmune syndromes, inflammation, and host defense. In the context of allogeneic HSCT, the characteristics and effects of double-negative T-cell recovery have not been documented. Material (or patients) and methods: 3 retrospective comparisons in lymphocytes recoveries and its correlation to EBV infection were performed in cohort patients who underwent allogeneic HSCT in our institute. Comparison 1: 29 patients received 6 mg/kg ATG versus 31 patients received 10 mg/kg ATG before haploHSCT from December 2010 to May 2012. These 60 Patients were diagnosed with standard-risk hematological malignancies. Comparison 2: 64 patients with EBV reactivation versus 192 patients without EBV reactivation after haploHSCT from January 2011 to December 2014. Comparison 3: 6 patients with EBV reactivation versus 18 patients without EBV reactivation after HSCT from HLA-matched unrelated donors since January 2011 to December 2014. Immunophenotypes of CD19 + B cells and CD3 + , CD4 + , CD8 + , CD4 + CD45RA + , CD4 + CD45RO + , CD4 + CD28 + , CD8 + CD28 + , and CD4 -CD8 -T cells were determined by flowcytometry. Results: We found that, compared to 6 mg/kg, 10 mg/kg ATG significantly hampered the recovery of CD4 -CD8 -T cells at 30, 60, 90, and 180 days after transplantation, respectively ( p40.05). Moreover, we showed that an increase in Epstein-Barr virus (EBV) infections, associated with the higher dose of ATG, was correlated with the delayed recovery of CD4 / CD8 double-negative T cells. In the context of allogeneic HSCT from either haploidentical or HLA-matched unrelated donors, impaired recovery of CD4 -CD8 -T cells was found in recipients with EBV reactivation at 30, 90, and 180 days after transplantation ( p40.05). By flowcytometry analysis of 3 representative haploHSCT recipients, we found 80% of recovered CD4 -CD8 -T cells were γδT cells. Conclusion: This study revealed a differential impact of different ATG conditioning doses on the recoveries of T cell subpopulations post-haploHSCT. Our finding was also the first to connect the recovery of CD4 -CD8 -T cells to the risk of EBV infection after allogeneic HSCT. Future prospective cohort studies should provide more unbiased evidences to clarify our findings and benefit the overall outcome in recipients of haploHSCT. Disclosure of Interest: None declared. effects by prolonging the half-life of IGF-1 and by modulating transport of IGF-1 to target tissues. Our aim was to investigate associations between the levels of plasma IGF-1 and IGFBP-3 and haematopoietic reconstitution in the early post-HCT phase. Material (or patients) and methods: 41 adults undergoing HCT at Rigshospitalet, Denmark from 2010-2013 were included. Diagnoses included malignant (n = 39) and benign (n = 2) diseases. Donors were MRD (n = 10), MUD (n = 27) or UCB (n = 4). Conditioning regimens were TBI-based (n = 34), busulphan plus cyclophosphamide (n = 6) or fludarabinebased (n = 1). Plasma levels of IGF-1 and IGFBP-3 were measured by chemiluminescence before transplantation and at day 0, +7 and +21 post-HCT, and were converted into age-standardised deviation scores (SDS). Haemoglobin, thrombocytes and leukocyte subsets were monitored daily during hospitalization, and T cells, B cells and NK cells were counted by flow cytometry at day +30, +60 and +90. T cell subsets included CD4+ T cells, CD8+ T cells, Th17 cells (CD4+CD161+CD196+), Tc17 cells (CD8+CD161 +CD196+) and Tregs (CD4+CD25 high CD127 low ). Results: IGF-1 rose from a mean value of -0.218 SDS (95%CI: -0.674;0.238) before conditioning to a peak of 0.671 SDS (0.219;1.123, P = 0.0002) at day 0 and then gradually declined. In contrast, IGFBP-3 decreased from 0.004 SDS (-0.455;0.463) before conditioning to a minimum of -1.097 SDS (-1.575;-0.618, P = 0.0001) at day +21. We investigated if IGF-1 and IGFBP-3 levels were associated with cell counts measured in parallel, and found that IGF-1 and IGFBP-3 levels at day +21 were positively correlated with the level of haemoglobin (r = 0.35, P = 0.040 and r = 0.40, P = 0.016, respectively) and thrombocyte counts (r = 0.50, P = 0.002 and r = 0.55, P40.001, respectively). No associations with neutrophil or absolute lymphocyte counts were found. Next we investigated if IGF-1 levels were predictive of lymphocyte subset reconstitution. IGF-1 and IGF-BP3 levels at day +21 correlated positively with NK cell counts at day +30 (r = 0.71, P40.0001 and r = 0.68, P40.0001, respectively). Furthermore, high IGF-1 levels at day -7, 0 and +7 correlated with the proportion of CD8+ T cells at day +90 post-HCT, and high IGFBP-3 levels at day +21 were associated with a high proportion of Th17 cells among the CD4+ T cells at day +90 (r = 0.43, P = 0.026). Nine patients (22.5%) developed aGVHD grade 3-4. High levels of IGF-1 on day +21 were associated with decreased risk of severe aGvHD (OR = 0.46 per 1 SDS increase in IGF-1, P = 0.046) in multivariate analysis. Conclusion: This study shows for the first time that high IGF-1 and IGFBP-3 levels in the early post-HCT phase predict a faster reconstitution of erythrocytes, thrombocytes and lymphocytes and a reduced risk of aGvHD. Assuming a causal relationship, IGF-1 and IGFBP-3 may increase haematopoiesis after allogeneic HCT by limiting epithelial damage and by direct effects on haematopoietic cells. Disclosure of Interest: None declared. Introduction: Complete hematopoietic chimerism is the aim of hematopoietic stem cell transplantation. In solid organ transplantation, micro and macrochimerism has been observed. Its impact on tolerance is still debated. Material (or patients) and methods: A multi-organ transplantation (liver, stomach, pancreas, spleen,smal intestineand colon) was performed for a 62 years-old woman. Donor was a 12 years old male. At 2 years post transplantation, a deep hemolytic anemia occurred with a positive direct antiglobulin test. Treatment based on corticotherapy and intravenous immune globulin failed; secondary use of Rituximab was efficient. Retrospectively, a blood group change from A Rh D+ C+ E-c+ e+ K-(recipient) to A Rh D+ C+ E+ c+ e+ K-(donor) was enhanced. This suggested an engraftment of the recipient marrow from hematopoietic stem cells coming from deceased donor. This hypothesis was confirmed by chimerism study on both blood and bone marrow. Results showed 0% of recipient's cells on blood at 2 years post transplantation and 0.85% of recipient's cells on CD34 cells and 0.2% of recipient's cells on CD3 cells in marrow. Results: During solid transplantation, donor leucocytes are cotransferred into recipients and can be detectable in recipient's blood in early post transplantation period [1] . Microchimerism is defined by the presence of o1% donor cells and its effect on tolerance induction remains unknown. Clinical trials studying simultaneous transplantation of donor hematopoietic stem cell and solid organ (kidney, liver) was performed to induce macrochimerism and thus, immune tolerance [2] [3] [4] . Mixed chimerism was induced in most patients and in a significant number of them, immunosuppression was successfully withdrawn. This case report is the second case published of total chimerism after solid organ transplantation without simultaneous hematopoietic stem cell transplantation [5] . Stem cells may have migrated from donor's liver and/or spleen. Engraftment was possibly facilitated by: (i) post transplantation immunosuppression. (ii) infection during the early post transplantation period, wich may have immune-modifying effect. (iii) Number of stem cells from liver and/or spleen related to donor's young age. Conclusion: Macrochimerism and, furthermore complete chimerism, remains a promising concept for tolerance induction in solid organ transplantation. Tolerance acquisition may result in an early imunosupressive therapy withdrawal. Thus, long lasting immunosuppressive complications (infections, secondary lymphoproliferative disorders, metabolic disorders) could be avoid without graft rejection. Introduction: HSCT is of benefit in pediatric patients with high-risk ALL in first or further remission. The prospective ALL SCT I-BFM Study, coordinated by Peters in Vienna, was initiated within the International BFM Study Group in order to assess whether the outcome of HSCT from a matched compatible donor (MD, 9 or 10 out of 10 HLA alleles) was inferior to the outcome of HSCT from a matched sibling donor (MSD) in children or young adults with ALL in CR carrying very high risk eligibility criteria for HSCT. Primary endpoint was EFS and secondary endpoints were NRM and incidence of aGVHD and cGVHD. Material (or patients) and methods: Between 2007 and 2013, in 10 countries (Czech Republic, Denmark, France, Israel, Italy, The Netherlands, Poland, Sweden, Slovakia, Turkey) 342 consecutive patients, o18 ys (68% M, median age 9 ys; 66% o12 ys) were enrolled in the core arm (MD vs MSD) of the study being transplanted from MSD (133; 48% in CR1, 48% in CR2, 4% 4CR2) and from a MD (209; 46% in CR1, 46% CR2, 8% 4CR2), either related (8%) or unrelated (92%), at a median of 189 and 197 days, respectively, after diagnosis/relapse. As per protocol, conditioning regimen consisted of TBI-VP, in patients 42 ys, or BU-CY-VP, in patients o2 ys and GVHD-prophylaxis consisted of CSA only for MSD and CSA/MTX/ATG for MD recipients. 24% and 33% CMV+ recipients have been grafted from CMV-MSD and MD, respectively. Median observation time was 4 years. Results: 3-yr EFS was 65% (SE 5) for MSD vs 57% (SE 4) for MD recipients (p 0.192), overall survival (OS) 74% (SE 5) vs 68% (SE 4) (p 0.121), cumulative incidence of relapse (CIR) 24% (SE 4) vs 26% (SE 4) (p 0.656) and NRM 10% (SE 3) vs 15% (SE 3) (p 0.184), respectively. No statistical significance was reached in the multivariate analyses for the 4 endpoints between MSD and MD. Being CMV-rec/+ don vs other combinations was associated with higher NRM (HR 2.95, CI 1.37-6.34, p 0.006) and lower OS (HR 1.84, p 0.018) . Adolescents had higher risk of NRM (HR 2.79, CI 1.29-6.06, p 0.010). Grade II-IV aGVHD occurred in 42% and 44% and grade III-IV in 19% and 17% of the MSD and MD recipients, respectively. 36% of the MSD and 25% of the MD recipients experienced chronic GVHD, which was severe in 25% and 14%, respectively. Cumulative incidence of extensive cGVHD was 22% (SE 4) for MSD and 12% (SE 2) for MD recipients (p 0.026). Being transplanted from a MD, compared with a MSD, was significantly associated with a reduced risk of extensive cGVHD (HR 0.41, p 0.003) , despite a similar risk of relapse. Grade III-IV aGVHD had a statistically significant impact on the risk of NRM (HR 7.29, Po0.0001) and EFS (HR 1.82, ; p 0.033), after adjusting for donor type and phase. Conclusion: Outcome of transplantation of MD pediatric recipients with VHR ALL in CR was not inferior to the outcome of MSD recipients in terms of EFS, OS, NRM and CIR, with probability of cGVHD being lower for MD versus MSD recipients and severe acute GVHD being associated with increased NRM and lower EFS but similar CIR. For nonmalignant diseases, there were 9 PID, 3 FA/SAA and 1 Thalassaemia major. The median age of patients was 7.5 years (range 3 months to 17 years). There were 23 males to 9 females. All had myeloablative conditioning except for the 5 cases of SCID. Results: Majority (75%, 24 cases) of cord units were from Singapore and Taiwan cord blood banks and 25% from other cord banks (6 from NMDP and 2 from Australian CBB). The racial distribution of recipients were 28 (87%) Chinese, 1 (3%) Malay and others 10% (1 Indonesian, 1 Australian, 1 Middle Eastern) . During this period we performed 8 MUD HSCT, in 6 Chinese and 2 Indian patients. Median neutrophil engraftment for UCBT was 26 days (14 to 86) and platelet engraftment 37 days (15 to 82). There were 4 graft failures (12.5%); 3 had nonmalignant conditions (1 with FA had successful 2 nd HSCT from haploidentical parent, 1 Thalassaemia major had autologous reconstitution, 1 with ALL had autologous reconstitution and is in ongoing DFS for 10 yrs and 1 chronic granulomatous disease refused 2 nd transplant and subsequently died 1.5 years later from pneumonitis. TRM 100 days was zero while 2/32 (6%) died from late TRM, 1 died of chronic GVHD of liver and gut, while another died from leukoencephalopathy. Of the 19 with malignant conditions; 5 had relapse of disease and died. The 5 year OS for malignant and nonmalignant diseases was 63.2% and 78.8% respectively. When compared with MSD HSCT for malignant diseases (5 year OS 58%) in our centre, the results are comparable. Conclusion: Unrelated cord blood as stem cell source contributed to 80% of our unrelated HSCT. The main source of UCB units was Singapore and Taiwan cord blood banks. Our 5 year OS for UCBT in malignant diseases was comparable with MSD HSCT. The nonmalignant conditions had excellent outcomes of 78.8%.With better supportive care and improved technology the haploidentical parent is an alternative source of stem cells for patients with no MSD or MUD donors. However unrelated cord blood remains a good option in our local population. Disclosure of Interest: None declared. Introduction: Mutations in IL7R cause an autosomal recessive form of severe combined deficiency (SCID) which is characterized by a T-B+ NK+ phenotype. Hematopoietic stem cell transplantation (HSCT) is the only curative option. Material (or patients) and methods: We retrospectively analysed data on clinical presentation and outcome of HSCT in 8 patients with IL7R-deficiency transplanted in our institution between 1996 and 2014. Results: Overall survival (OS) was 88% (7/8) with a median follow up of 15 years (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) . Donors were HLA-identical siblings (n = 2), a HLA-matched unrelated donor (n = 1), a HLAmismatched unrelated donor (n = 1) and HLA-mismatched haploidentical parents (n = 4). In vitro T-cell depletion was performed in all haploidentical grafts. Four patients received primary pretransplant conditioning, using busulfan in combination with either cyclophosphamide or fludarabine, and four patients were transplanted without conditioning, two from HLA-identical siblings and two from haploidentical donors. The latter 2 patients with haploidentical donors experienced primary graft failure. In both cases retransplants were performed successfully after conditioning (busulfan and cyclophosphamide). One patient died on d+51 after haploidentical transplantation due to generalized mycobacterial infection before T-cell reconstitution. Stable B-cell function without the need for Ig-substitution was noted in all 7 longterm survivors irrespective of B-cell engraftment. Conclusion: In patients with IL7R-deficiency conditioning was not needed in order to establish posttransplant B-cell function but was required to avoid graft failure after haploidentical HSCT with T-cell depleted grafts. The role of potentially alloreactive autologous NK cells in this setting needs to be further elucidated. Disclosure of Interest: None declared. Introduction: Chronic granulomatous disease (CGD) is a rare primary immunodeficiency characterized by recurrent lifethreatening infections and autoimmune sequelae. The most established curative option for CGD, currently, is allogeneic hematopoietic stem cell transplantation (HSCT). A myeloablative conditioning regimen may be associated with both transplant-related mortality (TRM) and relevant morbidity, especially in patients with high-risk disease, while a reduced intensity conditioning may result in an increased risk of graftfailure. Therefore, we explored a new preparative regimen, which enables appropriate myeloablation with reduced toxicity. Material (or patients) and methods: We report on 8 patients with CGD who underwent an allogeneic HSCT between February 2013 and September 2015 at the Bambino Gesù Children's Hospital (OPBG), Rome. The median age at S251 diagnosis was 3 years (range 0.1-9 years), while the median age at HSCT was 4 years (range 1-10 years). Pre-transplant medical history included lymphoadenitis, cutaneous abscesses, Nocardia-related pneumonia with lobectomy, Criptococcus-induced pneumonia, S. Aureus cellulitis and osetomyelitis, fungal multiple liver and CNS lesions. Five patients were transplanted from an HLA-identical sibling, whereas 3 received HSCT from a 10/10 allelic matched unrelated donor (MUD). All patients received bone marrow (BM) cells. The original conditioning regimen consisted of treosulfan 14 g/m 2 given once daily i.v. on days − 6 through − 4 (total dose, 42 g/m 2 ), fludarabine 40 mg/m 2 given once daily i.v. from day − 6 to day − 3 (total dose, 160 mg/m 2 ) and thiotepa 10 mg/kg i.v in two doses on days -8 and -7. As prophylaxis of GVHD, patients were given Cs-A and short-term methotrexate. All MUD recipients received anti-thymoglobulin (rabbit ATG from Neovis) at the dosage of 10 mg/Kg on days − 5 through − 3. Results: Two patients developed grade 2 mucositis and 2 other children experienced grade 2 skin rash not related to infection or GvHD. Only one patient developed transient elevation of liver enzymes. Remarkably, no life-threatening (grade 4) toxicities occurred. All patients had sustained primary engraftment of donor cells, the median time to reach neutrophil and platelet recovery being 16 days (range 12-23 days) and 19 days (range 12-26 days), respectively. No patient died due to transplant complications. Only 1 patient experienced grade II, skin-only acute GVHD, while no patient had chronic GVHD. None of the 8 children showed CMV or adenovirus reactivation, while 2 developed EBV reactivation which did not require any treatment because of low viral load. With a median follow-up of 13 months (range 3-34 months), all patients are alive and disease-free. Six out of the 8 patients had a full donor chimerism, while the two ramaining patients had 80% and 75% donor chimerism respectively. Conclusion: This study shows that treosulfan combined with fludarabine and thiotepa is an effective preparative regimen for children with CGD transplanted from either an HLAidentical sibling or a 10/10 MUD. The low incidence of treatment-related toxicity, the absence of transplant-related fatal episodes and the low incidence of acute GvHD in our cohort suggest that this approach is a suitable treatment option in particular in high-risk patients with CGD. Disclosure of Interest: None declared. Introduction: Blood stream infection (BSI) is one of the most serious complication of hematopoietic stem cell transplantation (HSCT). In 13-60% of HSCT recipients, BSI has been observed. Although HSCT related BSI usually occurs in the first 100 days, patients are not free of infections even after 100 days. Several risk factors have been described in different studies but these studies are not in agreement. Especially, published studies for pediatric patient population are still insufficent. Our study was performed to describe the occurrence of BSI in pediatric HSCT patients to define risk factors and assess the outcomes of BSI. Material (or patients) and methods: This study was a retrospective cohort analysis of 212 HSCT from January 2010 to December 2014 in Akdeniz University School of Medicine Pediatric HSCT Unit. Date of HSCT and date of BSI were recorded in order to calculate the time-to-infection in the period following HSCT. All patients received antimicrobial prophylaxis with a fluoroquinolone, fluconazole and acyclovir for 9 to 12 months after HSCT. After engraftment, patients received prophylaxis for Pneumocystis jirovecii pneumonitis with trimethoprim-sulfamethoxazole until 1 year after HSCT. Intravenous immunoglobulin was administered to patients according to immunglobulin G plasma levels. For the comparison between groups, the Pearson Chi-Square and T-test were used. overall survival between groups were assessed via Kaplan-Meier method and Cox regression. All analyses were performed using SPSS 20 and p-value of 0.05 was considered statistically significant. Results: In total 212 HSCT procedures, 31% (n = 66) were hematologic malignancies, 14% were solid tumors and other 55% were non-malign hematological diseases and immun deficiencies. Most of HSCT were performed from unrelated donors (%42). 108 (%51) has at least one BSI epizode. Most of them occured in the first 100 days (%68). Forty-nine percents of BSI had multiple episodes. Fifty-five percents of isolates were Gram positive cocci. According to statistical analysis, anti-thymocyte globuline (ATG) doses were higher in BSI group (P = 0,02). BSI in first 100 days were significantly lower in match sibling donor group (P = 0,03). In cord blood group, BSI in first 100 days were significantly higher (P = 0,01). In univariate analysis, full HLA matching was associated with reduced BSI risk in first 100 days and two or more HLA mismatching were associated with increased risk (P = 0,001). We have found that BSI is associated with increased mortality after HSCT (HR 1.76, 95% CI 1.07, 2.74, P = 0.02). Survival expectancy was better in no BSI group. Conclusion: There are few studies in the literature about BSI in pediatric HSCT. This study is also unique with the large number of patients and five year data. Our findings showed that BSI is associated with increased mortality after HSCT. Although the association between ATG and viral infections is well known, our findings also show the association of bacterial BSI with the use of ATG. It is interesting that BSI occurence is not associated with the use of ATG only but higher doses of ATG causes increased BSI rates. It is obvious that BSI is still a major complication of HSCT and further studies and prevention on reducing BSI will decrease mortality in HSCT. Disclosure of Interest: None declared. Material (or patients) and methods: We studied the transplantation outcomes of 15 pediatric patients (12boys, 3girls) including 12 DBA, 2 CAMT and one SCN patients whom were done HSCT between 2010 and 2015. The median age at the time of transplantation was 4 years (range, 7 months -15 years). All of the transplants were done using a non-TBI containing regimen including Busulfan and Cyclophophamide +/-ATG, just SCN received Fludarabine, ATG and Melphalan. Results: 9(60%) received their transplantation from an HLAmatched sibling donor, 3(20%) from HLA-matched otherrelated donor and 3(20%) from HLA-matched unrelated donor. Source of transplantation was in 9(60%) patients from peripheral blood, 4 from bone marrow and 2 from cord blood. The median time to neutrophil recovery was 12 days (range, 10-15 days) and to platelet recovery was 15 days (range, 8-23 days) . One of DBA patients did not achieve neutrophil engraftment. The 100-day mortality was 13%. Grade II to IV acute graft-versus-host disease occurred in 9 (60%) and chronic graft-versus-host disease in 2(13%).The 13 (87%) patients are still alive. Cause of death was GVHD grade IV (n = 2). Both of them were above 14 years old. Conclusion: To best of our knowledge it is the first time that transplantation results of single cytopenia patients are reported from Asia and Pacific region. Our data are compatible with issues which were published in other centers. Transplantation in the younger age before receiving blood products may improve survival. Disclosure of Interest: None declared. Although allogeneic HSCT was known as the effective therapy in these patients, the optimal myeloablative conditioning regimen is unknown. Even though myeloablative conditioning with total body irradiation (TBI) has been widely used for pediatric ALL, due to the high probability of late effects of TBI in children it is important to find a suitable myeloablative conditioning regimen without TBI. Material (or patients) and methods: During 1997-2014 a total of 46 pediatric patients with very high risk ALL who were in their first complete remission underwent non-TBI hematopoietic stem cell transplantation in our center. Median age was 12 years (range: o1-15); 32 patients (69.6%) were boys and half of donors were male; median age of donors was 19.5 years (range: 5-53); 38 patients (82.7%) had B-lineage leukemia; 42 patients (91.3%) received stem cells from their HLA-matched siblings and in 4 patients (8.7%) HLA-matched other related donors participated; in 44 patients (95.7%) peripheral blood was used as the source for stem cells and 2 patients (4.3%) received bone marrow stem cells. The conditioning regimen consisted of busulfan and cyclophosphamide. For the patients with other related donors, we used rabbit antithymocyte globulin as part of conditioning regimen. All patients received cyclosporine with methotrexate as GVHD prophylaxis. Results: Thirty-four patients (73.9%) developed acute GVHD, of which more than 90 percent recovered; 18 patients (39.1%) had chronic GVHD. Five-year overall survival, leukemia-free survival (LFS), relapse rate and transplantation-related mortality were 58.96% (95% confidence interval [CI]: 42.82-71.96%), 52.33% (95% CI: 36.52-65.94%), 38.98% (95% CI: 24.33-53.37%) and 8.70% (95% CI: 2.73-19.07%), respectively. Conclusion: TBI-free conditioning regimen consisting of busulfan/cyclophosphamide provides tolerable outcomes with encouraging leukemia-free survival rates, similar to what is reported in existing literature using TBI and it isn't associated with some of adverse effects of radiation, particularly in very high risk ALL in pediatric patients. Multi-centre studies are needed to better clarify the differences between radiationbased and TBI-free conditioning regimens in paediatric ALL HSCT. Disclosure of Interest: None declared. Introduction: Relapse is the major cause of treatment failure after allo-SCT in children with ALL. Treatment options are limited in this situation and a standardized approach to open a curative perspective is currently not available. The major challenge is to identify patients, who still have a chance to be cured in this difficult clinical situation. We hereby present an algorithm to meet this challenge. Material (or patients) and methods: 101 children with ALL received first allo-SCT in our institution between 2005 and 2014. Patients were grafted from MUD (n = 68), MSD (n = 21), MMFD (n = 11) and MMUD (n = 1) in CR 1 (n = 54), CR2 (n = 30), 4CR2 (n = 16) and NR (n = 1). Preparative regimens were myeloablative (n = 91) or RIC (n = 10) and based on TBI (n = 84) or chemotherapy (n = 17). After relapse, high-dose-chemotherapy (HDCHT) followed by second allo-SCT was offered if the patient was in good clinical condition. If second SCT did not seem to be feasible, a combination of low-dose-chemotherapy and donorlymphocyte-infusions (LDCHT+DLI) was offered. Results: For the whole cohort (n = 101), pEFS (4 years) was 66% (+/-5%, events 32), 4y-CIR was 25% (+/-4%, events = 23) and 4y-NRM was 9% (+/-3%, events 9). 23/101 patients relapsed after first SCT. 8/23 received HDCHT, followed by second SCT in 7/8 patients. 10/23 were treated with LDCHT+DLI, 5/23 received palliative care. 7/10 treated with LDCHT+DLI had suffered relapse before and 3/10 beyond day 365 post-SCT (d365). Amongst those, who relapsed before d365, 1/7 remained in CR, whereas those, who relapsed beyond d365, remained in CR in 3/3. Based on these findings, we developed an algorithm to guide the evaluation of treatment options in a relapse situation after allo-SCT. Further therapy can have (a) curative or (b) potentially curative intention: (a) If the patient is in good clinical condition, HDCHT followed by second allo-SCT could be offered. If second SCT seems not to be feasible, LDCHT+DLI can be done if relapse occurred beyond d365. (b) If second SCT seems not to be feasible and relapse occurred before d365, LDCHT+DLI can be offered. Application of this algorithm resulted in following therapies and outcomes: 11/23 patients were treated in (a) curative intention: 8/11 received HDCHT, consisting of IDA/DNX-FLAG in 3/11, blinatumomab in 3/11 and was clofarabine-based in 2/11. After induction of remission 7/11 children received second allo-SCT from haploidentical donor. 3/11 were treated with LDCHT+DLI. These curative approaches resulted in 3y-OS of 73% (+/-13%, events = 3). 7/23 patients treated in (b) potentially curative intention with LDCHT+DLI achieved 3y-OS of 14% (+/-13%, events = 6, P = 0.008). This algorithm to treat post-transplant relapse lead to 4y-OS of 76% (+/-4%, events = 23) for the whole cohort of 101 patients. Conclusion: In a relapse situation after first allo-SCT it is a major challenge to identify patients, who could benefit from further curative treatment. Prerequisites for therapeutic options is good clinical condition, an interval between SCT and relapse of at least Introduction: Fanconi anemia and Diamond-Blackfan anemia belong to the inherited bone marrow failure syndromes, a group of rare hematological disorders, with ineffective hematopoiesis and high risk of malignancy. Allogeneic hematopoietic stem cell transplantation is the only curative treatment for this patients but is related with high risk of rejection and transplant-related mortality. Material (or patients) and methods: From 1/1991 to 6/2015 in Polish pediatric transplant centers, 29 children with FA and 23 with DBA underwent allo-HSCT. The median age at transplant (Tx) was 10 years (range 2-18) and 3 years (range 1-14), respectively. In 29 patients with FA 34 allo-HSCT were performed. Three patients was transplanted twice, one thrice. Seven (21%) children were transplanted from MSD, 26 (76%) from MUD, and one (3%) from MMFD. HSC source were bone marrow (BM) (n = 19; 55%), peripheral blood (PB) (n = 12; 35%), cord blood (CB) (n = 2; 6%), and CB+BM (n = 1; 3%). Median CD34+ cells dose was 3.5 x 10 6 /kg of recipient b.w. (range 1. .6 x 10 6 /kg). Conditioning for first Tx was based on low dose Busulfan in 17 children (59%) and on Fludarabine with or without low dose Cyclophosphamide in remaining 12 (41%). All patients underwent in vivo T-cell depletion. For GvHD prophylaxis CsA was given, in addition 8 patients received MTX, 6 MP, and 4 MMF. In 23 children with DBA, 27 allo-HSCT were performed. One patient was transplanted twice and one three times. HSC were collected from MSD (n = 6; 22%), MUD (n = 15; 56%), and MMFD (n = 6; 22%). HSC derived from BM (n = 12; 44%), PB (n = 14; 52%), and BM+PB (n = 1; 4%). Median CD34+cell dose was 6.4 x 10 6 /kg of recipient b.w. (range 1.7-38.8 x 10 6 /kg). Preparative regimens prior first allo-HSCT was based on Busulfan in 14 (61%) patients, and in 9 (39%) on Treosulfan. All patients underwent in vivo T-cell depletion, and -except one -received CsA for GvHD prophylaxis along with MTX in 19 and MP in 3. Results: In FA group, one patient died before engraftment. All but two remaining patients engrafted (n = 26; 93%). Two patients developed secondary graft failure. Acute GvHD II-IV rate in alived, engrafted patients was 19% (n = 5/26), whilst cGvHD occurred in 21% (5/24) of them. Seven (24%) patients died (10, 21, 22, 57, 67, 110 and 855 days after allo-HSCT). The reasons of death was related to GvHD (n = 3), infections (n = 2), MOFS (n = 1) and accident (n = 1). Twenty two children are alive, with survival time ranging 12.5-182 months (median 42) after HSCT. For FA probability of overall survival at 10-years is 0.76. In DBA group all but two children engrafted 91% (n = 21). Three (13%) children died, two without engraftment (78 and 390 days after first allo-HSCT) and one from kidney toxicity and CMV reactivation 153 days post transplantation. Rate of aGvHD was 38% (n = 8/21) and 20% (n = 4/20) in case of cGvHD. Twenty one children are alive, with survival time ranging 5.5 -234.6 months (median 58 Introduction: Sickle cell anemia (SCA) remains associated with high risks of morbidity and early death. Children with SCA are at high risk for ischemic stroke and transient ischemic attacks, secondary to intracranial arteriopathy involving carotid and cerebral arteries. Children with Black African variant SCA are prone to invasive infections caused by S. pneumonia, H. influenzae and Plasmodium falciparum (in malarias areas). In Africa, malaria contributes substantially to the early mortality of patients with SCA. Allogeneic hematopoietic stem cell transplantation (HSCT) is the only curative treatment for SCA. We report our experience with transplantation in a group of Nigerian children affected by SCA. Material (or patients) and methods: This study included 37 consecutive SCA patients who underwent bone marrow transplantation from human leukocyte antigen (HLA)-identical sibling donors between 2010 and 2015 following a myeloablative-conditioning regimen. Patients received fludarabine (30 mg/m 2 /day) for 5 days and a conditioning regimen including targeted intravenous busulfan (14 mg/kg total dose) and cyclophosphamide (200 mg/kg total dose). Blood samples were collected in different Nigerian Hospitals and shipped to the Laboratory of the IME Foundation in Italy where were processed for DNA preparation on a fully automated system (Maxwell, Promega, Madison, WI). Low resolution HLA-A, -B, -C, -DRB1 and -DQB1 typing was performed using the polymerase chain reaction-sequence specific oligonucleotide (PCR-SSO) technique (LABType -One Lambda, Canoga Park, CA). Results: The median patient age was 10 years (range 2-17 years). Before transplantation, seventeen patients had recurrent, painful, vaso-occlusive crisis; twelve patients had recurrent painful crisis in association with acute chest syndrome; three patients experienced ischemic stroke and recurrent vaso-occlusive crisis; two patients experienced ischemic stroke; one patient exhibited leucocytosis; and one patient exhibited priapism. Of the 37 patients, 32 survived without sickle cell disease, with Lansky/Karnofsky scores of 100, following a myeloablative-conditioning regimen. The probabilities of survival, SCA-free survival, and transplantrelated mortality after transplant were 87%, 87%, and 13%, respectively. All surviving patients remained free of any SCArelated events after transplantation. Within the frame of the HSCT program for the treatment of SCA, a total of 124 Nigerian families with 2 to 11 children (average 2.5) were typed for five HLA loci (A, B, C, DRB1 and DQB1) and the phased genotypes were unambiguously determined. Thirty-six percent of the patients had at his disposal within the family a HLA compatible sibling. Conclusion: The protocols used for the preparation to the transplant in thalassemia are very effective also in the other severe hemoglobinopathy as in the sickle cell anemia. If a SCA patient has a HLA identical family member, the allogeneic cellular gene therapy through the transplantation of the hematopoietic stem cells should be performed as soon as possible, before the disease progresses through a treatmentrelated irreversible organ damage. Disclosure of Interest: None declared. IL-15/4-1BBL activated and expanded Natural Killer cells in paediatric patients with refractory acute leukaemia D. Corral 1 , J. Valentín 2 , D. Bueno 3 , P. Carrasco 4 , R. de Paz 5 , R. Rodriguez 6 , L. Fernandez 7 , A. Balas 8 , J. L. Vicario 9 , D. Campana 10 , J. Martinez 11 , A. Pérez Martínez 12,* 1 Pediatria, Universidad Autónoma de Madrid, 2 Inmunidad Innata, Instituto de Investigación La Paz, 3 Unidad de Hemato-Oncología Infantil. Hospital La Paz., Hospital Universitario La Paz, 4 Hemato-Oncología Pediátrica Molecular, Instituto de Medicina Molecular y Genética (IMMGEN), 5 Hematologia, 6 Unidad de Inmunologia, Hospital Universitario La Paz, 7 Investigación clínica, Centro Nacional de Investigaciones Oncológicas (CNIO), 8 Histocompatibilidad, 9 Unidad de Tipaje HLA, Centro de Transfusiones de la Comunidad de Madrid, Madrid, Spain, 10 Department of Paediatrics, National University of Singapore (NUS), University of Singapore (NUS), Singapore, Singapore, 11 Servicio de Hematologia, Hospital Universitario 12 de Octubre, Hospital Universitario La Paz, Madrid, Spain Introduction: Current therapies fail in most children with refractory or relapse leukaemia. Novel therapies are needed. Leukaemia cells are susceptible to be killed by activated Natural Killer (NK) cells. Highly activated/expanded NK cells can be generated ex-vivo by stimulation with the human leukocyte antigen-deficient cell line K562, genetically modified to express 41BB-ligand and membrane-bound interleukin-15, developed at SJCRH (Memphis, TN). We tested the safety and feasibility of haploidentical activated and expanded NK cell therapy in this heavily pre-treated paediatric population in two phase I/II trials (EudraCT: 2012-005146-38 and EudraCT: 2012-000054-63) . Material (or patients) and methods: Nineteen children with a median age of 12 years who had refractory (10 of them after allogenic stem cell transplantation) acute leukaemia (6 B cell lymphoblastic leukaemia, 7 T cell lymphoblastic leukaemia and 6 acute myeloid leukaemia) were treated with rescue chemotherapy followed by the infusion of fresh activated and expanded NK cells obtained from haploidentical donor peripheral blood followed by three times a week subcutaneous administration of low dose of IL-2. Rescue chemotherapy consisted in Cy-Flu (4 patients), clofarabine-based (6 patients), FLAG-IDA (3 patients) or nelarabine-based (6 patients). Results: To date, we have infused 48 NK cell products containing a median of 11.74x10 6 /kg. All infusions were well tolerated. Thirteen patients have completed treatment and this is ongoing in 1; in 5 additional patients treatment could not be completed because of leukaemia progression (n = 2) or chemotherapy-related toxicities (n = 3). Among the 13 patients who completed the treatment, 6 achieved negative minimal residual disease, 5 had cytological remission and 2 had disease progression. All patients that achieved negative minimal residual disease received an allogenic stem cell transplantation. Four of them are and leukemia-free with a median of 200 days post-trasplant. Conclusion: Infusion of fresh activated and expanded NK cell therapy was feasible and safe in a heavily pre-treated paediatric population, and should be further investigated in patients with high-risk leukaemia. References: 1. Rubnitz JE et al. Pediatr Blood Cancer. 2015; 62:1468 -72. 2. Rubnitz JE et al. J Clin Oncol. 2010 20; 28:955-9. 3. Curti A et al. Blood. 2011; 118:3273-9 . Disclosure of Interest: None declared. Introduction: Inmune anaemia (IA) is a well-known, although infrequent, haematological complication after allogeneic hematopoietic stem cell transplantation (HSCT). Paediatric information regarding this topic is scanty. Aim: To study incidence, risk factors, treatment and clinical outcome of IA in our series. Material (or patients) and methods: Descriptive and retrospective study. We have included every consecutive patient undergoing allo HSCT in our Unit. The aim of the study was twofold; to explore children's healthcare experiences of HSCT in order to guide improvements to local inpatient services, and to inform plans for the development of PROMs for paediatric HSCT services. Material (or patients) and methods: A qualitative pilot study was undertaken adopting a range of child-centred, participative methods for data collection with children. Semi-structured interviews were conducted using play, photo elicitation and drawing. The topic guide explored positive and negative aspects of hospitalisation and the HSCT, and areas for improvement. 5 children and 5 parents took part. Interviews took place between day -1 to +33 from transplant, depending on the wellbeing of the child. Engagements took place over one or two occasions and lasted on average 70 min (range 22-188) for children, and 49 min (range 26-65) for parents. All children taking part were boys and ages ranged from 5-12 years, the parents comprised 3 fathers and 2 mothers. Data was analysed using a constant comparison approach. Results: Both children and parents demonstrated an understanding and acceptance of the intense HSCT regimen and shared a hope that this would help them (their child). They also reported frightening and difficult experiences. Children described understanding that HSCT will make them ill before it makes them better: having to go into hospital when they are not "feeling ill" means HSCT is particularly difficult, compared with previous hospitalisations. Children perceived the hospital as 'a world of illness', a painful disruption to normal life. Staying in touch with day-to-day ordinary life was a priority for children, and play provision a particularly valued element of this. Conclusion: The findings suggest areas for local service improvements in relation to play provision and support for children to maintain as much of their normal daily life as possible during their HSCT hospitalisation. The study demonstrates the value of exploring children's views of the services they receive and that further engagement with children and families would be feasible as part of developing a PROMs for paediatric HSCT services. The use of qualitative, participative methods, which allows considerable freedom of expression proved particularly valuable in engaging with the children. A challenge for the next stage of this work will be to develop a PROM which engages children in the same way. References: Arbuckle, R. and Abetz-Webb, L., (2013 5,5 years (0,7-15,7) . The most prevalent underlying diagnosis was ALL (13), 3 in CR1, 7 in CR2 or more and 4 with active disease. Haplo SCT was indicated in 8 patients for advanced disease because of the potential inmunological benefit and in 6 as the only option due to the absence of a matched donor. For 6 patients it was their second transplant. Regarding the donor, mother was chosen in 11 cases and father in 4. KIR allorreactivity was detected in 10 cases (all of them in GvH direction). In all cases, unmalipulated peripheral blood was used as stem cell source. All patients were conditioned with mieloablative regimes. GVHD prophylaxis consisted of post infusión ciclophosphamide on days +3 and +4 plus tacrolimus and MMF from day +5. Median lapse of time for neutrophil engraftment was 14.3 days (SD+/-1.9). We did not have any engraftment failure. Complications: 12 patients presented acute GVHD (2 grade I, 5 grade II and 5 grade III) and 6 patients presented chronic GVHD (3 mild, 2 moderate and 1 severe). Three patients presented autoinmune haemolytic anaemia, one of them associated with thrombocytopenia (Evans syndrome). Thrombotic microangiopathy was diagnosed in three patients and sinusoidal obstructive sindrome in one. Thirteen patients suffered infections (9 viral, 4 bacterial, 1 fungal). Current status: 7 patients are alive and disease-free. 7 patients died, 5 for infectious complications before day+100, 1 for infection after day +100 associated with chronic GVHD and 1 died for relapse and desease progression. Conclusion: Even though our study is conditioned by the small number of patients: S258 after MFD-BM 1 ± 2%, and after uCB 17 ± 4% (P = 0.001). In multivariate analysis only uCB as the donor source (P = 0.048) and no previous chemotherapeutic treatment (P = 0.036) remained significant predictors ( Figure 1 ). Ongoing (and symptomatic) low blood cell counts required extensive immunosuppressive treatment for many months (24 patients received IVIG+prednison, 17/24 patients also received Ritux-imab+ongoing IVIG substitution, and 2 of these even required additional treatment with Fludarabine). With this management, the overall survival was excellent, 85% ± 8% compared to 67 ± 3% in patients without AI cytopenia (NS, P = 0.07). TRM was also unaffected, 10% ± 7% in patients with AI cytopenia compared to 22 ± 3% in patients without (NS, P = 0.16). Conclusion: uCB and "no chemotherapeutic pretreatment" were significant predictors for development of AI-cytopenia in pediatric HCT. With aggressive standardized immunosuppressive management, TRM is comparable to patients without AIcytopenia. Developing strategies to prevent AI-cytopenia, should primarily focus on these at-risk groups. Disclosure of Interest: None declared. Introduction: Patients with an hematological disease often receive multiple transfusions of 'packed red cells' (PRC) which can lead to iron overload defined as serum ferritin (SF) 41000 μg/L. Recent literature concludes that this has an effect on morbidity and mortality in adult SCT patients and iron chelation might be a therapeutic option. In pediatric patients data and information on HSCT and iron overload is limited. Material (or patients) and methods: We conducted a retrospective study of 58 pediatric patients of the pediatric hemato-oncology ward at the University Hospital of Ghent who underwent allogeneic HSCT between 2007 and 2012. Pts with hemoglobinopathy and SCID pts were excluded. Pts characteristics and relationship of iron overload with late organ damage (mostly to the liver) and survival were studied. We also looked at the spontaneous evolution of SF-levels posttransplant. Results: 8 out of 58 patients (13,8%) suffered TRM, TRM was responsible for 8 out of 22 (36,4%) deaths. All other deaths were due to relapse of malignant disease. We found no correlation between TRM and pretransplant SF (P = 0,67), even when corrected for pretransplant CRP (P = 0,39).Seven pts developed VOD which was reversible with supportive therapy. We did not find a correlation between SF4 1000 μg/L and VOD (P = 0,17) nor on GvHD or relapse.SF pretransplant was highly variable with a mean of 932 μg/L (15-11695) and significantly related to the number of PRC transfusions ( p40,0001). In 18 pts, pretransplant SF was more than 1000 μg/L. None of the pts was treated with chelation therapy pre SCT. There was no correlation between the number of transfusions post-transplant and an elevated SF in the immediate period post-transplant and at 6 months posttransplant (P = 0,14). At 12 months post-transplant we did find a significant correlation between these variables (P = 0,01). At 24 months the correlation was borderline significant (P = 0,0675).There was no significant association between elevated AST/ALT and SF 41000 μg/L in the first year post SCT. However a significant association between elevated AST/ ALT and SF at 12 (P = 0,03), 18 (P = 0,05) and 24 months (P = 0,005) was noted.Our study of the natural evolution of ferritin post-transplant shows that there is a spontaneous lowering of ferritin in the first two years after HSCT. Pretransplant, the average SF was 923 μg/L with a peak at 3 months post SCT and a spontaneous decline to an average SF of 1003 μg/L at 12 months and 445 μg/L at 24 months post-SCT. Conclusion: We conclude that high serum ferritin (41000 μg/L) has no influence on TRM in children. We did however mark a correlation between high transaminases and SF in the later period post-transplant. Therefore we strive for more follow-up concerning possible liver damage in patients who show a persistently high SF. This should be done using T2*MRI and/or liver biopsy. Future research should also include whether chelation could be adventitious to certain patients. Although chelation is already used in pediatric HSCT, there exists insufficient current evidence to do so. Clear guidelines for detection and possible treatment of iron overload in children undergoing allogeneic HSCT are needed. Disclosure of Interest: None declared. Subsequently, an unselected aliquot of 10x10 6 /kg T cells was added to the purified stem cells resulting in a median T cell depletion of 2 log. The diagnoses were: acute lymphatic leukemia (n = 9; CR1 = 4, CR2 = 5), NHL (n = 1) acute (n = 5; CR1 = 1, CR2 = 4), chronic myeloic leukemia (n = 3), MDS (n = 3) and Wiskott-Aldrich (n = 1). Median age was 10 years (0.5-18 Chronic GvHD occurred in only one patients (4,5%). 14/22 patients are alive with a median follow up of 9,8 years (range 5 to 11.5 years). Five year EFS was 64% (all patients), 50% (ALL/ NHL) and 73% (myeloic leukemias /MDS). Relapse probability at five years was 23%, probability of TRM at five years was 14%. Causes of death were relapse (n = 5) and infections (n = 3). Thus, stable and favorable survival rates with a low incidence of GvHD were achieved. Conclusion: The method allows the administration of clearly defined T cell numbers independent from the volume of the apheresis product. This may be of advantage in particular in small children and if the donor does not accept a bone marrow harvest. All available stem cells can be infused without the limitation of intolerable high T cell numbers. Disclosure of Interest: None declared. respectively. An increased bilirubin was recorded in 13/24 patients. Two intra-alveolar hemorrhages (HIA) and 4 delayed onsets of pulmonary arterial hypertension (PAH) were diagnosed. One patient died of VOD-related toxicity. The hospitalization duration was 32 (22-108) and 26 (16-46) days in patients with or without VOD, respectively (P = 0.02). In a univariate analysis, gender (P = 0.5), age at diagnosis (P = 0.7), burden of the disease (stage 4, P = 0.6; bone metastasis, P = 0.2; bone marrow involvement, P = 0.4; NMYC amplification, P = 0.9), and previous treatments including previous HD-Thiotepa (P = 0.3) had no significant impact on incidence and severity of VOD. However, even if not significant, the administration order of the Bu-Mel administration and the alkylating agent-dose intensity may have impacted the incidence of VOD. Conclusion: In our study, we reported a high incidence of VOD in children with neuroblastoma in spite of defibrotide prophylaxis after Bu-Mel with ASCT. The accurate assessment of this complication in our department may explain the persistent high incidence of VOD. The toxicity-related death rate was comparable to that reported in the SIOPEN/HR-NBL1 population (1.7% vs 3%, respectively). In this specific population, already identified as at high-risk for VOD development, no additional risk factor was displayed. A prospective randomized study is required to evaluate the potential benefits of the defibrotide prophylaxis in this specific population. Disclosure of Interest: None declared. Introduction: Allogeneic hematopoietic cell transplantation (HCT) is the only potential curative option for some hematologic chilhood cancers, but many children do not have a HLA-matched donor. Haploidentical stem cell transplantation (haplo-HCT) has become an interesting strategy to avoid this problem, but the required T-cell depletion produce a profund and long-lasting inmunosupression, as well as graft failure and rejection. As naïve T-cells (identified by CD45RA expression) are believed to cause graft versus host disease (GvHD), while memory T-cell (CD45RA-) provide inmediate anti-infection, anti-leukemia and anti-rejection effects, allogeneic HCT using CD45RA depletion arises as a novel approach to haplo-HCT. Material (or patients) and methods: Six children, 5 of them with high-risk malignancies and 1 with a immunodeficiency were transplanted following nonmyeloablative conditioning regimen. Each patient received two cell products, one created by CD34 selection and the other through CD45RA depletion from the CD34-fraction by CliniMACS device. Results: Graft consisted in a mean of 7x10 6 /kg CD34+ cells/kg (4,5-9,89x10 6 /kg), 3x10 5 /kg CD3+ cells/kg (6,9x10 3 -1x10 6 /kg), 1x10 6 /kg CD45RA+ cells (1x10 4 -4,98x10 6 /kg) and 7x10 8 /kg CD45RO+ cells (3,8x10 8 -1,9x10 9 /kg). CD45RA depletion resulted in a mean of 3 log (2,2-4,29 log). All six patients engrafted with a mean of 11 days (10) (11) (12) and rapidly achieved 100% donor chimerism. No graft failure was observed. Only one patient developed GVHD4II that was sensible to steroids. T-cells led immune recovery. They achieved values higher than 700 cells/mcL and 1000 cells/mcL at day 30 and 60 respectivelly. Most of T cells were both CD8+CD45RA-and CD4+CD45RA-T cells (median of 395x10 6 /mm 3 and 243x10 6 / mm 3 , respectively), while the number of CD8+45RA+ and CD4 +45RA+ cells remained low (median of 0.7x10 6 and 0.07x10 6 respectively), recapitulating the CD45RA depleted graft composition. A total of 4 patients presented CMV reactivation, but none progressed to disease. With a mean of 100 days of follow up two patients relapsed, both had minimal residual disease positive at haplo-TPH and two patient died, one debt to leukemia progression and one because cardiogenic shock. Conclusion: CD45RA depleted haplo-HCT is well tolerated with a rapid engraftment, minimal risk of GvHD and an accelerated inmunologic reconstitution However, in our personal experience MIOP patients older than 10 months are at high risk for non-engraftment and rejection. In order to favour engraftment in these particular setting, we chose an upfront serotherapy-based conditioning regimen prior to T-cell replete haploidentical bone marrow grafts and post-transplant cyclophosphamide. Material (or patients) and methods: Four consecutive patients with MIOP received T-cell replete haploidentical parental bone marrow grafts at a median age of 1.4 years (0.8 to 8.6 years). The first patient failed engraftment after two previous CD34+ selected T cell depleted haploidentical HSCT from his father and his mother (98 and 48 days prior to the third T replete HSCT with again his father as a donor). Splenectomy was performed for one patient prior to HSCT. The grafts contained a median of 6.45x10 8 total nucleated (4.5-8.5) , 14.8x10 6 CD34 + (14.7-23.7) and 8.7x10 7 CD3 + (7.2-19.5 ) cells per kg. Conditioning contained a serotherapy backbone of rituximab 375 mg/m 2 at day -12, alemtuzumab 0.4-0.5 mg/kg over two days (day-11/-10). All patients received fludarabine 150-160 mg/m 2 over 5 or 4 days respectively, cyclophosphamide 29 mg/kg, and busulfan with myeloablative AUC according to the EBMT/IEWP ESID guidelines. GVHD prophylaxis consisted of cyclophosphamide 2x50mg/kg on days +3 and +4, followed by CSA associated with MMF (n = 3) or steroids (n = 1). All patients received ursolvan and defibrotide (until day +21) as prevention for veino-occlusive disease. Results: After a median follow-up of 6.6 months (2.7-7.8) Introduction: Long-term complications related to bone metabolism has been increased after hematopoietic stem cell transplantation (HSCT). Children who do not reach optimal bone mineral density (BMD) are more likely to be at risk for osteoporotic fractures. Factors associated with bone loss after HSCT reported by recent studies are the type of conditioning regimen and immunosuppressive treatment, endocrine and metabolic complications, vitamin D (25-OHD) deficiency, low physical activity, and sun exposure. We aimed in this study to evaluate effect(s) of HSCT on BMD and 25 OHD status in children. Material (or patients) and methods: We retrospectively analysed children who underwent HSCT in our center from April 2010 to July 2013 with at least two years follow-up without relapse or progression of primary disease. Data were collected from medical records of the patients. We compared our findings in thalassemic patients with those in patients with transfusion dependent (TD) thalassemia major. (Table 1 ). At 6 th month of HSCT, there was a trend towards increasing in BMD. Further, BMD was found to be significantly increased at the 1 st year and 2 nd after HSCT (P = 0.02, and P40.001). BMD standard deviation score (SDS), which was calculated from the mean for age-and sexmatched controls, was significantly increased at two years after HSCT compared to pre-HSCT values (-0.59 ± 1.35 vs. -1.48 ± 1.28,P = 0.008). 25-OHD levels did not differ before and after HSCT. Sex, type of conditioning regimen, presence of chronic GvHD, steroid use, and level of 25-OHD were not associated with osteoporosis. Patients who aged 410 years had lower BMD SDS at the 1 st year after HSCT(-1.84 ± 1.55 vs. -0.36 ± 1.29, P = 0.001). In the initial assessment comparison of thalassemia patients' BMD and BMD SDS were not different between patients who had HSCT (20 patients) and TD (20 patients). However, 25-OHD levels were significantly lower in TD patients (P = 0.039). Mean total BMD was significantly increased in two years both in patients who had HSCT and TD (P = 0.027,and P = 0.005, respectively). There was no difference in serum 25-OHD levels and BMD changes between two thalassemia groups (Table 2) . Allogenic HSCT received full intensity Treo-based conditioning regimen. Diagnosis: AML -30% (n = 6), Hurler syndrome -20% (n = 4), SCID -15% (n = 3), CGD -5% (n = 1). Breastfeeding was not terminated during the conditioning period and within the whole early post-transplant period with complications. [P283] Results: The incidence of severe oropharyngeal mucositis and neutropenic enterocolitis (higher than grade 2) in all the survived patients (n = 20) was 15%. There were no severe infectious complications and inoculation of de novo pathological microflora that serve as indicators of breastfeeding complications observed in the analysed group. The incidence of intestinal form of graft versus host disease (GvHD) with allogenic HSCT (more than II gr.) was 14.3%. The performed analysis demonstrated that breastfeeding may be recommended for children at the BMT unit as it doesn't affect the process of early post-transplant period and doesn't increase the incidence of toxic and immune complications, protected gastrointestinal infectious complications and is the important element of psychological support in the "mother-child" system in a clinical setting. Disclosure of Interest: None declared. Low incidence of clinically significant hemorrhagic cystitis in patients after ex-vivo CD3+ or αβ+ T-cell depleted haploidentical stem cell transplantation in children Two patients after MUD had histologically confirmed acute GVHD grade 2 (skin stage 1-2 and gut stage 1) and needed topical steroids and prolonged tacrolimus treatment. One of these patients had histologically confirmed chronic gut GVHD and received budesonide. Acute or chronic GVHD more than grade 2 was not registred. Out of 7 patients, 2 developed CMV viremia post-transplant and received preemptive gancyclovir therapy. One patient after MUD had CMVchorioretinitis. Three patients after MUD had prolonged mixed chimerism without disease manifestation (maximum 98% patient-derived СD3+). After termination of immunosuppression two patients showed a trend to increase donor CD3+ chimerism from 2% to 60-80%. In two patients chimerism analysis in selected native T-cell subpopulation showed predominant donor chimerism (78 and 97%). At the moment of this report one patient has stable high-level mixed chimerism (90% autologous CD3+) without disease manifestation. Overall and disease-free survivals were 100% (follow up 4-33 months). The treosulfan-based regimen before TCR åß/ CD19 -depletion transplantation in inherited hemophagocytic lymphohistiocytosis is effective with low toxicity, abscence of severe GVHD and provides promising overall and disease-free survival rates. Disclosure of Interest: None declared. Biomarkers-based early prediction of response to steroids in pediatric aGvHD patients E. Jol-van der Zijde 1,* , A. Introduction: Severe acute graft versus host disease (aGvHD) is a major complication after allogeneic HSCT, and high dose steroids is the gold standard as first line treatment for aGvHD. However, about 50% of the patients progress to steroidrefractory aGvHD which is associated with a high mortality rate. In the last years the usefulness of several biomarkers have been reported to support diagnosis, to document response to therapy and to predict morbidity and mortality. In a previous paper from our group biomarkers were used to monitor the response to mesenchymal stromal cells (MSC) treatment. In the present study we focused on early prediction of steroidrefractory patients in an attempt to improve our risk assessment. Material (or patients) and methods: In this retrospective study we quantified 10 biomarkers sera from 34 pediatric patients with severe aGvHD grade II-IV; this cohort consist of 16 steroid-refractory patients and 18 steroid responders. From these patients, sera were selected pre-conditioning, at w+1, 2, and 3 post HSCT, at one week before start steroids (2 mg/kg), at day of start steroids, and 1, 2 and 4 weeks thereafter. Sera from non-GvHD HSCT patients (n = 13) were used as controls. IL-2Ra, TNFR1, HGF, IL-8, ST-2. VEGF and ANG-2 were measured by Luminex assay (R&D), Elafin, REG3a and CK18 by ELISA. For statistical analysis log transformed data were used. Results: Prior to start of the conditioning regimen, patients who developed steroid-refractory aGVHD were characterized by significantly higher serum levels of IL-2RA ( p40.01), HGF ( p40.01) and REG3a (P = 0.05). IL-8 was significantly higher in refractory patients from start steroids onwards ( p40.01). Levels of TNFR1, Elafin, CK18 and VEGF didn't differ at any time point between steroid responding and refractory patients. In the post-transplant phase, preceding the onset of aGvHD symptoms, levels of ST-2 (left panel Figure) and ANG-2 were higher in steroid refractory patients at any time point. At 1 week post HSCT ANG-2 was 2.3 times higher (7300 vs. 3200 pg/mL; P40.01), and ST-2 was 2.5 times higher (49.000 vs 19000 pg/mL; P = 0.02). ST-2 levels (from 1 week pre-start steroids (P-1 W) to 4 weeks after start) and the ANG-2/VEGF ratios (from 2 to 4 weeks after start steroids) were significantly higher in steroid-refractory patients (P = 0.01). A cut-off level of 35000 pg/mL at P-1 W is predictive for steroid refractory with a sensitivity of 92% (95% CI: 64-99%) and a specificity of 80% (95% CI: 52-96%) (right panel Figure) . [P287] (GvHD) prophylaxis was carried out with Mycofenolate until day 60, if residual T-cells in the graft exceeded 25000/kg BW. 4/21 patients rejected the haplo graft and needed reconditioning with Fludarabine (3x40mg/m 2 ), Thiotepa (1x5mg/kg), ATG and/or OKT3, 7 Gy TLI and a second stem cell donation from a different donor. Results: Final engraftment was achieved in 21/21 patients. Median time to reach 500/μl neutrophiles was 10 days (8) (9) (10) (11) (12) (13) (14) (15) . Independence from platelet substitution was reached after 9 days (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) . 17/21 patients (81%) had no signs of GvHD or GvHD grade I, 1 patient (5%) had GvHD grade II, 3 patients developed grade III (15%). TRM at day+100 and after 3 years was 0% and 15%, respectively. Event free survival (EFS) of the total cohort after 3 years was 18% (4 out of 21). EFS of RMS was 29% after 3 years, whereas patients with Ewings sarcoma showed a poorer outcome with only 1 out of 11 patient surviving at 2 years, which was presumably due to the higher portion of Ewing patients being transplanted in non remission (NR) (9 out of 11). According to remission status at SCT none of the patients with progressive disease or NR survived, whereas patients with complete remission (CR) had a more favorable outcome of 47% (4 out Material (or patients) and methods: The study included clinical observation of the post-transplant course in ten patients with acute myeloid leukemia (AML) and 7 healthy donors. Bone marrow nucleated cells were selectively harvested two weeks prior to BMT, followed by monolayer culture in alpha-MEM culture medium with 20% fetal bovine serum. Upon growth of fibroblast-like cell colonies (CFU-F), their hematopoiesis-supporting activity was determined in the agar-drop/liquid culture system, as well as their differentiating ability along adipogenic and osteogenic pathways. We have also analyzed the relative expression of selectin and CXCR4 genes in these cells. Results: When comparing functional characteristics of BMSC from healthy donors and AML patients, an increased hemostimulatory activity of the latter was noted, as reflected by an increase in large and small CFU-GM numbers (Po0.02). In addition, an increase in differentiation along adipogenic and osteogenic pathways was observed in AML patients (P = 0.03). Moreover, the number of CFU-Fs, capable for adipogenic differentiation was inversely correlated with platelet recovery time (P = 0.05). In contrast, higher numbers of osteogenic colonies in culture were associated with an increased time to leukocyte lineage recovery (P = 0.05). Furthermore, a direct correlation was observed between CFU-F capable of osteogenic differentiation, and the number of blast cells at the time of transplantation (P = 0.05). When analyzing gene expression in BMSC population, a decreased expression of the CXCR4 gene responsible for the homing effect, with age of the patient's (P = 0.05) was noted. CMV induces in patients the expansion of NK cells, which expresses activating and inhibitory killer-immunoglobulin-like receptors, NKp46, NKG2C and NKG2D receptors as well as the proliferation of γ/δ T cells and CD8+ T cells co-expressing also PD-1, CD56, CD57 and NKG2C. These CMV specific NK-and T cells persist over a long time-period and counteract probably AML blasts by cross reactivity and intensifying graft-versusleukemia effect independent of GVHD. Conclusion: CMV reactivation reduces the relapse risk of patients with AML. Patients with advanced stages of AML may benefit especially from the induction of a tolerated CMV reactivation. This could be associated with an increased NRM rate. The results of studies evaluating pretransplant CMV serostatus of patient or donor are controversial in regard of relapse rates, NRM or OS rates.thor. Introduction: The use of tyrosine kinase inhibitors in patients with Philadelphia chromosome-positive (Ph+) acute lymphoblastic leukemia (ALL) should be considered after allogeneic haematopoietic stem cell transplantation (alloHCT), as either primary prophylaxis or preemptive therapy after detection of BCR-ABL transcript. The data on the use of dasatinib (DAS) in posttransplant maintenance therapy in Ph+ ALL are limited to several single case and case series reports. Material (or patients) and methods: We retrospectively analyzed the efficacy, safety and tolerability of DAS given after alloHCT as either primary prophylaxis or preemptive treatment following BCR-ABL transcript detection in patients with Ph+ ALL who were transplanted in the TKI era in the nine Polish centers. Results: The study group consisted of 19 patients with median age of 31 years (range, 21-60 years) in first complete remission (CR) (n = 14) or with advanced Ph+ ALL, beyond first CR (n = 5) who underwent first (n = 15) or subsequent alloHCT (n = 4) with HLA-matched sibling (n = 7) or unrelated donors (n = 12) and received DAS in the posttransplant maintenance therapy between July 2007 and July 2015. Minimal residual disease (MRD) was detectable at the molecular level in 12 patients (63%) before alloHCT. Fifteen patients were treated with DAS preemptively following MRD detection after alloHCT with a median interval between transplant and initiation of DAS treatment of 7 months (range, 1-22 months). Four patients with undetectable molecular MRD after alloHCT received DAS as prophylactic therapy which was started after a median time of 5 months (range, 2-6 months) after transplant. DAS was initiated at the dose of 100 mg in all patients. Temporary dose reduction or therapy withhold were required in 6 patients (32%) due to grade 2 hematologic toxicities (n = 2), pleural effusion (n = 1), liver toxicity (n = 1) and infectious complications (n = 2). The median duration of treatment was 11 months (range, 2-39 months). Fourteen patients still continued therapy at the last follow-up. Molecular response has been assessed in 18 patients so far. Sixteen patients (89%) achieved (n = 13/14) or maintained (n = 3/4) a complete molecular response. One patient who was given DAS prophylactically lost his molecular response and the other one did not achieve BCR-ABL transcript eradication despite preemptive treatment. Material (or patients) and methods: Here, we report the outcome of second allo-HSCT from the same donor in 21 patients (10 women, 11 men) with acute leukemia relapsing after allo-HSCT, treated in our center between 2004 and 2014. Sixteen patients (76%) were diagnosed with acute myelogenous leukemia (AML), five (24%) with acute lymphoblastic leukemia (ALL). Ages ranged from 15 to 67 years (median 35) at the time of first transplant. Patients received myeloablative (n = 15) or reduced-intensity (n = 6) conditioning regimen and peripheral blood stem cells (PBSC) (n = 8) or bone marrow cells (n = 13) from human leukocyte antigen-matched sibling donors (n = 14), matched unrelated donors (n = 5) or mismatch unrelated donors (n = 2). Only two patients presented chronic graft versus host disease (GvHD) symptoms. The median interval between first allo-HSCT and relapse was 12 (4-45) months. Six patients relapsed within 6 months, four later than 2 years. All patients received second allo-HSCT from the same donor as for the first transplant. Results: Chemotherapy was used as a reinduction for relapse in 17 patients, 5 of them did not achieve response to treatment. Disease status at the time of second allo-HSCT was: complete remission (CR) (n = 14), and nonCR (n = 7). Patients received myeloablative (n = 6) or reduced-intensity (n = 15) conditioning regimen. All patients but one received PBSC as a cell source. Neutrophil engraftment was achieved in all patient. No patients presented symptoms of acute GVHD. Chronic GVHD we observed in 5 patients. Only one patient died due to transplant related reason (hemorrhage to central nervous system). Relapse of disease occurred in eleven patients (52%), nine patients died due to relapse. Median time between second allo-HSCT and relapse was 6 months (range 2-30). After the median follow-up time of 20 months (6-96 months) ten patients (48%) remain alive in remission of disease. Median overall survival was 31 months (95% CI 11-50). The one-year and two-year overall survival was 79% (95% CI 62-96) and 52%(95% CI 26-77), respectively. (GVHD) prophylaxis was applied all of them as cyclosporine plus methotreksat (N = 37) or cyclosporine plus mycophenolate mofetil (N = 1) or tacrolimus plus mycophenolate mofetil (N = 1). GVHD was occurred in 41% of them (28.2% acute GVHD, 17.9% chronic GVHD). Grade I skin aGVHD was occurred most frequently of them (N = 9). Veno-occlusive disease (VOD) was occurred in 3 patients and only one of them received prophylaxis for VOD as defibrotide. Transplant related death was observed in 10 patients. Overall survival (OS) and disease-free-survival (DFS) were evaluated as median 9 (range, 0.5-92) and 8 (range, 0-92) months, respectively. One year OS and DFS were 41% and 38.5%, respectively. Overal survival in patiets who diagnosed over 25 years old were worse than others (kaplan meier; long rank; P = 0,009). Analyse of OS and DFS was illustrated in figure. Ten patients are alive without disease. Conclusion: Allo-HSCT is an effective treatment for increasing of long term survival in both newly diagnosed and relapsed/ refractory ALL. Transplantation related death and GVHD are most frequent complications so choice of patient's risk group must be well defined. Disclosure of Interest: None declared. The Ecotropic Viral Integration Site 1 (EVI1) protein is located on chromosome 3q26 and is overexpressed in 6 to 11% of Acute Myeloid Leukemia (AML). The most significant chromosome 3 lesions are t(3;3) and inv (3), which confer a negative prognosis. EVI1 expression can also be increased without chromosome 3 abnormalities, and predicts a poor outcome. Allogeneic hematopoietic stem cell transplantation (HSCT) is the most effective treatment in first complete remission (CR) in case of EVI1 overexpression, but whether HSCT is sufficient to erase its negative impact remains unanswered. Material ( Tai-Cheng Stem Cell Therapy Center, National Taiwan University, 2 Internal medicine department/Hematology division, National Taiwan University Hospital, Taipei, Taiwan, Province of China Introduction: Acute myeloid leukemia (AML) is a clinically and biologically heterogeneous hematologic malignancy, characterized by uncontrolled hematopoietic precursors proliferation and impaired differentiation. In order to provide the best treatment for every patient, risk-stratification is very important. Recently, more and more prognostic factors were found and many of them changed the indications and timing of hematopoietic stem cell transplantation (HSCT). However, literature about the prognosis of HSCT for patients achieving the first complete remission (CR) after two cycles of induction chemotherapy remained scarce. Material (or patients) and methods: Totally, 254 adult patients at the National Taiwan University Hospital with complete clinical, cytogenetic, genetic mutations, and laboratory data were recruited for this study. The induction chemotherapy include a standard "3+7" regimen as Idarubicin 12 mg/m 2 for three days and Cytarabine 100 mg/m 2 for seven days. Adjusted dosage to 2+5 for elderly patients was also classified as standard induction chemotherapy. Regarding to the pre-HSCT status, 104 patients had the first CR after one cycle of induction chemotherapy (abbreviated as CR1), 67 patients had the first CR after two cycles of induction chemotherapy (abbreviated as delayed CR1), and 83 patients had relapse and re-gained CR after re-induction chemotherapy (abbreviated as CR2 In multivariate Cox proportional hazards regression analysis for OS, reduced intensity conditioning, acute graft-versus-host disease more severe than grade two, and delayed CR1 were independent poor prognostic factors (HR = 2.5, P = 0.011; HR = 3.4, P = 0.023; HR = 2.0, P = 0.038, respectively). [P304] S273 Conclusion: AML patients achieving CR1 after two cycles of induction chemotherapy had a significant poorer OS and higher relapse rate. Further clinical trials are warranted to improve the prognosis of these patients. Disclosure of Interest: None declared. The prognosis of Isolated extramedullary relapse is better than bone marrow relapse for acute leukemia post allogeneic stem cell transplantation at complete remission In our cohort, 28 (9.1%) patients had extramedullary relapse in terms of isolated (n = 19) or concurrent with bone marrow relapse (n = 9). The top 3 extramedullary relapse sites were CNS, soft tissue, and mediastinum. The median overall survival for patients with any relapse was much worse than that for patients without relapse (15.9 m vs not reached, and 5-yr OS 18.4% vs 76.8%, P40.0001). We further dissected patients with any relapse into 3 groups: isolated extramedullary relapse group (iEM), bone marrow relapse group (BM), and concurrent extramedullary plus bone marrow relapse group (EM+BM). iEM group has significant better progression free survival (Fig 1 A, PFS, 11.2 m vs 6.9 m vs 3.7 m, P = 0.0118) and overall survival (Fig 1B, OS, 61.1 m vs 15.9 m vs 11.6 m, P = 0.0158), respectively. We also analyzed the outcome according to site of isolated extramedullary relapse, and in fact, PFS and OS were similar for CNS vs non-CNS relapse group (Fig 1C, We also tried to identify the risk factors for isolated extramedullary relapse using univariate and multivariate analysis, and we found that extremely high white count (4100k at diagnosis) was a risk factor (HR = 2.823, 95% CI = 1.073 to 7.423, P = 0.0354). Of note, isolated extramedullary relapse was not affected by acute GVHD or chronic GVHD (P = 0.6923 and P = 0.6439, respectively). 10 Institute of Transfusion Medicine, University of Ulm, Germany, 11 DRST -German Registry for Stem Cell Transplantation, Ulm, Germany Introduction: Even though the role of HLA-E in both, innate and adaptive immunity has long been described, its effects in hematopoietic stem cell transplantation (HSCT) remain markedly elusive. The aim of this study was to investigate the impact of HLA-E as alloreaction-mediator on HSCT outcome in an acute leukemia (AL) setting, as it has been reported that HLA-E expression in leukemic cells can be retained despite the downregulation of HLA-class I molecules, often observed in these cells. Material (or patients) and methods: 552 AL (AML:329, ALL:151, AL:72) patients (P) undergone 10/10 HLA matched unrelated HSCT (MUT) and their donors (D) were HLA-E genotyped by sequence based typing. Three alleles (01:01, 01:03, 01:07) and four genotypes (01:01/01:01, 01:01/01:03, 01:03/01:03, and 01:01/01:07) were identified. The effect of P and D HLA-E genotype as well as of P-D HLA-E match grade were assessed using univariate Kaplan-Meier (KM), multivariate Cox regression, and competing risks analyses. OS (overall survival), DFS (disease free survival), RI (relapse incidence) and TRM (transplantation related mortality) were set as endpoints, and statistical significance was set to a Po0.05. Results: The HLA-E frequencies found in our cohort for both P and D were in accordance with those previously reported elsewhere for Caucasian populations (01:01 = 55.5-56.4%, 01:03 = 43.5-44.5%, 01:07 = 0.1%). 348/552 (63%) pairs were HLA-E matched (M) and 204/552 (37%) were HLA-E mismatched (MM) with balanced distribution of patients in each group with respect to age and disease stage. KM analysis revealed notably improved 5y OS (52.6% vs 37%, P = 0.001) and DFS (44% vs 32.7%, P = 0.008) rates in the HLA-E MM compared to the HLA-E M pairs. Multivariate Cox regression analysis confirmed these results for both, OS (HR = 0.66, CI = 0.51-0.87, P = 0.003) and DFS (HR = 0.73, CI = 0.58-0.93, P = 0.012). No difference in RI was observed between the two groups, however, the MM group exhibited significantly lower TRM rates as shown by both, univariate (18.3% vs 27.5%, P = 0.022) and multivariate (HR = 0.65, CI = 0.43-0.98, P = 0.04) analyses. P as well as D HLA-E genotypes showed no particular effect on any of the study endpoints. Conclusion: Our results are indicative of an association between HLA-E MM and better HSCT outcome in AL patients. A putatively enhanced NK-mediated alloreactivity due to HLA-E MM may account for the observed beneficial effect, as it has been demonstrated that NK alloreactivity is associated with lower GvHD rates. In vitro studies could prove helpful in unraveling the biological mechanisms underlying our observations. In conclusion, our study is to our knowledge the first to describe the effect of HLA-E matching in an AL MUT context. Further confirmational studies on independent large cohorts are required before further conclusions can be drawn. Disclosure of Interest: None declared. Introduction: Pneumonitis intestinalis is a very rare clinical condition that may occur following hematopoietic stem cell transplantation (HSCT). We aimed to review this topic because of our three cases. Material (or patients) and methods: Pneumonitis intestinalis was found in 3 out of 382 patients (0.8%) undergoing allogeneic HSCT in the Hematology, Anadolu Medical Center Hospital, Bone Marrow Transplantation Center between June 2010 and July 2015. The cases are presented below. Case #1. The patient diagnosed as having AML underwent allogeneic HSCT from his HLA-matched sibling. Abdominal pain occurred on the 54 th post-transplantation day. Pneumonitis intestinalis was found on the abdominal computerized tomography (CT) examination ( Figure 1 ). The patient for whom no surgical intervention was considered was recovered with antibiotic and supplementary treatment and was discharged from the hospital on the 80 th posttransplantation day. Case #2. The patient with diagnosis of ALL had had fever on the 178 th post-transplantation from HLA-matched HSCT. Fever was controlled by antibiotic treatment but the patient developed abdominal pain and distention. Pneumonitis intestinalis was found on the abdominal CT examination. Oral intake was stopped and antibiotic and supplementary treatment was started. The symptoms disappeared in 10 days. On the abdominal CT, the findings were observed to disappear. Case #3. The patient with diagnosis of AML developed abdominal pain, diarrhea, fever, and dyspnea on the 134 th post-transplantation of allogeneic haploidentical HSCT. CMV pneumonia and sepsis were considered. Pneumonitis intestinalis was found on the abdominal CT examination of the patient whose abdominal pain continued despite treatment. No surgical intervention was made and antibiotic treatment, fluid replacement, positive inotropic agents, and renal replacement therapy (CRRT) were given and parenteral nutrition was continued. Symptoms regressed within 15 days. The findings were observed to disappear on the abdominal tomography taken 10 days later. The most commonly implicated causes for development of pneumonitis intestinalis include chemo-and radiotherapy, GVHD, and corticosteroids. It may manifest with abdominal distention, pain, diarrhea, and rectal bleeding. The right intestine is preferably involved. Radiological involvement is usually in linear manner parallel to the intestinal wall. Most subjects recover spontaneously with conservative approaches while surgery is rarely indicated. Conclusion: "Pneumonitis intestinalis" should be considered in differential diagnosis for the patients developing clinical condition similar to acute abdominal pain following HSCT. For such patient: 1) Abdominal CT examination is recommended for making diagnosis of pneumonitis intestinalis. 2) Oral intake should be stopped, parenteral nutrition should be started, nasogastric catheter should be inserted and gastric decompression should be applied, 3) Effective antibiotic treatment should be given, and 4) Recovery is usually observed in 2 to 4 weeks with efficient treatment. Introduction: Endothelial Progenitor Cells(EPCs) play a significant role in blood vessel integrity. Chemotherapy is related with side effects such as obesity and hypertension. Additionally recent studies show that obesity has negative correlation with the circulating EPCs,suggesting a negative affection in vascular repair. Aim:The study of circulating EPCs in children who received chemotherapy for acute lymphoblastic leukemia(ALL) and solid tumors(ST) and the evaluation of their levels in correlation with patients' Body Mass Index(BMI) and blood pressure(BP) regarding the years following treatment. Material (or patients) and methods: Peripheral blood from children with acute lymphoblastic leukaemia (ALL,n = 48) and children with Solid Tumours at diagnosis(ST,n = 38) were studied.Four colour flow cytometry was performed to evaluate the subpopulations CD34+CD45negdimCD133+, CD34+CD45negdimVEGFR2+ and CD34+CD45negdimCD133 +VEGFR2+ of circulating endothelial cells. The mean time from the end of therapy in ALL was 2.79 years and ST 3.03 years. The BMI of the patients was calculated and the BMI percentile was established specific by the age and gender. Normal weight defined with BMI percentile over 5 th and bellow 85 th percentile, and overweight/obesity over 85 th percentile. The systolic blood pressure(BP) was measured and the percentile was calculated specified by the age and gender.Normal BP was defined BP over 5 th and bellow 90 th percentile, and hypertension with a BP over 90 th percentile.The post treatment period of time divided in three groups under or equal of 1 year, 1 to 3 years, and equal and over 3 years. Results: In ALL group 62.5% of children were overweighted and 39.58% hypertensive.The percentage of CD34(+)VEGF(+) were 0.0044% and CD34(+)CD133(+)VEGF(+)0.0039%. In ST group obesity estimated in 63.15% and hypertension in 39.48% of children.The CD34(+)VEGF(+) percentage was 0.008% and CD34(+)CD133(+)VEGF(+) levels 0.006%. No statistical significance was found between the ALL and ST in relation with the type of EPCs.Although in ALL as well as ST group there was no statistically significant difference regarding the years after treatment,there was a trend according which the 1 st year values were higher than the following years' groups.In ALL patients with normal BMI the CD34(+)VEGF(+) value was 0.005%,CD34(+)CD133(+)VEGF(+) 0.0048% and in obese children CD34(+)VEGF(+) 0.004%,CD34(+)CD133(+)VEGF (+) 0.0032% respectively. There was no significant difference between normal and obese but it appears a trend with higher percentage of EPCs in normal weight group related to the obese. In ST, although there was no significant difference between normal and obese regarding the EPCs populations, it appears an opposite trend compared with the ALL patients, with the normal weight children to have lower percentage of EPCs than the obese(normal:CD34(+)VEGF(+)0.0066%,CD34(+) CD133(+)VEGF(+)0.0048%, Obese:CD34(+)VEGF(+)0.0094%, CD34(+)CD133(+)VEGF(+)0.0072%). Concerning the blood pressure, there was no significant difference between normal S276 and hypertensive in ALL, but the trend in ST was opposite, the children with normal BP evaluated to have lower percentage of EPCs than the hypertension group. Conclusion: Circulating EPCs in both ALL and ST seem to be higher during the 1 st year after treatment. ALL patients with normal weight and BP have higher levels of EPCs, in contrast with ST, results that need further elucidation. Disclosure of Interest: None declared. Factors Predicting Outcome after Allogeneic Transplant in Refractory Acute Myeloid Leukemia: A Retrospective Analysis of the Gruppo Italiano Trapianto Di Midollo Osseo E. E. Todisco 1,* , F. Ciceri 2 , C. Boschini 3 , F. Giglio 2 , A. Bacigalupo 4 , F. Patriarca 5 , I. Donnini 6 , E. P. Alessandrino 7 , W. Arcese 8 , A. P. Iori 9 , P. Marenco 10 Introduction: Outcome of primary refractory (PRF) AML patients is poor with a minor proportion rescued by allogeneic hematopoietic stem cell transplantation (HSCT). The identification of pre-HSCT variables helps to identify PRF AML most likely to benefit from HSCT. We analyzed PRF AML patients transplanted in Italy between 1999-2012. Material (or patients) and methods: Outcome of 242 patients transplanted in 26 GITMO centers was analyzed. Patients disease status at HSCT included PRF AML defined as failure to achieve a complete response (CR) after one or more chemotherapy cycles. The cytogenetic and molecular risk was defined according to the European LeukemiaNet. The main clinical and outcome follow up data were retrieved from the GITMO database. The main end-points of the study were overall survival (OS) and leukemia-free survival (LFS). Results: The median age at HSCT was 49 years (18-72). Before HSCT, 58% received ≤ 2 chemotherapy cycles. After HSCT, 137 patients (57%) achieved CR, 71 patients did not (29%), and 34 (14%) died early; 69 non-responder patients died, the remaining 2 patients were alive at last contact, while among remitters, 66 patients maintained CR, of whom 32 died in remission and 34 survived. Seventy-one patients relapsed, 65 died of disease and 6 survived. Up to 42 patients were alive, 38 in CR and 4 with active disease. The median OS of the whole patient cohort was 5.7 months (IQR 4.5-7.1), while median follow up of survivor was 28 months (1.8-148) . LFS of responder patients is 23% at 3 years and OS of the whole patient population is 15%. At 100 days from transplant, cumulative incidence of aGvHD was 36% (10% grade III-IV). Cumulative incidence (CI) of cGvHD at 1 year after HSCT was 25% (11% extended). The 3-years CI of NRM and RI was 27% and 59%, respectively. By multivariate analysis risk factors significantly associated with survival were:42 chemotherapy cycles to achieve CR, (HR: 1.97; 95% CI: 1.30-2.96; P = 0.0012), a number of BM blasts greater or equal than 25% or any level of blasts in PB (HR: 1.72; 95% CI: 1.16-2.55; P = 0.0068), a Karnofsky PS less than 90 (HR: 1.50; 95% CI: 1.04-2.19; P = 0.0320) and being more than 60 years-old at transplant (HR: 1.60; 95% CI: 1.00-2.56; P = 0.0492). On the basis of these adverse prognostic factors, we set up a score that stratifies the population in 3 groups: the group with score 0 (no adverse prognostic factor or 1 prognostic factor), shows a 27% survival at 3 years. For patients with score 1 (2 adverse prognostic factors) 17% survival at 3 years, while patients with score 2 (3 or 4 adverse prognostic factors), 7% survival at 1 year ( Figure 1 ). Conclusion: The clinical outcome of PRF AML remains poor. The new simple clinical GITMO score helps indentifying patients most likely will benefit or not from the HSCT. Disclosure of Interest: None declared. Overall survival at 5 years was 69.1% for the 1 st CR group and 35.6% for the non 1 st CR group (P = 0.037). Event-free survival at 6 and 12 years was 61% for the TKI group as compared with 35.7% and 25.5% for the non-TKI group (P = 0.025); Event-free survival at 6 and 12 years was 58.7% and 42% for group in 1 st CR and 25.9% and 25.9% for the non 1 st CR (P = 0.0016). Transplant-related mortality at 100 days and 1 year was 18.6% and 28.6%. Moderate to severe acute GVHD was observed in 38.5% and cronic GVHD was 14.3%.In the multivariate analysis, we found the pre-transplant status as a risk factor: [HR 2.38 IC 95% (1.11-5.12 ); P = 0.026]; and the use of TKIs as a protective factor on overall survival: [HR 0.33 IC 95% (0.15-0.71); P = 0.0047]. Conclusion: Allo-SCT is nowadays the therapy of choice for pediatric Ph-ALL patients, showing better results basically for those in CR1 at the time of allograft. Patients treated with TKIs improved outcomes in OS and EFS. Our results match the current publications about the subject. Disclosure of Interest: None declared. Cytomegalovirus induces strong antileukemic effect in acute myeloid leukemia patients following sibling HSCT without ATG-containing regimen H. Qiu 1,* , X. Bao 1 , S. Xue 1 , X. Hu 1 , X. Ma 1 , F. Chen 1 , D. Wu 1 1 The First Affiliated Hospital of Soochow University,Jiangsu Institute of Hematology, Suzhou, China Introduction: A considerable number of studies have demonstrated that cytomegalovirus (CMV) reactivation after allogeneic hematopoietic stem cell transplantation (Allo-HSCT) could enforce graft-versus leukemia (GVL) effect in acute myeloid leukemia (AML) patients. However, the use of antithymocyte globulin (ATG) as part of graft-versus-host disease (GVHD) prophylaxis may dampen this beneficial effect of CMV replication. Material (or patients) and methods: In this context, we retrospectively analyzed the effect of CMV reactivation on relapse, survival and prognosis in a total of 227 AML patients who received a myeloablative (MA) conditioning regimen at a single research center between January 2010 and April 2013. Results: Of these 227 patients, 110 cases received non-ATGcontaining regimens and 117 cases received ATG-containing regimens. CMV reactivation occurred in 45 patients (41%) among non-ATG regimen group and 73 patients (62%) among ATG regimen group (P = 0.001). At a median time to follow-up of 27.5 months, a lower risk of cumulative relapse incidence associated with CMV reactivation was observed in non-ATG group in multivariate analyses (OR 0.28, 95% CI 0.10-0.79; P = 0.016). However, CMV reactivation after transplantation did not significantly decrease the cumulative incidence of relapse in our ATG group (OR 0.28, 95% CI 0.10-0.79; P = 0.016). Conclusion: Collectively, our results demonstrate that in AML patients following sibling HSCT, the CMV-induced beneficial effect on relapse occurs only in the MA regimens containing no ATG, although ATG promotes CMV reactivation. Disclosure of Interest: None declared. [P314] frequencies were quantified in each sample in order to estimate tumor burden for clonal cell populations. Conclusion: These findings support the idea of the acquisition of additional somatic mutations required for AML initiation. These data demonstrate the loss at DCL relapse of subclones present at DCL onset, which are eradicated by the therapy. On the other hand, other subclones grow after therapy, either because they are primary resistant to chemotherapy or due to the acquisition of new mutations leading to resistance. Some of the mutations identified affect genes participating in several well-described pathways that are known to contribute to cancer pathogenesis. On the contrary, the role of other variants it is not clear in this context, making necessary to study its recurrence in other AML samples. However, further studies would be necessary to rule-out potential preleukemic variants in the UCB donor. Disclosure of Interest: None declared. Introduction: EBMT Acute Leukemia Working Party published recently a registry based data mining prediction model to predict mortality after alloHSCT in acute leukemias (AL). In the model, 9 variables are used to assess the day100 and 2-year TRM, including age, Karnofsky, disease and its status, time from Dg to transplant (more or less than 142 days), CMV serostatus of donor and patient, donor type, conditioning and center experience (more or less than 20 annual transplants). We retrospectively evaluated the usefulness of this prediction model in allotransplanted patiens with AL at our institution. Material (or patients) and methods: The data on 94 consecutive alloHSCT patients with AL during 2010-2014 were evaluated, and the TRM score by the EBMT prediction model was created through the web calculator (http://bioinfo.lnx.biu.ac.il/ ∼ bondi/web1.html) together with the Gratwohl score. The scores were compared between the patients who succumbed in the posttransplant period and those without TRM. Results: The characteristics of the patients were: 53 males/41 females with AML (n = 70) or ALL (n = 24), median age 53 (range: 16-68) years, median time from Dg to transplant 5.5 (2.3-208) months, disease status CR1 in 61 (65%), CR2 in 17 (18%) and advanced disease in16 (17%) patients, sibling donor 28 (30%) and MUD 66 (70%), and MA conditioning for 62 (66%) and RIC for 32 (34%) patients. Of the 94 patients 17 patients succumbed in post-transplant complications: three within 100 days (3% TRM100d) plus 14 more within 2 years (18% 2-year TRM; Group 1); the rest (n = 77) experienced no TRM (Group 2). There were only three patients who met with day100 TRM. Their predictive probabilities were 3.8, 10.6 and 25.9%. The median EBMT predicition scores for Groups 1and 2 were: 100-day TRM probabilities 6.4 (3.7-25.9 ) % vs 7.5 (2.95-25.9 ) %, and 2-year probabilities 15.7 (9.8-28.3) % vs 18.6 (9.8 -28.3) %. The median Gratwohl scores were 3 for the both groups. The patients with the highest calculated TRM probabilities (20-30%) were distributed as follows: 1 out of 3 who met with 100 day TRM (33%), two out of 17 who met with 2-year TRM (12%), and among the patients without TRM 10 were calculated to have the high score for day100TRM probability (13%) and 31 patients for 2-year TRM (40%). Conclusion: In spite of the large registry data basis for the TRM prediction model it seems that in real life the value of the predictive scoring at an individual level is very limited. An apparent explanation is that several critical items in the TRM probability scoring have already been taken into account in the decision making by the transplant team, reflected by the fact that the highest predicted TRM probability in our cohort was 28.3%.. Disclosure of Interest: None declared. higher for patients transplanted in CR versus those transplanted in active disease: 21.9% vs. 15.8% (P = 0.05) and 32% vs. 17.5% (P = 0.06), respectively. In multivariable Cox analysis CR at transplantation was the only factor associated with LFS and OS. Complete remission at transplantation was also the only significant factor for decreased NRM (P = 0.02) and improved GRFS (P = 0.01). Conclusion: The analysis of allo-SCT in AML pts with t(3;3)/inv (3) indicates that the t(3;3)/inv(3) leukemia remains an extremely high risk disease with low LFS and OS probability even with transplantation. Results are better for pts transplanted in CR. Novel agents and treatment modalities are most probably indicated to improve outcome especially for pts not achieving CR before transplantation. Disclosure of Interest: None declared. requiring inotropic supportive care, developed in two cases (16%). Five (42%) patients had aGvHD 2-4, one of them after DLI, in 2 cases (12%) aGvHD III-IV was observed. Chronic GvHD was observed in 4 cases (36%). Eleven (91%) patients are alive, in complete remission with a median follow up of 5 months (range 2,7-11). Conclusion: We demonstrate that tocilizumab can be safety administered to children with acute leukemia in the context of Treosulfan/Melphalan-or TBI/VP-based conditioning regimens. This limited experience suggests that post-transplant tocilizumab does not provide acute GVHD control equivalent to polyclonal serotherapy. Long-term follow-up will demonstrate if the combined administration of tocilizumab and bortezomib post-TCRαβ+/CD19+-depleted grafting will improve graftversus-leukemia effects without exessive GVHD-related morbidity and mortality. Disclosure of Interest: None declared. Introduction: Allogeneic Hematopoietic Stem Cell Transplantation (allo-HSCT) is potentially curative for Acute Myeloid Leukemia (AML), mainly through the "graft-versus-leukemia" (GVL) effect, which leads to reduced risk of relapse. In Brazil, intermediate risk AML patients are usually submitted to this procedure if a matched sibling donor is available while patients without a sibling donor are submitted to consolidation chemotherapy. Material (or patients) and methods: The aim of this study was restrospectively compare the results of Allo-HSCT or chemotherapy in intermediate risk AML patients in first Complete Remission (CR1) at the Hematology Service and Bone Marrow Transplantation from Hospital de Clínicas de Porto Alegre, southern Brazil, from April 1st 1999 to October 1st 2014, with at least, one year follow-up after treatment. Results: Among the 69 patients analyzed, 45 were submitted to consolidation with chemotherapy (Intermediate risk AMLnon allo-HSCT) and 24 submitted to allo-HSCT (Intermediate risk AMLallo-HSCT). The average age of Intermediate risk AMLnon allo-HSCT was 47.8 years old and Intermediate risk AMLallo-HSCT was 35.5 years of age ( p40,001). There was no difference regarding gender between the two groups. The median follow-up in the non allo-HSCT was 1.1 years (interquartile rage of 0.4 to 2.5) and in the allo-HSCT cohort of patients was 2.7 years (interquartile rage of 0.4 to 5.5) (P = 0.236). Survival in one year for the non allo-HSCT group was 52.3% and 62.5% for the allo-HSCT. The 2 and 5 years survival was 31.7% and 21.1% for the non-allo-HSCT and 58.3% and 53.8% for the allo-HSCT, respectively. LongRank test indicates a statistically significant difference in survival between groups after 5 years, with hazard ratio (HR) for death of 2.2 (IC95% 1.1-4.2)(P = 0.027), but when we adjust to age, this association loses statistical significance (HR:1.6 95%CI: 1-1.1; P = 0.246). Conclusion: Our results, in accordancy with the literature, suggested a better survival rate for the group submitted to allo-HSCT. However, despite loosing statistical significance when adjusting for age, the adjusted Hazard Ratio (HR) remains higher to the non allo-HSCT group. Although it can be explained by the small number of the patients in our cohort, identifying which patients will benefit from allo-HSCT is increasingly challenging. With the advent of nonmyeloablative conditioning as an alternative to older patients with comorbidities, better supporting therapy, better understanding of molecular marker in risk classificantion, and based in our experience, we need to be less conservative in indicating allo-HSCT. Disclosure of Interest: None declared. immunosuppression is an effective approach and should be considered in the patients allografted for ALL. University Hospital of Salamanca-IBSAL, Salamanca, 2 Hospital de León and Ibiomed, León, 3 Hospital Miguel Servet, Zaragoza, 4 Hospital Nuestra Señora de Sonsoles de Avila, Ávila, Spain Introduction: Mutations impacting epigenetic mechanisms, such as DNMT3A mutations, are recurrent in acute myeloid leukemia with intermediate-risk karyotype (AML-IR), although their prognostic value remains unclear. The prospective determination of DNMT3A mutations may provide an important step forward individualized therapies using hypomethylating agents or highdose daunorubicin. However, the clinical benefit of allogeneic stem cell transplantation (HSCT) has not been established. Aims: 1) To evaluate the incidence of DNMT3A mutations, their prognostic impact as well as the interaction with other molecular markers in AML-IR patients; 2) To determine if allogeneic HSCT in first remission improves the outcome of patients with somatic mutations in the DNMT3A gene. Material (or patients) and methods: A cohort of 127 patients with AML-IR (median age 55 years) (PETHEMA AML-99-2010). Analyses of DNMT3A mutations (exons 10 to 23) were characterized by direct sequencing in diagnostic bone marrow (BM) samples. One hundred and thirteen patients (89%) achieved complete remission (CR) and 72 of them underwent an allogeneic bone marrow HSCT, 58 in first remission. Results: Of the 127 patients included in this study, 36 (28%) had mutations in DNMT3A (25 had R882 mutations and 11 other missense mutations). DNMT3A mutations were frequently detected in FLT3-ITD mutated patients (48% vs. 21%, P = 0,007). No significant differences were found in both groups with regard to age, WBC counts or BM blast percentages. The median follow-up time was 693 days (range, 28-4.610). AML-IR patients with R882 mutations had shorter relapse-free survival (RFS) (21% vs. 78%, P = 0,013) at 5 years than the non-mutated subgroup (n = 80) (Figure 1a) . We next examined the relapse risk after allogeneic HSCT in first CR in order to elucidate if poor outcome of DNMT3A mutations was overcome. Again, DNMT3A R882 mutations identified a population of patients with substantially shortened RFS at 5 years (42% vs. 73%, P = 0,031) (Figure 1b) . In multivariate analysis, only FLT3-ITD mutations ( p40.0001) had a negative prognostic impact on RFS. Conclusion: 1) R882 mutations in DNMT3A identify a subgroup with poor prognosis in the intermediate-risk group. 2) Although based on small numbers of patients, these data suggests that mutations in DNMT3A identify a significant fraction of HSCT recipients with poor survival, for whom alternatives to standard transplantation options should be considered. Disclosure of Interest: None declared. Introduction: Allogeneic stem cell transplantation (ASCT) appears to be a treatment of choice in patients with intermediate-risk and high-risk acute myeloid leukemia (AML). Many studies show that the prognosis and clinical outcome in these patients was assessed by detection of minimal residual disease (MRD) in the bone marrow by multicolour flow cytometry both before and after ASCT. The aim of the study was to evaluate the clinical relevance of the level of residual leukemic cells in the bone marrow of AML patients before and at day 100+ after ASCT defined by 8-colour flow cytometry. Material (or patients) and methods: In the present study, 28 patients with AML who received ASCT at the National Haematology Hospital, Sofia were included. The patients' cohort comprised 19 males (68%) and 9 females (32%) at a median age of 41 (ranging 22-59). Transplantation from related donors was performed in 14 patients and from unrelatedin 14. The level of MRD was determined by 8colour combinations of monoclonal antibodies matching the patients' specific phenotypes and flow cytometry (FACSCanto II, BD) before ASCT and at day 100. Patients were stratified into two groups, according to the MRD levela "low MRD" group with MRD levelo 0,1% of bone marrow nucleated cells, and a "high MRD" group with residual leukemic blasts 40,1%. The following events were recorded and included in the analysis: conditioning regimen, peripheral cell engraftment, acute (a) GvHD, viral reactivation and deaths. Overall survival (OS) was defined as the time from transplantation until death, irrespective of the cause; and Disease-free survival (DFS)as the time to relapse/recidive. Results: MRD was evaluated prior to ASCT in 27 pts. In 10/27 pts (36%) the quantity of residual leukemic blasts was o0,1%, and in 17/27pts (60%): 40,1% (ranging 0,004%435%). The MRD evaluation at day 100+ after ASCT revealed levels o0,1% in 21/28 pts (75%), and only 7pts (25%) 40,1% (ranging 0-2,9%). Importantly, 11 pts, in whom the MRD prior to transplantation was high 40,1%, converted to the "low MRD" grouPo 0,1% at day 100, while one "low MRD" patient become positive after transplantation and presented with clinical relapse. The amount of residual leukemic cells before transplantation did not correlate with engraftment, the donor Introduction: aplastic anemia (AA) is a life-threatening hematological condition. Immunosuppressive therapy (IST) with ATG plus cyclosporine A (CsA) is usually employed for patients non-candidate for first line allo-HSCT. The reported overall response (OR) to 1 st course of IST has ranged from 37% to 80%, with around half of them complete responses (CR). Response rate after a 2 nd IST in patients unresponsive to a first course has been published to be of 77% (Di Bona, British Journal of Haematology 1999) . The aim of this study was to know the chances and timing of response of AA patients receiving a 2 nd IST course. Material (or patients) and methods: a total of 50 AA patients received a second IST course with ATG plus CsA during the 2000-9 decade in several Spanish centers. The characteristics of the series are shown in table 1. The type of ATG received by the majority of patients was Thymoglobuline at first (76%) and second course (82%); the median dose was generally higher after the 2 nd compared to the 1 st IST course (3.5 versus 2.5 mg/ kg/day/5 days). Results: overall responses at 3, 6 and 12 months were 38.8%, 49.0%, and 55.3%, respectively. Complete responses at 3, 6 and 12 months were 0%, 4.3%, and 14.9%, respectively. Six patients (12%) died during the first year after the 2 nd IST, the majority of them (5 out of 6) from infections. Five patients underwent allo-HSCT after failing the 2 nd course, two of which are alive at last visit. Conclusion: 1) In our experience, responses after a 2 nd IST course (49% at 6 months), although lower than previously reported (77%), are high enough to be considered a therapeutic choice for some patients. 2) As a consequence, we suggest that the decision-making process for the rescue treatment of AA patients unresponsive to first line IST should be performed in an individual basis, and include: an alternative donor HSCT, a 2 nd course of IST, and the employ of new drugs as Eltrombopag (alone or combined with ATG/CsA). Disclosure of Interest: None declared. Introduction: Severe aplastic anemia is a rare disease in childhood. First line treatment depends on patient age, the availability of a matched sibling donor and the severity of the disease. Since there is an available matched sibling donor, allogeneic transplantation is the preferred treatment. If no matched sibling donor is available, immunosuppressive treatment consisting of antithymocyte globulin and cyclosporin is the therapeutic option. However, if there is no response to immunosuppressive treatment, hematopoietic stem cell transplantation (HCT) with an alternative donor should be performed.Cord blood is used for unrelated allogeneic stem cell transplantation, however there are a few reports of successful autologous cord blood transplantation in patients with SAA. Material (or patients) and methods: We describe 2 cases with SAA treated successfully with autologous cord blood transplantation.Patient 1: A 15 month male infant was diagnosed with severe aplastic anemia (SAA). SAA was considered secondary to H1N2 infection. No matched sibling donor was found, however autologous cryopreserved cord blood was available for the infant and autologous cord blood transplantation was considered feasible for the patient. The conditioning regimen consisted of Cyclophosphamide+ATG with Cyclosporine A for immunosuppression. Cellularity of the graft: mononuclear cells: 1,6 x 10 7 /Kg, CD34: 1,6 x 10 5 /Kg). There was successful engraftment with neutrophils4500/μL on day27 and platelets4 20.000/μL on day37. With a follow up of three years the patient is well with normal hematopoiesis.Patient 2: A 2-year old male was diagnosed with SAA 3 months after vaccination. As there was no matched sibling donor for the patient, he was given immunosuppressive treatment with ATG and Cyclosporine A with no response so he was referred to our unit for further treatment. Since there was cryopreserved autologous cord blood available, this graft was given to the patient. The conditioning regimen included Cyclophosphamide+ATG with Cyclosporine A for immunosuppression. Cellularity of the graft: mononuclear cells: 3,8x 10 7 / Kg, CD34: 1,25x 10 5 /Kg. There was successful engraftment with neutrophils4500/μL on day29 and platelet 420.000/μL on day27. Three years after transplantation the patient has full hematological and immunological recovery. Results: Cryopreserved autologous cord blood for personal use is not indicated for transplantation with the only exception of SAA in childhood. Conclusion: Under certain circumstances and when a matched donor is not available autologous cord blood transplantation could be a successful therapeutic option for patients with SAA. Disclosure of Interest: None declared. Introduction: Aplastic anemia (AA) is a rare, and life-threatening hematological disease. It is known that its incidence and prevalence might vary substantially among different geographic regions, and its presentation has been sometimes linked to environmental exposures. Montané et al reported an incidence of 2.34 per million inhabitants per year in the metropolitan area of Barcelona (Haematologica 2008) . Aim: to know the incidence and epidemiology of AA in a well-defined population. Material (or patients) and methods: Four general hospitals from the provinces of Álava, Guipúzcoa, Navarra, and Vizcaya, that cover together a population of around 1.97 millions inhabitants, participated in this study. All the patients diagnosed with AA were included. Data bases of the Hematology, Pathology, and Pharmacy Departments were used as a source of information. Results: During the period of time 2006-2015, 49 new cases were identified (2-8 cases/year). Incidence of AA was 2.49 per million inhabitants per year. Median age at diagnose was 51 years old , and gender was distributed almost equally between female and male. Almost half of the cases were severe or very severe, and 70% of the patients had transfusion needs. The most frequently employed approach at first line was immunosuppressive therapy with ATG/cyclosporine A (CsA) (65.3%). With a median follow-up close to four years, two thirds of the patients were alive. Period Introduction: Allogeneic hematopoietic stem cell transplantation (ASCT) from a matched sibling donor is the treatment of choice for severe aplastic anemia. In developing countries, the limitations include the high cost of the transplant and the post-transplant care and the lack of stem cells banks which eliminates the possibility of unrelated donors. In this abstract, we review the experience with transplant for constitutional and acquired aplastic anemia at our center. Material (or patients) and methods: A retrospective chart review of all patients diagnosed with severe aplastic anemia who underwent ASCT between August 2005 and August 2015 at American University of Beirut Medical Center was performed. Results: 24 patients (13 children and 11 adults) with severe aplastic anemia underwent ASCT. 22 donors were fully-matched siblings; the remaining 2 patients received bone marrow from HLA-high-resolution-matched first cousin and mother. Seven patients had received immunosuppressive therapy (IST) prior to transplant. Conditioning regimens included Cyclophosphamide/ ATG/Fludarabine in 20 patients as per Bacigalupo A et al (1) , TBI was added to the regimen in 9 adults and one adolescent. Material (or patients) and methods: Both patients have a previously described c.1246 T4C; p.Cys416Arg mutation of PIK3CD associated with HyperIgM syndrome and a variety of non-Hodgkin lymphomas 1 . UPN1 presented at 10 months with failure to thrive, LN, HS, pancytopenia, hypocellular bone marrow, low IgG with elevated IgM and was commenced on immunoglobulin supplementation. At 2 years RSV pneumonia necessitated ventilation. He also had marked cervical LN, chronically elevated IgA (≤ 29.3 g/L), elevated CD4+CD8+ dual positive T-cells (12%), recurrent dacryoadenitis and gross splenomegaly requiring splenectomy. LN controlled well with ciclosporin (CSA) therapy but subsequent complications included development of multiple autoantibodies (ANA, ANCA, SMA, anti-mitochondrial), AIHA, hepatitis, bronchiectasis, focal segmental glomerulosclerosis and colitis necessitating protracted inpatient TPN. His sister UPN2 had numerous problems with chest, ear and sinus infections, oronasopharyngeal lymphoid proliferation, LN, progressive panlymphopaenia, hypogammaglobulinaemia, autoantibody formation, recurrent shingles, progressive food intolerance with elevated IgE, and eczema. Results: UPN1 underwent 10/10 MUD BMT following fludarabine (Flu), melphalan and Alemtuzumab (Alem) at 9 yrs but developed primary graft rejection. Enteropathy necessitated second BMT from an alternative 10/10 MUD at 15 yrs following treosulfan 36 g/m 2 , Flu 150 mg/m 2 , Alem 0.9 mg/kg with CSA for GVHD prophylaxis. CD34-selected PBSC top-up was required due to ongoing transfusion requirements at 6 months post-BMT, resulting in transfusion independence and sustained full donor chimerism; mild anaemia secondary to compensated haemolysis persists at 8yrs post-BMT. HyperIgM resolved and IgA is minimally elevated. UPN2 underwent uncomplicated 9/10 MUD PBSC transplant at 22 yrs following Flu 160 mg/m 2 , iv Busulfan 12.8 mg/kg and Alem 1 mg/kg conditioning with MTX/CSA GVHD prophylaxis and is a full donor chimera at 3 yrs post-BMT. Both patients were able to cease immunoglobulin substitution and immune suppression, infection rates have fallen radically, gammopathy has resolved and autoantibodies have steadily decreased. They now only require annual hospital review. Conclusion: Despite severe pre-transplant morbidity these siblings with PIK3CD GOF mutations tolerated reduced intensity transplantation well, with major amelioration of immune deficiency, lymphocyte / immunoglobulin dysregulation and autoimmunity. . Ibrutinib irreversibly inhibits both Bruton's tyrosine kinase (BTK) in B cells and interleukin-2-inducible kinase (ITK) in T cells. These observations led us to hypothesize that ibrutinib ITK inhibition is augmenting the graft-versusleukemia (GVL) benefit by depleting Th2 skewed T cells while augmenting cytotoxic CD8 cells providing anti-tumor Graft-vleukemia (GVL) benefit. Material (or patients) and methods: Thus far, 12 alloHCT patients with relapsed CLL are receiving Ibrutinib 420 mg daily with prospective blood samples collected on an IRB-approved research protocol. Here we present blood immune phenotype characterization for 6 patients with at least 9 months follow-up. Results: B cell summary shows that Ibrutinib rapidly depletes CLL and CD19+CD38+CD27+IgD+ cells, previously described as pre germinal center (preGC), which have been shown to associate with cGVHD. In contrast, CD19+CD27+IgD-CD38memory cells are abundant and persist throughout Ibrutinib S291 therapy as does total IgG.T cell summary included 15 antibodies overall. Total CD3 + CD8 + T cell frequency is unchanged at frequencies averaging 630 cells /ul. In contrast, CD4 T cells are 50% reduced by 6 months Further analysis of CD4 intracellular promoter stains shows a 75% reduction of GATA3 expressing Th2 while in contrast Tbet Th1 CD4 cells remain unchanged. The CD45RA and CD45RO staining of CD4 T cells showed no relative change comparing memory and naïve phenotypes (not shown). In addition, initiation of ibrutinib causes an immediate depletion of CD4 + CXCR5 + BCL6 follicular T cells from circulation. CD4 + CD25 hi CD127 -Tregulatory cells are expectedly infrequent and persist following Ibrutinib therapy. Conclusion: This first comprehensive immune assessment of ibrutinib therapy following alloHCT demonstrates a dramatic depletion of Th2 cells and T follicular helper cells while CD8 T cells and Tbet+ Th1 cells persist. This reduction of T follicular helper cells and preGC B cells may benefit chronic GVHD patients as we are currently testing in a Phase I/II trial of Ibrutinib for steroid refractory cGVHD (NCT02195869). Ibrutinib therapy following alloHCT is well-tolerated with few infectious complications possibly due to persistent memory B cells and cytotoxic CD8 T cells. These persistent CD8 T cells may provide GVL benefit with observed conversion to full donor T cell chimerism and CLL MRD eradication. These results support clinical trials of ibrutinib maintenance following alloHCT testing GVL augmentation and cGVHD prevention. Disclosure of Interest: None declared. Introduction: Allogeneic hematopoietic stem cell transplantation (HSCT) is currently the only curative treatment option for patients with myelofibrosis (MF) although it is still associated with significant morbidity and mortality related to transplant complications, namely Graft-versus-Host-Disease (GvHD). In vivo and ex vivo T-cell depletion (TCD), including use of antithymoglobulin (ATG) or alemtuzumab, could contribute to prevent GvHD and hopefully improve the outcome. The aim of this study was to analyze the impact of TCD on transplant outcomes in the Swiss cohort of patients transplanted for MF. Material (or patients) and methods: We performed a retrospective study on 73 patients who underwent allogeneic HSCT for primary or secondary myelofibrosis in Switzerland between 1997 and 2015. 16 patients received T cell repleted (non-TCD) grafts. 57 patients received TCD grafts, obtained by in vivo ATG administration in the conditioning regimen (n = 54) and/or ex vivo TCD by incubation with alemtuzumab in vitro washed before infusion followed on day +1 by an add-back of donor T cells (n = 11). Differences between survival curves were determined using Log-rank Mantel-Cox test. Cox regression was used to examine the independent effect on OS of clinical factors including age (less or more than 65 years), form of MF (primary or secondary), disease status (AML transformation), DIPSS at HSCT (low, intermediate or high), conditioning (RIC or MAC), donor type (related or unrelated donor), matching (matched or mismatched), stem cell source (PBSC or BM), year of transplantation (before or after 2010) and T-cell depletion. Cumulative incidence estimates of clinically relevant (grade II-IV) acute GvHD (aGvHD) or chronic GvHD (cGvHD) were calculated with death as competing events. Cumulative incidence estimates of relapse were calculated with death from other causes defined as competing events. Results: TCD was associated with significantly improved 5-year OS (68%, 95%CI 48%481%) compared to non-TCD transplantations (39%, 95% CI 15%462%; P = 0.00749) [ Figure 1 ]. The positive effect of TCD on OS remained highly significant (HR 0.018, 95%CI 0.003-0.122; P = 0.00004) in multivariate analysis performed taking into account the aforementioned clinical factors. Cumulative incidence of aGvHD (grade II-IV) was significantly reduced in TCD-transplant recipients (34%, 95%CI 22%447%) compared with recipients of non-TCD grafts (65%, 95%CI 34%484%; P = 0.0124). Conversely, no significant impact of TCD was observed on cGvHD, NRM or relapse cumulative incidence. Conclusion: TCD appears to reduce aGvHD incidence and to improve OS in patients undergoing allogeneic HSCT for myelofibrosis. Disclosure of Interest: None declared. Introduction: Allogeneic hematopoietic stem cell transplantation (HCT) is a potentially curative treatment for CLL and is performed in fit patients (pts) with high-risk features. Almost exclusively, outcomes on matched related and unrelated donor transplantations in CLL have been published and data of alternative donors are scarce. Recently, mismatched related donors are gaining interest because of the better outcome of haploidentical HCT with post-transplantation cyclophosphamide (PTCY). Material (or patients) and methods: All patients with CLL who received a first allogeneic HCT with a mismatched related donor and whose data were available in the EBMT registry were analyzed. Median values and ranges are reported for continuous variables and percentages for categorical variables. The probabilities of overall survival (OS) and progression free survival (PFS) were calculated using the Kaplan-Meier method and the log-rank test for univariate comparisons. Relapse/ predominant joint involvement without CNS-injury (type2/3). Only few cases of FD submitted to HSCT are described in literature. Material (or patients) and methods: We describe a case report of 13 y. o. boy with FD type 2/3. The first symptoms of FD appeared at 6 months. The patient had widespread granulomatous polyarthritis with contractures, hoarseness, subcutaneous and tongue granulomas. Juvenile idiopathic arthritis (JIA) was suggested. All antirheumatic therapies (Steroids, Methotrexate, Cyclosporine A, Hydroxychloroquine) failed. At the age of 12 y.o. analysis of ASAH1 gene showed c.760 A4G mutation in exon 10 (p.Arg254Gly). At 13 y.o. allo-HSCT from partially matched unrelated (DRB1-missmatch) was performed. By the time of allo-HSCT patient had severe osteoporosis, bone destruction, joint subluxation and joint ankylosing, intensive pain syndrome, high paraclinical activity of disease, body weigh deficiency 45%. Reduced toxicity conditioning regimen, consisted of Treo 42 g/ m 2 , Flu 150 mg/m 2 , ATG 60 mg/kg, was used. CSA and short course of MTX were used for aGVHD prophylaxis. Results: Early post-transplant period was complicated with severe bacterial infection. After first allo-HSCT of BM engraftment was achieved on D+23, initial chimerism study on D+25 showed 70-79% of donor cells. Escalation of immunosuppressive therapy resulted in increasing of donor chimerism (90-97% of donor cells on D+33), clinically patient had gradually regression of subcutaneous granulomas, absence of pain syndrome. But then progressive decreasing of donor chimerism with graft rejection at D+80 was observed. On D+111 after first allo-HSCT the second allo-HSCT of PBSC from the same unrelated donor was done. A reduced intensity conditioning regimen (RIC) based on Flu 150 mg/m 2 , Mel 140 mg/m 2 , Thiotepa 10 mg/kg with serotherapy (ATG 60 mg/kg, Rituximab-375 mg/m 2 ) was used. Post-transplant cyclophosphamide (PT CY) 50 mg/kg on day+3,+4 and combination of tacrolimus and sirolimus were used for aGVHD prophylaxis. Engraftment was achieved on D+23 and PCR SRT showed full donor chimerism. During next months patient had complete resolution of granulomas, improvement of mobidity and joint motility. 16 months later patient is alive with full donor chimerism, without signs of GVHD. Conclusion: In previous reports myeloablative (conventional or reduced toxicity) conditioning regimens were used for FD pts, because early experience suggests that RIC are associated with high incidence of graft failure in metabolic diseases. In our case we observed graft rejection after MAC reduced toxicity conditioning. Our experience showed successful engraftment of FD patient and stable donor chimerism after RIC with additional immunosuppression with PT CY, tacrolimus and sirolimus. We suggest, that escalation of immunosupression may play role for successful engraftment after RIC. Introduction: Red blood cell exchange (RBCx) is indicated in multiple clinical manifestations of sickle cell disease (SCD). According to the 2013 American Society for Apheresis (ASFA) guidelines, RBCx is recommended for the treatment of several complications of the disease, such as ischemic stroke, acute chest syndrome (ACS) and the prevention of iron overload in patients with high transfusion requirement. In other situations, such as in the preparation of major surgery or during vasoocclusive crises (VOC), the treatment is controversial and therefore it must be individualized. Allogeneic hematopoietic stem cell transplantation (HSCT) is the only potentially curative treatment in SCD, even though it is burdened with frequent complications. Supportive therapy is essential to reduce the adverse effects and the transplant related mortality. In this regard, at the Pediatric Transplant Center of Padua, RBCx was introduced as a common practice in the pre-transplant period, with the aim of reducing HbS and inhibiting endogenous hematopoiesis. Material (or patients) and methods: Between March 2010 and August 2015 8 patients (3 males, 5 females, genotype: 7 HbSS and 1 HbS/β 0 ), aged 3 to 17 years, mean weight of 30 kg (range 16-55), were treated with a session of automated RBCx (Cobe Spectra/Spectra Optia, Terumo BCT) during the preparatory phase of HSCT. The indication to HSCT was cerebrovascular disease in 4 patients; in the remaining cases, the indications were recurrent ACS with pulmonary hypertension, hemolytic crisis and splenic VOC not responsive to hydroxyurea. The HSCT conditioning regimen included thiotepa, treosulfan, fludarabine. GvHD prophylaxis was obtained with anti-lymphocyte serum (Fresenius ATG) and cyclosporine. Five patients underwent transplantation from a related donor, 2 from a haploidentical parent, one from an unrelated donor. In 5 cases the patient and the donor had the same blood group, 2 patients had an major ABO incompatibility (B → A, A → O), one a minor ABO incompatibility (O → B). Results: All the procedures were well tolerated without complications. The mean value of HbS pre-and post-RBCx was, respectively, 42% (range 19-69) and 15% (9-23). The median time to neutrophil and platelet engraftment was 20 (range 16-30) and 22 (12-28) days, respectively; the median number of packed red blood cell units and platelet concentrates transfused was 4.5 (range 3-6) and 5 (2) (3) (4) (5) (6) (7) (8) , respectively. One patient experienced grade I acute intestinal GvHD, in another case reactivation of CMV and EBV associated with BK virus hemorrhagic cystitis occurred. All patients are alive at a median follow up of 21 months. Conclusion: In our experience in patients with SCD with organ involvement, pre-HSCT RBCx was a well tolerated procedure, able to substantially reduce the percentage of HbS. Although not included in the most recent ASFA guidelines, this procedure could be a precautionary tool to improve transplant outcomes. Disclosure of Interest: None declared. Material (or patients) and methods: Nine consecutive patients suffering from SCA who underwent HSCT between March 2014 and November 2015 were included in the study. Three underwent matched sibling donor bone marrow transplant (BMT), one patient underwent mismatched unrelated donor peripheral blood stem cell transplant (PBSC) and one underwent matched sibling cord blood transplant (CBT) using Busulfan@3.2 mg/kg/ day x 4 days, cyclophosphamid@50 mg/kg/day x 4 days, hATG (PfizerATGAM)@30 mg/kg/day x 3 days. One adult patient underwent matched sibling donor PBSC using reduced intensity conditioning with busulfan@3.2 mg/kg/day x 4 days, cyclophosphamid@60 mg/kg/day x 2 days, ATG(PfizerATGAM) @30 mg/kg/day x 3 days. Immune suppression for BM/PBSC patients was cyclosporine@3 mg/kg/day in 2 divided doses starting D-3 and methotrexate@10/m2 on D+1 followed by 7 mg/m2 on day 3, 6 and 11 post BMT and cyclosporine and methylprednisolone for CBT. Three patients underwent haploidentical HSCT using hypertransfusion (target Hb 11-13gm/dl) and hydroxyurea (20 mg/kg) from day -45, conditioned with Thiotepa 10 mg/kg in two divided doses (D-7) , fludarabine 30 mg/m2 (D-6 to D-2) , cyclophosphamide 14.5 mg/kg (D-5, D-4) , TBI 2 Gy with thymic shielding (D-1) , rATG (Genzyme Thymoglobulin 1.5 mg/kg (D-9 to D-7) . GVHD prophylaxis included PTCy 50 mg/kg/day on D3 and 4, tacrolimus to maintain a level of 5-15 ng/ml (till 6 months post HSCT) & MMF (till D35) starting from D5. Introduction: Mantle cell lymphoma (MCL) is associated with high relapse rates and poor survival when treated with conventional chemotherapy. The role of autologous hematopoietic cell transplantation (auto-HCT) is controversial. In this retrospective revision we want to share our experience in treatment with induction chemotherapy followed by consolidative auto-HCT. We will then compare different conditioning regimens. Material ( The conditioning chemotherapy regimens were BEAM (BCNU, etoposide, cytarabine, and melphalan) for 7 patients and BUCY 2 (busulfan, cyclophosphamide) for the other 13. Response rate to induction at auto-HCT time was 100% and complete response was reached in 70%. Two patients needed two lines of chemotherapy to achieve response. The overall survival (OS) and progression-free survival (PFS) were estimated by the Kaplan-Meier method. PFS was measured from diagnosis to disease relapse or progression and OS was measured from diagnosis to death by any cause. Results: Median PFS was 8.34 years (95% CI: 2.7-13.9). There is not significant difference between BEAM (median 8.33 years) and BUCY 2 (median 6.63 years) conditioning regimens (P = 0.958, long-rank test). The 5-year PFS rate was 63.8%, 53, 6% and 73.3% for all, BEAM and BUCY 2 chemotherapy groups respectively. Median overall survival was not reached. The 5-year OS rate was 85% in both groups, being 71% for patients who received BEAM conditioning schedule. There were no deaths in the group receiving BUCY 2 chemotherapy. Of the 4 patients which died, two did it due to disease progression and two from treatmentrelated side effects (AML and astrocytoma). . The goal of this study was to identify factors impacting the safety and efficacy of AHCT in the elderly NHL patients in order to better select those who will benefit from this intervention. Material (or patients) and methods: This single-center, retrospective study examine outcomes of AHCT in elderly patients ( ≥60 years old) with NHL. Between January 1st, 2008 and January 1st, 2015, 90 patients met the inclusion criteria S297 and were included in the study. Patients signed an informed consent and the ethics committee of our institution approved the study. Progression-free-survival (PFS) and overall survival (OS) were analyzed according to age at the time of transplantation, HCT-CI, lymphoma histology and disease status at the time of transplant. Toxicities were analyzed according to age and HCT-CI. Results: Median age at time of NHL diagnostic was 60 years (range 42 to 68) and 63 years at time of transplant (range 60 to 69). One-third (33%) of our cohort was ≥ 65 years old. Histologic subtype was mainly composed of follicular (36%), mantle cell (20%) and large B-cell lymphoma (38%). HCT-CI risk score was low in 34%, intermediate in 40% and high in 26%. The incidence of febrile neutropenia was 92% with 2% admission to the intensive care unit (ICU) with no difference between patients younger or ≥ 65 years old. Age ≥ 65 year was not associated with an increased transplant-related mortality (TRM) and was surprisingly associated with less total parenteral nutrition (P = 0,046) and narcotics use (P = 0,011). The median follow-up was 27 months (range, 1 to 87), median PFS was 46 months (CI 95% 24,4-67,6) and OS is not reached (graph 1). The estimated 5 years OS is 62% and PFS is 40%. TRM was only 1% at 100 days and 2% at 1year after transplant. The 1-year progression rate was 30% and mortality rate only 12%. Progressive disease status following first line therapy was associated with a worse PFS compared to the achievement of a partial or complete remission (HR 2,77; CI 95%, 1,18; 6,49). Progressive disease status at the time of transplant was also associated with a lower PFS (HR 9,30: CI 95% 2,55 to 33,92) and OS (HR 13,44: CI 95% 2,68 to 67,48). International Prognostic Index, age and treatment type did not influence PFS or OS. Surprisingly, HCT-CI score did not correlate with toxicities, morbidity or mortality. Conclusion: AHCT is safe and effective in selected elderly NHL patients. Progressive disease at the time of transplant was associated with worse PFS and OS. HCT-CI did not allow to predict outcome. Our data suggest that age alone should not exclude patients from transplantation. However, this approach should be reserved to patients with chemosensitive disease and avoided in the elderly patient with progressive disease. Disclosure of Interest: None declared. Introduction: Salvage multidrug therapy R-ESHAP with subsequent consolidation with autologous stem cell transplantation (ASCT) remains today the main regimen for relapsed/refractory patients with aggresive non-Hodgkin lymphoma (NHL) Multiple variables like response to salvage treatment, duration of first response, age,.. affect the outcome of the disease. We retropectively analyzed the influence of certain clinical and biological variables in response to both salvage treatment and its subsequent consolidation in relapsed and refractory diffuse large B-cell lymphoma (DLBCL). Material (or patients) and methods: We retrospectively analyzed 57 consecutive patients with refractory or relapsed DLBCL after first line treatment, who were treated with salvage chemotherapy R-ESHAP scheme between 2004-2014, and consolidated with ASCT (n = 30). Results: At the time of treatment with R-ESHAP, 37 patients had relapsed lymphoma and 20 refractory lymphoma. The median number of cycles of R-ESHAP received was 4 (1 to 6). The main factor influencing OS and EFS of patients with DLBCL is conducting ASCT (Figure 1 ) OS patients who received ASCT one year 66% vs 22.2% in those who did not take place; OS at 5 years 50% vs 12% (P = 0,000). SLE 1 year 56% vs 26% transplanted patients in the non-transplanted; DFS at 5 years 48% vs 6% (P = 0,000). ASCT was performed (with BEAM conditioning scheme) in 30 of the 57 patients. Reasons for not performing transplantation after the salvage therapy (n = 27) were: death of the patient because of infectious complications (9 patients), chemoresistant disease (11 patients), not candidate for ASCT because of poor medical condition (6 patients) and patient refusal (1 patient). In patients who underwent ASCT (n = 30), the main prognostic factor associated with OS and DFS was the status of pretransplant disease (figure 2), with 1 and 5 year-OS of 75% and 61% in patients in complete remission (CR) (n = 25) vs 21% and 0% for patients in partial remission/ refractory disease (n = 5) (P = 0.002). 1 and 5 year EFS was 63% and 58% in patients CR vs 20% and 0% in patients with partial remission/ refractory disease (P40.05). It has not been proved relationship between OS and EFS and other variables analyzed in the study, such as the disease status at the time of receiving treatment with R-ESHAP (relapsed vs refractory disease), or time from relapse ( o1 year vs41 year). Introduction: Allogeneic stem cell transplantation (alloSCT) is not considered a standard therapy for multiple myeloma (MM), but this intensive therapy can be beneficial for some patients (pts) as it offers the benefit of graft versus myeloma effect. Reduced-intensity conditioning (RIC) has broadened the use of alloSCT pts with MM as it reduces the toxicity associated with the procedure allowing older and heavily pretreated pts to receive alloSCT. Material (or patients) and methods: We retrospectively analyzed 95 pts, 85 with MM and 10 with plasma cell leukemia, receiving single or double UCBT in EBMT centers from 2001 to 2013. Results: Out of the 45 pts with available cytogenetic data, 32 had abnormal karyotypes including 11 with high risk alterations (e. g. t (4, 14) and del17p). Eighty-two pts received proteasome inhibitors or immunomodulatory drugs before UCBT. The median age at UCBT was 53 years (yrs) and 36% of pts was at stage III ISS. Thirty-six pts (38%) received a single UCBT and 59 (62%) a double UCBT (dUCBT); 89 pts (96%) received at least one previous ASCT. Disease status at UCBT was 1 st complete remission (CR) in 11%, 2 nd CR in 11%, very good partial response (VGPR) in 22%, partial response (PR) in 41%, stable or progressive disease in 15% of pts. Seventyseven pts (82%) underwent RIC, mostly (n = 60) cyclopho-sphamide+fludarabine+total body irradiation (TBI). Anti thymocyte globulin (ATG) was given to 24% of the pts. The median number of infused total nucleated cells (TNC) was 3.3 x10 7 /kg. The median follow-up was 41 months. The cumulative incidence (CI) of neutrophil (PMN) and platelet (PLT) engraftment was 97 ± 3% at 60 days and 72 ± 5% at 180 days after UCBT, respectively. Seven pts failed neutrophil engraftment, from which, 3 were alive at last follow-up (2 after ASCT rescue) and 4 died in a median time of 21 months after UCBT. The CI of 100-day acute graft-versus-host disease (aGVHD) grade II-IV was 41 ± 5%. The CI of chronic graft-versus-host disease (cGVHD) at 2 years was 22 ± 4%. The 3-year CI of relapse and of non relapse mortality (NRM) was 47 ± 5% and 29 ± 5%, respectively. Progression free survival (PFS) and overall survival (OS) at 3-year was 24 ± 5% and 40 ± 5%, respectively. Sixty-three pts died: 30 of relapse and 33 of transplant related causes (infections, n = 16, GvHD, n = 5 and others causes, n = 12). In multivariate analysis, ATG use was associated with decreased incidence of aGVHD (HR = 0.25, 95%, CI = 0.08-0.80, P = 0.020), higher NRM (HR = 3.35, 95% CI 1.44-7.81, P = 0.005), decreased OS (HR = 4.03, 95% CI = 2.13-7.64, P40.001) and decreased PFS (HR = 2.73, 95% CI = 1.48-5.05, P = 0.001). Patients with high cytogenetic risk had higher relapse rates (HR = 3.83, 95% CI = 1.26-11.61, P = 0.018) and worse OS (HR = 2.99, 95% CI 1.31-6.83, P = 0.009) and PFS (HR = 2.88, 95% CI 1.26-6.57, P = 0.012). Conclusion: This study shows that UCBT it is feasible and potential curative for MM pts. Results obtained after UCBT for MM are similar to what has been reported with other alloSCT sources. Disclosure of Interest: None declared. Progression free survival in patients with Multiple Myeloma treated with induction therapy followed by Autologous Stem Cell Transplant 5 years experience C. Plaza Meneses 1,* , T. Villaescusa de la Rosa 1 , J. L. Lopez Lorenzo 1 , C. Teran Benzaquen 1 , M. S. Sanchez Fernández 1 , E. Prieto Pareja 1 , E. Askari 1 , A. Velasco Valdazo 2 , P. Llamas Sillero 1 1 Hematology, Hospital Universitario Fundación Jiménez Díaz, 2 Hematology, Hospital Rey Juan Carlos, Madrid, Spain Introduction: Multiple Myeloma as today is considered an incurable disease. Multiple studies have demonstrated that the treatment that offers the longest remission is an induction therapy followed by an autologous stem cell transplant (ASCT) (6). Our group uses VCD (Bortezomib+Cyclophosphamide +Dexamethasone) as standard induction therapy, followed by an ASCT as front line therapy in eligible patients. Material (or patients) and methods: We analyzed data from January 2011 to November 2015 from four hospitals, in which we include all patients with multiple myeloma that have gone through an autologous hematopoietic stem cell transplant. The sample has 55 patients overall, divided in subgroups depending on the response to the induction therapy at the time of the ASCT. The subgroups are Complete Remission (CR), Very Good Partial Response (VGPR), Partial Response (PR), and Progression Disease (PD). We analyze the progression free survival time (PFS). The PFS event was defined as the time when the patient needed a new line of treatment. Results: Results show a mean PFS of 24 months (95% confidence interval 18, 9) . The subgroups are distributed in CR 16%, VGPR 36%, PR 42%, PD 6%. PFS after subgroup stratification was 36 months in CR patients, 21,9 months in VGPR patients, 21,1 months in PR patients and 0.3 months in PD patients. (Table 1) . Conclusion: After the statistical analysis, we concluded that multiple myeloma is still a disease that´s not very well controlled with standard care. Even with newer regimens that are coming it is still considered an incurable disease. In our study the progression free survival curves are very similar to the ones described in larger studies, the difference seen in relapse time is statistically significant (P = o0,001) when the patients have a complete remission during the induction phase of the treatment. In comes to our attention the practically nonexistent difference between VGPR and PR in progression free survival. We will confirm these finding after more time has passed since many of our patients do not have too much follow up time. We hope that the new lines of treatment will have the same impact that show in clinical trials can be replicated in he every day practice so we can improve the survival a PFS curves. with Durie-Salmon stage IIIA. Median follow-up was 67 months (95% Confidence Interval (CI) 60.0-74.0), median OS 62 months (95% CI 51. 4-72.7) , and median PFS 33 months (95% CI 27. 0-39.0) . TRD rate was 3%. Patients older than 60 years at time of ASCT had a worse PFS in comparison to younger patients (26 vs. 41 months, P40.001). 41% of the patients (N = 130) reached a very good partial remission (VGPR) or better at day +100 post-ASCT and showed a significant advantage in median OS compared to 127 patients responding with partial remission (PR) or worse (89 vs. 53 months; P = 0.001). Multivariate analysis of the entire cohort revealed age ≥ 60 years (vs. o60 years; P = 0.030; Hazard Ratio (HR) 1.922, 95% CI 1.1-3.5) and higher International Staging System stage (3 vs. 1; P = 0.006, HR 2.599, 95% CI 1.3-5.1) as adverse risk factors regarding PFS. Patients with one line of induction therapy before ASCT had a better median OS than patients with 2 or ≥ 3 lines of induction therapy (72 vs. 54 months vs. 30 months; P = 0.003). Median OS of myeloma patients receiving ASCT in the period from 2005 to 2012 was longer than in patients receiving ASCT in the period from 1996 to 2004 (71 months vs. 52 months, P = 0.027). Moreover, 65.1% of the myeloma patients achieved at least a VGPR on day +100 post-ASCT in the second time period compared to 29.5% in the first time period ( p40.001). Median OS was significantly worse in newly diagnosed MM (NDMM) patients (N = 158) receiving first induction therapy with a traditional regime than in patients (N = 87) treated with a novel compound, e.g. bortezomib, thalidomide, and/or lenalidomide (58 vs. 69 months, P = 0.01). Conclusion: Our analysis confirms HDCT-ASCT as an effective therapeutic strategy with low TRD rate in an unselected large myeloma patient cohort. Prognosis of HDCT-ASCT recipients has been improved by the introduction of novel induction strategies. Disclosure of Interest: None declared. Introduction: Autologous hematopoietic stem cell transplantation (ASCT) is a standard treatment option for newly diagnosed young patients (pts) with multiple myeloma (MM). The objective of this single-center study was to analyze prognostic factors predicting outcome of ASCT in pts with MM. Material (or patients) and methods: Between 2007 and 2015 100 consecutive pts with proved MM were enrolled in the study. Peripheral blood stem cells (PBSC) were collected and the first ASCT was performed independently of response to standard chemotherapy (CT). The second ASCT was recommended in case of partial response (PR), stable disease (SD), progression of disease (PD) or relapse after the first ASCT. Associations of patient and graft characteristics with outcome were evaluated in multivariate analyses, using Cox proportional hazards. Results: A total of 126 ASCT were performed in 100 patients. The median age of the patients was 51 years (range 29-64) and approximately half the patients were male (n = 49).The time interval between diagnosis setting and first ASCT was 5-43 mo (median 10 mo). In a period of 3,5-41 mo (median 7 mo) 26 pts received the second ASCT. Disease status at ASCT: complete remission (CR) 14%, PR56%, SD12%, PD18%. Conditioning regimen included melphalan in a standard dose of 200 mg/m 2 in all cases. The graft consisted of PBSC, median number of infused PBSC was 4,3 х10 6 CD34+cells /kg (range, [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] 8) . The median time to neutrophil recovery to more than 0,5х10 9 /l was 14 days (range, [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] . Recovery of platelets to more than 20х10 9 /l after transplantation occurred on day 12 (range, . Analysis of early post ASCT phase revealed that in most cases hematological and nonhematological toxicity was insignificant (1-2 grade*) to correct basic supportive therapy. Median duration of hospitalization was 22 days (range, 14-83). There were no cases of early transplant-related mortality. In a period of 1,5-3 mo after ASCT 33% pts achieved CR, 52%4 PR (including 31% very good PR), 9%4 SD, 6%4 further PD. After ASCT 12% of pts received thalidomide, lenalidomide or bortezomib as maintenance therapy. Estimated 5-year overall survival (OS) and progression-free survival (PFS) for all 126 ASCT was 69% and 45%, respectively. Median OS hasn't been reached, median PFS -36 mo (95% CI 16, 2) . On multivariate analysis achievement of a CR or very good PR after 1ASCT was associated with an improved OS (HR 2,7; 95% CI 1,1-6,7; P = 0,03), and PFS (HR 1,9; 95% CI 1,0-3,5; P = 0,04). Multivariate analysis revealed no significant relationship between outcome measure (OS and PFS) and features at diagnosis of MM, interval between diagnosis and 1ASCT, response to standard CT and all the remaining variables analyzed in this study (P40,05). Conclusion: Given the relative safety and efficacy of ASCT, this option should be included in the treatment program of all newly diagnosed young pts with MM with no absolute contraindications to the high doses of chemotherapy, regardless of the antitumor effect of the standard CT. *National Cancer Institute common toxicity criteria for autologous transplants, v.1.0. References: Child J. A., NewEnJMed, 2003 A., NewEnJMed, , 348(19):1875 Attal M., NEJM, 1996, 335 (2):91-7. Disclosure of Interest: None declared. Outcomes after autologous stem cell transplantation for patients with multiple myeloma from 2009-2013 in single cener Bulgaria (CIC 859) G. Arnaudov 1,* , P. Ganeva 1 , Y. Petrov 1 , G. Mihaylov 1 on behalf of Mincho Mintchev, Ivan Tonev, Chavdar Botev, Svetla Ivanova, Mntonia Mihova, Lidia Gartcheva, Margarita Guenova 1 National Hospital for Active Treatment of Hematological Diseases, Sofia, Bulgaria Introduction: Despite the introducing of new drugs in the treatment of multiple myeloma (MM), autologous stem cell transplantation (ASCT) remains a standard of care for patients under the age of 70 years. We summarize and analyze data of patients with MM who underwent ASCT at our clinic as primary therapy for 5 years period (2009) (2010) (2011) (2012) (2013) . Material (or patients) and methods: The analysis include 157 transplants in 145 patients with MM; 133 patients underwent a single ASCT and 12 double ASCT, (3 were tandem ASCT and for 9 the second ASCT was performed because of relapse). The median age was 56.4 years (41-67 years range); male/ female ratio was 85/72. Mobilisation was done according to standard procedure with GSCF (83%), Endoxan +GSCF(10.3%), Mozobil +GSCF (6.2%); mean yield of CD34+ 3.79 x10 6 bw. Conditioning regimen included Melphalan (dose according to the risk group or age). At the time of autograft, 55.5% had overall response rate (ORR = CR+VGPR); 38% had partial response (PR), 6% had relapse/refractory disease. We estimated overall survival (OS) and progression free survival (PFS) regarding the patient'age, ISS score, the time to ASCT( o 12 months of diagnosis and ≥ 12 months). The results were summurised according to EBMT criteria's for response. Results: Transplant related mortality (TRM) was 3.5%. Following ASCT, 71.3% of patients obtained ORR, 27.7% had PR or minimal response (MR) and only 1 patient (0.7%) progressed. OS was 68% at 4 years. PFS at 5 years was 18.5% whereas 50% PFS was 29 months. No prognostic significance was estimated for age with cutoff value of 55 years (P = 0.626) and time to ASCT with cutoff value of 12 months (P = 0.876) for OS. survival with major mismatch (P = .056). This was reflected in patients with pure red cell aplasia also having an improved PFS. The reasons for this are unclear and need to be studied in larger studies. Conclusion: T cell deplete allogeneic stem cell transplantion results in good overall survival in the MDS/MPD u categories with low TRM and relapse rates. Major ABO mismatch between donor and recipients impacts positively on PFS and needs to be validated in larger studies. Disclosure of Interest: None declared. The benefits of early allogeneic hematopoietic cell transplantation for the high/very high risk myelodysplastic syndrome patients who received hypomethylating agent therapy J. Y. relevant cause of cancer-related death in women worldwide. While the less aggressive variants of BC can nowadays be treated through a series of well-consolidated therapy strategies arousing generally a fairly good expectation of survival, the more severe disease formssuch as Triple-Negative BC (TNBC)actually do not benefit from well-assessed treatments. Recent data have shown that autologous HSCT may be associated with better outcome than standard chemotherapy for TNBC. Results: Since its original compartmentalization as a treatment for hematologic diseases, immunotherapy has emerged in the last decays as a novel effective therapeutic strategy also for some solid malignancies such as melanoma and prostate carcinoma. For breast carcinoma, an immunologic therapeutic approach has been barely evaluated as of today, although there are the bases to hypothesize that it would be a successful novel strategy. Many solid neoplasms have indeed been observed to express PD-1 and its ligand PD-L1 at high levels. The clinical potential of anti PD/PD-L agents arises on the fact that a great number of solid tumors express PD-L1, including breast, colorectal, NSCL, renal-cell and pancreatic carcinomas. Conclusion: Based on recent data, it can be postulated that cell therapy and HSCT may be combined in order to enhance the response rate of several solid tumors, and we do believe that this approach will provide excellent results in the coming years. Balsalobre 1,2 , M. Kwon 1,2 , L. López-Corral 3 , I. Heras 4 IiSGM, Instituto Investigación Sanitaria Gregorio Marañón Hospital Universitario Ramón y Cajal, Madrid, 9 Hematology, HGU Morales Meseguer, Murcia, 10 Hospital Clínico Universitario, Valencia, 11 Hematology Jonigk: None declared, I. Hamwi: None declared, L. Hambach: None declared, P. Schweier: None declared, I. Türüchanow: None declared, C. Rotmann: None declared, D. Ihlenburg-Schwarz: None declared, A. Papkalla: None declared, J. Raad Employee of: mosaiques, A. Durban: None declared, M. Dakna: None declared Tolerance and chimerism after renal and hematopoietic-cell transplantation. The New England journal of medicine HLA-mismatched renal transplantation without maintenance immunosuppression. The New England journal of medicine Correction of the hyper-IgM syndrome after liver and bone marrow transplantation. The New England journal of medicine Disclosure of Interest: None declared. Results: The study results show that OS and DFS was 82.4% and 72.4% respectively, in all patients included in this study, while subgroup analysis of cases with HCV PCR positivity had OS and DFS of 78.2% and 74.7% respectively, while those with negative HCV PCR had OS and DFS of 82.4% and 71.7% respectively (p value 0.509).Patients with serum ferritin below 2000 ng/dl were 89.5% and 80.9% for OS and DFS respectively, and for those with levels above 2000 were 75.1% and 64.6% for OS and DFS respectively (p value 0.025*). Regarding OS and DFS in different risk class groups, we reported 92.1% and 84.8% for OS and DFS, respectively in risk class 1 patient, and 82.7% and 70.4% for OS and DFS, respectively in risk class 2 patient, while for cases with risk class 3 the OS and DFS was 69.6% and 64.9% respectively ( p value 0.092). Conclusion: As regarding the impact of HCV status on the OS and DFS it shows no significant impact on both OS and DFS. There is a significant impact of pre-transplant ferritin level on the outcome of SCT of cases with BTM cases transplanted using PBSCs transplantation from a matched sibling donor regarding Thalassemia intermedia: An overview How I treat thalassemia Disclosure of Interest: None declared. P305 Impact of EVI1 Overexpressing Acute Myeloid Leukemia Patients undergoing Allogeneic Stem Cell Transplantation: An SFGM-TC study B. Papoular 1,* , M. Michallet 14 Hôpital d'instruction des Armées Percy Introduction: NRM is the first cause of treatment failure after UCBT following myeloablative conditioning (MAC) In comparison to MAC recipients, RIC recipients were almost 2 decades older (median age 52.5 vs 33.7 yrs, P40.001), were more often transplanted for AML (80% vs 57%, P = 0.001), received more frequently 2 cord blood units (61 vs 32%, P40.001), received more frequently units with4or = 2 HLA-mismatches (69% vs 58%, P40.001), received more TNC (median 3.5x10E7 vs 2.9x10E7, P40.001), and received less frequently ATG in the conditioning (23% versus 57%) Conclusion: These data suggest that LFS and OS might be as good with RIC than with MAC in adults AL pts offered UCBT. These observations could serve as basis for future prospective randomized studies TKIs incorporation into chemoterapeutic regimens has improved the prognosis, with a high percentage of molecular remission. So the use of TKIs prior to allo-SCT appears to be the most promising strategy. However, its real impact in the DFS for these patients is yet to be determined.We analysed the clinical outcomes in children with Ph+ ALL undergoing allo-SCT Material (or patients) and methods: We describe the outcome of 70 patients with Ph+ ALL who were submitted to an allo-SCT in 10 spanish Pediatric-Hematology-Oncology centers and were reported to GETMON between A total of 42 patients received TKI therapy pre-HSCT (TKI group): 30 patients received pre-HSCT only and 12 pre-and post-HSCT. While 28 patients were in the non-TKI group: 24 received no TKI (n = 24) and 4 patients only after relapse. Overall survival, event free survival (event: death or relapse) and MRT were calculated using SPSS package. Results: Overall survival at 5 years was 76.1% for the TKI group compared with 39 P320 Allogeneic Stem Cell Transplantation for t(3;3)(q21;q26)/ inv(3)(q21;q26) Acute Myeloid Leukemia -Report from the Acute Leukemia Working Party of the EBMT K Inv (3) was diagnosed in 66%, while t(3;3) in 34% of the patients, respectively. Concomitant cytogenetic aberrations most frequently included chromosome 7 abnormalities (monosomy, deletion or not otherwise specified) and complex karyotype and were observed in 74/106 (70%) and in 23/106 (23%) of pts, respectively. Fifty seven pts (53.8%) were transplanted while in active disease (refractory -40, first relapse -17) and 49 were transplanted in CR. 66 pts (62.3%) received myeloablative conditioning Presented study compared an outcome after both transplant approaches (reduced and myeloablative) in adults with high-risk ALL. Material (or patients) and methods: 41 patients with ALL and the median of age 29 (range, 19-58) years underwent allogeneic SCT (all but one allografted with peripheral blood stem cells). 21 recipients after RIC combining fludarabine (150 mg/m 2 ), melphalan (140 mg/m 2 ) and thymoglobulin (4.0 or 4.5 mg/kg) followed by earlier withdrawal of immunosuppression (if there was no evidence of GVHD) were retrospectively compared with 20 ones after various SMC (CY-TBI 3x, BU-CY 15x, BU-MEL 2x) 03), chronic GVHD (14% P = 0.009) and NRM 002) were observed after RIC. Despite statistically comparable cumulative incidence of relapse 3), low NRM after RIC resulted in higher probability of 2-year EFS (74% 01) and OS (94% Conclusion: MRD evaluation of bone marrow samples in AML patients at day 100+ after ASCT using multiparameter flow cytometry provides additional clinically relevant information and can be recommended in the follow up panel Disclosure of Interest: None declared Atg Plus Cyclosporine in Aplastic Anemia Patients Unresponsive to A First Immunosuppresive Therapy in AA transplants. Regular monitoring post transplant for chimerism status may help to identify patients at increased risk of graft rejection. Material (or patients) and methods:: IR was assessed in 11 children with severe AA who received allogeneic HSCT from fully matched sibling donors between Cell lineage specific chimerism ensured decreasing chimerism level. There was a wide range of CD3, CD4 and CD8 T cell numbers/μl blood. Mean Levels of CD3+ T cells were (795 ± 470), (1157 ± 495), and (1696 ± 531) at 6, 12, and 18 months. NK (CD56+/16+) cells recovered in all patients by 6 months. The mean NK at 6, 12, and 18 months were (100 ± 34), (83 ± 38), and (99 ± 49) respectively. Mean CD8 was highest at 6 months (527 ± 274). CD8 recovered by 6 months in all patients Normal IgG and IgM developed in 60% of children by 6 month post transplant while IgA did not normalize in half of the patients up to 18 months. Conclusion: Immune reconstitution in NK, CB19 B cell, CD3 T cell, and CD8 T cell, as well as IgM and IgG immunoglobulins production were early and complete by one year.Young age, use of ATG and absence of cGVHD might be responsible besides the presence of complete chimerism. Disclosure of Interest: None declared. P332 Incidence and epidemiology of aplastic anemia Introduction: Philadephia-negative Bcr-Abl negative chronic myelogenous leukemia (atypical CML, aCML) represents an aggressive disease entity with dismal prognosis, for which allo-SCT represent the only curative treatment option. Here we describe allo-SCT outcomes in a relatively large series of patients with aCML reported to the EBMT registry. Material (or patients) and methods: 42 patients with aCML who underwent allo-SCT from 1997 to 2006 were included in this retrospective analysis. Estimates of outcomes and hazard ratios where obtained by KM method and univariate Cox models. Transplant-related mortality (TRM) and relapse incidence were based on cumulative incidence estimates. Results: At the time of transplantation median age was 46 years (range 25-67 years), with 38% being older than 50 years. Males were 55%. Karyotype abnormalities other than the Phmatched unrelated in 36% (n = 15). A reduced intensity conditioning (RIC) regimen was employed in 24% (n = 10) of patients. Bone marrow represented the stem cell source in 33% of patients, being more often selected within the MAC then in the RIC transplant settings (41% and 10%, respectively). A T-cell depletion strategy was applied in 52% of cases, more frequently in patients transplanted from MUD than from HLA-identical sibling Overall survival was affected by patient age and the EBMT score, with HR of 3.62 when comparing patients classified in the "high-risk" group with those in the "low-risk". Relapses were less frequent in MUD then in HLA-identical sibling SCT (HR 0.22, C.I. 0.06-0.79, P = 0.021) with a consequent higher RFS probability (HR 0.41, C.I. 0.17-0.99, P40.05), while no difference in TRM was observed between the two subgroups. Conclusion: Considering the current lack of effective treatment options for aCML, this study confirmed that allogeneic SCT represents a valid strategy to achieve the cure in a reasonable proportion of patients, with young ones having a low EBMT risk score possibly being the best candidate. Also, this study raises speculations on a possible more effective "graft-versus-aCML" in the MUD setting Disease status at HCT was CR in 14%, PR in 40% and SD/PD in 46%. Karnofsky score was known in 69 pts, of these 58% had a score of 90 or more at the time of HCT, and only 4% had a score lower than 70%. Sixty percent of pts received reduced-intensity conditioning For the whole cohort of pts OS at 2 and 5 yrs was, respectively, 46% and 35%. PFS at 2 and 5 yrs was, respectively, 39% and 29%. Transplantation method (TCD vs. other) did not have a significant impact on OS or PFS. CI of NRM at 2 and 5 yrs were 41% and 46%, respectively. CI of relapse at 2 and 5 yrs were 20% and 25%, respectively. The use of PTCY did not lead to better outcomes, but the number of pts was very limited. The occurrence rate of acute graft-versus-host disease (aGVHD) at 100 days was 30% for grade II-IV and 15% for grade III-IV with a median time of onset of 23 days (range 4-57). Conclusion: In conclusion, mismatched related donor HCT results in long term PFS in a third of the pts. This result seems only slightly inferior to matched donor transplant (5 yrs PFS 29% vs. 37%). NRM in this group is high, but comparable to other studies with haploSCT without use of PTCY Hospital Puerta de Hierro, 6 Hematology Young and fit patients with high-risk chronic lymphocytic leukaemia (CLL) have a significantly shortened life expectancy. In them, allogeneic haematopoietic cell transplantation (alloHCT) remains a valid option with curative potential, although the results are not entirely satisfactory. The addition of rituximab, an anti-CD20 monoclonal antibody (MoAb) has improved the outcome of patients with B-cell malignancies undergoing alloHCT. Ofatumumab is a novel anti-CD20 MoAb with enhanced anti-CLL activity in vitro compared to rituximab. We hypothesized that adding ofatumumab to the conditioning regimen of patients with CLL undergoing alloHCT could improve the outcome of these patients even further. Material (or patients) and methods: Patients with high-risk CLL, defined by 17p-deleted progressive disease (17pD) or disease refractory or relapsing within 12 months of the last chemoimmunotherapy regimen (RR12), were eligible. The conditioning regimen included ofatumumab There have been no relapses on study. Both overall survival and PFS were 52% (95% CI 28-76%) at 12, but also at 24 months. MRD at 3 months from transplantation was negative in all patients evaluated. Conclusion: The addition of ofatumumab to the conditioning regimen of patients undergoing alloHCT for high risk CLL was feasible. After a median follow-up of 2 years, ofatumumab reduced the relapse rate completely, but the TRM was too high to recommend this therapeutic approach. Moreover, novel agents (ibrutinib, idelalisib, venetoclax) have a remarkable efficacy in patients with high-risk CLL with much less toxicity, even though none of them possesses curative potential. Alternative strategies with a lower toxicity should be explored for this high-risk population Russian Federation Introduction: Farber's disease (FD) is an extremely rare lysosomal storage disorder with progressive, multisystem involvement caused by deficiency of the lysosomal enzyme acid ceramidase, leading to accumulation of lipids in the joints, tissues and CNS. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) has been described as a curative option for patients with FD with Blood and Marrow Transplantation All had morphologic evidence of porphyria including cutaneous photosensitivity/scarring, liver dysfunction, or hematological manifestation. Donors were HLA matched unrelated (n = 4), haploidentical (n = 1), or 6/6 matched cord (n = 1). Bone marrow graft was used in 3, peripheral blood in 2 (one was supported by third party mesenchymal stem cells), and cord blood in 1. GVHD prophylaxis included in-vivo T cell depletion (anti-thymocyte globulin Except for one transplant in 2001, all others were performed between 2007-2013. Results: Median time to neutrophil and platelet engraftment was 23 days (range:6-32), and 27 days (range:14-89) respectively. Only one patient developed acute GVHD (grade II), and no one developed chronic GVHD. Only one patient (1/6) with CEP rejected the graft within one year of transplant and is alive with disease at 3 years post-transplant. One patient with a T-cell depleted haploidentical transplant (preceded by liver transplant 4 months prior to HCT) suffered non-relapse mortality due to VZV and adenovirus infection despite attaining 100% donor chimerism. Transplant-related toxicity also included rotavirus shock (n = 1). Neither BK virus cystitis nor CMV reactivation was observed (CMV paired status was negative/negative in 4/6 patients). Five patients were alive with median follow-up duration (of surviving patients) of 4.7 years. The 3-year overall and disease-free survival 83% and 67% respectively. Conclusion: Alternative donor allo HCT can be curative for erythropoietic porphyria. Only mortality encountered in our small series was due to overwhelming infection after T-cell depleted haploidentical HCT following liver transplant Successful match-unrelated donor bone marrow transplantation for congenital erythropoietic porphyria (Gunther disease) Employee of: Consultant for Recordati Rare Diseases, Inc, A. R. Gennery: None declared. P344 Hematopoietic Stem Cell Transplant in Class III Thalassemia Major: Does Splenectomy Improve Outcome? G Hematopoietic stem cell transplant (HSCT) to cure thalassemia major (TM) is an established modality. The success in class I or II has been very good but the success in class III patients is still a matter of concern considering the performance status of these patients. The role of splenectomy in affecting the outcome has yet not been discussed. We present our experience of HSCT in class III TM splenectomized vs non splenomegaly cohort. Material (or patients) and methods: Forty seven of class III TM underwent BMT between 14; BM/SC 1.11) and median MNC count was 9 Results: Median polymorphonuclear cell and platelet engraftment were seen on D+15 and D+26 respectively (BM/S 17days and 29days BM/SC 12days and 25days PB/S 15days and 25days Conclusion: From our experience class III TM splenectomized patients fared better as compared to patients with splenomegaly who underwent HSCT. No further deductions can be made due to smaller cohort Sickle cell anemia (SCA) 1,* , F. Martini 1 , R. morganti 1 , F. caracciolo 1 , N. Cecconi 1 , G. cervetti 1 , D. Caramella 1 , P. Erba 1 Out of 3 cases of negative US 2 positive cases of CT scan regarded M involvement and 1 A Ly. Nevertheless, overall, there was not a statistically (ST) significant difference (SD) in detection of Lph relapse between US and CT or PET/CT (P = 0.186). Pts have been stratified according to BMI (BMI4 = 25 and o25): BMI does not influence concordance in relapse detection between US and CT or PET/CT (P = 0.8). There was not a ST SD in relapse detection according to histology. In the CR control group: N = 44/50 there was concordance US vs CT or PET/CT. In 3 pts PET/CT was a false positive and biopsy confirmed the US negativity. In 1pts/50 US was positive vs neg CT and the pts relapsed 3 months later. N = 2 pts US revealed increased and dishomogeneous spleen vs normal CT and both relapsed 2.8-3.2 months later Brentuximab vedotin in relapsed/refractory Hodgkin's lymphoma: the Italian experience and results of its use in daily clinical practice outside clinical trials Introduction: Allo-HSCT represents a curative strategy for patients (pts) with relapsed/refractory lymphoma. The use of reduced intensity conditioning decreased toxicity of this approach Here, we report on our experience with different salvage regimens which have been used in lymphoma pts relapsing after allo-HSCT subsequently relapsed and received salvage treatment were retrospectively analyzed. The primary endpoints were overall response rate, response duration, rate and severity level of adverse events and overall survival from relapse. Results: A total of 174 transplanted pts were identified in our database Burkitt lymphoma, n = 1; Sezary syndrome, n = 1; MALT, n = 1; Waldenström's macroglobulinaemia, n = 1) with median age of 51 years (range, 19-73; male, n = 21) who received salvage approaches were evaluated. A total of 20 pts (74%) developed early relapses ( o 12 mo) within a median of 3 mo (range, 1-11) post-transplant. Chemorefractory disease had been documented in 15 pts (56%) prior to allo-HSCT. Auto-HSCT had been performed in 15 pts (56%). The salvage options included: (i) immunomodulative/monoclonal antibody therapy Brentuximab alone, n = 2; Ofatumumab alone, n = 1) n = 5; (iii) 2 nd allo-HSCT, n = 3; (iv) radiotherapy followed by DLI (n = 1) or lenalidomide (n = 1). Overall response was observed in 11 pts Both pts after radiotherapy developed CR and maintain their responses at the last follow up. The remaining 3 pts who received 2 nd allograft achieved PR but died due to early toxicity. In the DLI group, 2/4 pts developed GvHD (cGvHD, grade II (NIH)); in the immunomodulation/monoclonal antibody group: after lenalidomide: 4/9 pts developed cytopenia (44%; ≥ grade 3, n = 2) and 3/9 pts developed exacerbation of GvHD (skin grade II, n = 2; skin grade III, n = 1) at a median of 1 mo (range, 1-3) from begin of therapy. One of 4 pts developed late-onset cytopenia after rituximab There were no GvHD related deaths. Conclusion: Salvage therapy based on the use of DLI, immunomodulation/monoclonal antibodies or radiotherapy post-transplant seems to be safe and might induce durable remissions in pts with high-risk lymphoma who relapse after allo-HSCT Spleen involvement in Hodgkin's lymphoma: assessment and risk profile Disclosure of Interest: None declared. P352 Response to Salvage Therapy, The Main Prognosis Factor in Patients with Refractory or Relapsed Diffuse Large B-Cell Lymphoma Treated with R-ESHAP and Autologous Stem Cell Transplantation Out of 6,476 retrieved articles, 5 trials with a total of 157 patients were met the inclusion criteria. Among these patients, upfront ASCT was significantly associated with improvement of PFS However, no statistically significant difference was observed in OS P = 0.10) and CR rates Conclusion: Among patients with T-cell lymphoma, upfront ASCT showed better PFS although we should consider that the above RCTs included B-cell NHL as well as T-cell NHL. References: 1 Comparison of autologous bone marrow transplantation with sequential chemotherapy for intermediate-grade and high-grade non-Hodgkin's lymphoma in first complete remission: A study of 464 patients Allogeneic Transplantation (allo-HCT) Failure in the Biologically Targeted Drugs Era: Prolonged Survival Outcomes and Salvage Therapy Analysis A. Mussetti 1,* , V. Montefusco 1 , F We report results of reduced-intensityconditioning (RIC) allo-HCT in a cohort of patients with MM. Further analysis was performed for post allo-HCT relapsed/refractory A HLAmatched related or unrelated donor was available for 41 patients (58%). RIC and PBSC as graft source were used for the whole study cohort. Overall survival (OS) and Progression Free Survival (PFS) were performed with Kaplan-Meier analysis. Neutrophil and platelets engraftment, acute Graft Versus Host Disease (aGVHD), chronic GVHD (cGVHD), Non-Relapse-Mortality (NRM) and Relapse Incidence/Progression of Disease (RI/POD) were obtained with competing risk analysis. Associations among patient-, disease-, and transplantation-related variables and outcomes of interest were evaluated using Cox proportional hazards regression. Results: Day +30 neutrophil and platelets engraftment was 98%(95%CI:82-100) and 96%(95%CI:84-99)respectively. The cumulative incidence of grade II-IV aGVHD at day +100 and cGVHD at 5-year was 13% Study on cord blood transplantation in multiple myeloma on behalf of Eurocord, CBC-Cellular Therapy & Immunobiology Working Party Department of Hematology, Erasmus MC-Daniel den Hoed Cancer Centre Disclosure of Interest: None declared. References: 1. Hematopoietic Stem Cell Transplantation for Multiple Myeloma: Guidelines from the American Society for Blood and Marrow Transplantation with T cell depletion. Material (or patients) and methods: Retrospective analysis performed on data collected from patients receiving HSCT for MDS/MPD unclassified at King's College Hospital between years 2000 -2015. Results: Thirty three patients with MDS/MPD unclassified overlap syndromes who have undergone allogeneic-HSCT were included in the analysis. The median age at HSCT was 55 (27-69), with up to 15 years follow-up. 19 patients were male and 14 female. Sixteen of the patients (48.5%) received reduced-intensity conditioning. 32/33 patients (97%) received T-cell depletion with ATG or Campath, 14 (42.4%) and 18 (54.5%) patients respectively. Transplants were from unrelated donors in 25 patients (75.8%) and from a sibling donor HLA-matched transplant, and 4 (12.1%) received a 9 Recipient features at diagnosis including age, haematological parameters, cytogenetic risk group and BM blast percentage were not found to correlate significantly with overall survival, relapse incidence or PFS. Similarly, factors relating to the HSCT including conditioning regime and intensity, as well as T-cell depletion, did not have a significant impact on outcomes. Of interest on univariate analysis, overall survival was impacted by CMV matching between donor and recipient ANCo 800/μL in 12 patients, hemoglobin (Hb) levelo8 gm/ dL in 14 and platelet count o50,000/μL in 15. A total of 31 patients were initially treated with HMT. Median time from diagnosis to transplant was 216 days (range, 93-890) and median cycles of HMT were 4 cycles (range, 1-20). The 2 patients achieved partial response (PR), 4 marrow complete response (CR), 17 stable disease, 5 progressive disease, and 3 relapse after CR/PR at the time of Allo-HCT. Sources of stem cells were peripheral blood in 29 patients, cord blood in 1 and BM in 1. The 14 patients had matched sibling donor. The 15 patients received a myeloablative regimen. Their median CD 34-positive cell count was 6.73 × 10 6 /kg (range, 1.3-11.56). The 30 patient showed neutrophil engraftment (ANC4500/μL), at a median time of 12 days (range 9-18). The incidence of acute graft-versus-host disease (GVHD) was 9.6% (3.2% for grade 3) and the incidence of cGVHD was 12.9% (9.6% for severe). The 13 patients died during follow-up, 5 patients due to AML transformation, 3 due to infection, 2 due to VOD, and 1 due to GVHD. The treatment related mortality was 19.3%. The 2 years OS were 87.4% (±8.4%) in early allo-HCT group and 51.1% (±15.8%) in delayed allo-HCT group (P = 0.009). The 2 years leukemic free survival (LFS) was 93.8 (±6.1%) in early allo-HCT group and 54% (±14.2%) in delayed allo-HCT group (P = 0.035). There was no statistical significance in the prognostic variables of IPSS-R and response to HMT United States Introduction: Relapse is the major cause of failure after allogeneic hematopoietic cell transplantation (HCT) in myelodysplastic syndromes (MDS) Material (or patients) and methods: We analyzed results in 258 MDS patients transplanted between 8%) had early relapse (more than 5% blasts on marrow aspirate). The 2-year overall survival (OS) from day 28 post-HCT of MRD positive patients and early relapsing patients was 21% and 5%, respectively, P = 0.001, Log-rank test. Despite a statistically significant difference, the outcome of both the subgroups was very poor. Conversely, the 206 MRD negative patients 2-year OS was 70%. We compared the probability of being MRD positive or relapsing early between patients who received low-versus high-intensity conditioning using stepwise logistic regression. Candidate variables included gender, age at transplant, pre-HCT treatments (induction chemotherapy given as first line treatment or as salvage, hypomethylating agents only, no therapy, or other), secondary MDS, revised international prognostic scoring system (IPSS-R) score at diagnosis, pre-HCT evolution to acute myeloid leukemia (AML). The odds of day 28 early relapse/MRD positivity was higher in patients treated with low Among patients with less than 5% marrow blasts, regardless of treatment, we defined two groups, cytogenetics positive and cytogenetics negative. The criterion for cytogenetics positivity was the presence of an identifiable abnormality by cytogenetics or FISH (excluding isolated deletion Y) Those patients should be considered for experimental treatments and clinical trials. Patients with 5% or more marrow blasts or cytogenetics positive at pre-HCT evaluation have a significantly higher risk of early relapse/MRD positivity at day 28 when treated with low-intensity conditioning Introduction: The number of MDS patients who receive allogeneic stem cell transplantation is steadily increasing. However, the main cause for treatment failure is relapse which exceeds 50%. Posttransplant strategies such as novel agents (5-azacytidine, HDAC inhibitor etc.) as well as adoptive immunotherapy (e.g. DLI) are currently under investigation to reduce the risk of relapse. Material (or patients) and methods: In order to have a valid tool for stratification in phase III studies, the CMWP of EBMT is developing a simplified "Relapse-risk score" for MDS patients. For this purpose 1638 patients with MDS who received an allogeneic stem cell transplantation from HLA-identical sibling or a matched unrelated donor between 1995 and 2012 and reported to EBMT registry were included. The median age of the patients was 54 years (range 18-76) and diagnosis were: RAR/RARS/RCDM-(RS) and RAEB. Variables which were taken into the analysis were: age, classification of MDS, donor source (HLA-identical sibling vs matched unrelated donors), stem cell source (PBSC vs bone marrow), T-cell depletion, intensity of the conditioning regimen (reduced intensity vs standard myeloablative), blasts in bone marrow at time of transplant, and cytogenetic: very poor (very poor according to IPSS revised or monosomal karyotype), poor (according to IPSSrevised), and good (according to IPSS-revised) and unclassifiable. To take the different risks of relapse depending on time from transplant into account we developed 4 different prognostic models: 1) relapse between SCT and 6 months after SCT, 2) relapse between 6 and 12 months post-SCT, 3) relapse between 12 and 24 months post-SCT and 4) relapse after 24 months post-SCT. Results: Multivariate Fine and Gray regression models were used to assess the impact of risk factors on the cumulative incidence of relapse the first conditioning regimen was carboplatin and etoposide, the second was thiotepa and cyclophosphamide, both with intraventricular/intrathecal metotrexat. Results: The median follow-up is 48 months (range, 5-173). The median time to engraftment was day +17 (range, 8-86) after auto-HSCT. Four of 5 patients with SD at the moment of auto-HSCT had disease progression within 8 months after HDCT. Twenty-two of 49 patients with CR or PR relapsed 1-24 months after HDCT, the other 27 patients are currently in CR or PR on the maintenance therapy. Cumulative incidence of relapse in 4 years accounted 45% (95% CI 21%460%). The conditioning regimens had acceptable toxicity. Complications grade 4 (COMMON TOXICITY CRITERIA 2014) were observed in 14% of cases. Four-year overall survival (OS) in all patient's group was 67% and disease free survival (DFS) was 55%. MB and germ cell tumors had better survival rate (DFS 65% and 64%, respectively) in compared to other embrional tumors (DFS 40%, P = 0,05). DFS was significantly better among patients 44year in compared to children o4year: 68% and 15%, accordingly ( p40,001) Highdose chemotherapy for germ cell tumors: do we have a model? An expert opinion on behalf of the European Society for Blood and Marrow Transplantation, Solid Tumors Working Party (EBMT-STWP). Expert Opinion in Biological Therapy Allogeneic haematopoietic stem cell transplantation for metastatic renal carcinoma in Europe Haploidentical stem cell transplantation (SCT) with t cell depleted grafts in advanced pediatric sarcomas H.-M. Teltschik 1,* , B. Gruhn 2 , T. Feuchtinger 3 , C. Urban 4 1 Pediatric Hematology and Oncology, Children`s University Hospital, Tübingen, 2 Pediatric Hematology and Oncology, University Children`s Hospital, Jena, 3 Pediatric Hematology and Oncology, Dr. v. Haunersches Kinderspital, Munich, Germany, 4 Pediatric Hematology and Oncology, University Children`s Hospital, Graz, Austria Introduction: The rationale for transplanting pediatric sarcoma patients with haploidentical grafts is a hypothesised graft-vs-tumour effect. We investigated a cohort of 21 pediatric patients with advanced sarcomas transplanted with T-and B-cell depleted peripheral stem cell grafts from haploidentical donors between 2005 and 2013. Material (or patients) and methods: 11 patients had advanced Ewings sarcoma, only 2 of them achieved complete remission (CR) before SCT. 10 patients had advanced Rhabdomyosarcomas or Rhabdo-like tumours (RMS), 6 of them achieved CR prior to SCT. Median follow up was 8.4 month. Median age at SCT was 14.3 years. Standard conditioning regimen consisted of Melphalan (2x70mg/m 2 ), Fludarabin (4x40mg/m 2 ), Thiotepa (1x10mg/kg), and OKT3 or ATG. Graft manipulation was carried out by direct depletion using anti CD3/CD19 or TCRαβ/CD19 coated magnetic microbeads. A median number of 10.4x10 6 CD34+ Progenitor cells and 38.8x10 3 Introduction: Allogeneic stem cell transplant (allo-SCT) provides a potentially curative therapy for patients with high-risk or chemorefractory acute myeloid leukemia (AML). In the last years several transplant strategies, using alternative donors or different conditioning regimens, have been introduced into clinical practice to improve outcome in haematological malignancies, especially in AML. The aim of the present study was to evaluate the impact of HLA identical siblings (SIB) unrelated donors (UD) and HLA haploidentical family donors (HAPLO) on survival, transplant related mortality (TRM) and quality of response in patients affected by acute myeloid leukemia (AML). Material (or patients) and methods: The study population was represented of 126 AML patients receiving allogeneic transplant (SCT) in a single Center. Of them, HAPLO were 22, SIB were 74 and UD were 30. Patients were stratified on the basis of disease phase: first remission (CR1, n = 66), second remission (CR2, n = 23) or refractory disease (Ref, n = 37) . The conditioning regimen was defined as myeloablative (MAC) (n = 113) or reduced intensity (RIC) (n = 13). In order to prevent Graft vs host disease (GvHD), the following prophylaxis were administered: unmanipulated bone marrow, post transplant cyclophosphamide (PT-CY) with cyclosporine (CsA) and mycophenolate (MMF) in HAPLO, CsA and methotrexate (MTX) in SIB and CsA +MTX+antithymocyte globulin (ATG) in UD. Fiftheen (68%) HAPLO recipients were refractory and six (27%) of them received a previous transplant (autologous n = 5; allogeneic SIB n = 1). The 3 -years OS on the entire study-cohort (n = 126) was 51,5%, although significant differences were noticed among the three donor types of transplant: 3 -years OS was 60%, 51% and 30,5% (P = 0,0393) for SIB, UD and HAPLO graft, respectively. When the response quality was considered, CR1 showed a longer OS compared to CR2 and refractory patients (73% vs 51% vs 19%, respectively ( p40.0001)).The cumulative incidence (CI) of TRM was 19% in all patients but TRM was lower in SIBs (12,1%) respect to HAPLO (27, 2%) and UD (30%). Similar results came from the comparison between disease phase and outcome: 13,6%, 26% and 24,3% for CR1, CR2 and refractory disease, respectively. Conclusion: In conclusion, the early phase of response was a positive predictor of a better OS compared to refractory disease. SIB showed the best outcome while no significant differences were found between HAPLO and UD transplantation. Disclosure of Interest: None declared. 1 Hematology and Bone Marrow Transplant Unit, Azienda Ospedaliera Papa Giovanni XXIII, Bergamo, 2 Haematology, AOU Città della Salute e della Scienza di Torino, Torino, 3 Haematology, Spedali Civili di Brescia, Brescia, 4 Hematology and BMT Unit, Ospedale San Bortolo, Vicenza, 5 U.O. Ematologia e Centro Trapianti, Ospedale Oncologico di Riferimento Regionale 'Armando Businco', Cagliari, 6 Hematology and BMT Unit, Central Hospital of Bolzano, Bolzano, 7 Introduction: Allogeneic stem cell transplantation (alloSCT) in first complete remission (CR1) remains the consolidation therapy of choice in Ph+ ALL. Evidence is emerging that S277 post-transplant relapse is influenced by the persistence of minimal residual disease (MRD), with an inferior outcome of patients undergoing transplantation with measurable level of MRD (1) (2) . Since a deeper molecular response can be achieved with innovative targeted therapies, such as second and thirdgeneration TKIs or immunotherapy, an accurate evaluation of MRD values before alloSCT is mandatory. The aim of this study was to evaluate the impact of MRD levels before transplant in Ph+ALL patients in CR1 on relapse incidence (RI), disease free survival (DFS) and overall survival (OS). Material (or patients) and methods: One hundred and six adult patients (median age 41.2, range 19-62) with newly diagnosed Ph+ ALL (as determined by cytogenetic or molecular analysis) were enrolled into 2 prospective NILG protocols (09/00 ClinicalTrial.gov Identifier: NCT00358072 and 10/07 ClinicalTrial.gov Identifier: NCT00358072) and were treated with chemotherapy and imatinib. One hundred (94%) achieved CR1, of whom 72 patients underwent an alloSCT in CR1 and are the subject of this report. MRD was determined by quantitative polymerase chain reaction (RQ-PCR) according to validated methods. Results: Among the 72 patients undergoing alloSCT, MRD status before transplant was available for 65 patients (90%). Twenty-four patients (37%) achieved a complete molecular response (BCR-ABL/ABL o 1x10 -5 ) at time of conditioning (MRD-group), while 41 (63%) remained carriers of any positive MRD level in the bone marrow or peripheral blood (MRD+ group), ranging from 1.2x10 -4 to 2x10 -1 . Patients' characteristics were similar between MRD+ and MRD-groups, except for a higher hemoglobin levels and a predominance of male gender in MRD-group. Thirty-five patients received alloSCT from a sibling and 37 from unrelated donor. The conditioning regimen to alloSCT was myeloablative in 85% and reduced intensity in 15% of patients. The stem cell source was the bone marrow in 19%, the peripheral blood in 78% and cord blood in the remaining 3% of patients. The OS of patients receiving alloSCT was 50%. The MRD negativity at time of conditioning was associated with a significant benefit in terms of risk of relapse with a RI of 8% compared to 39% of patients with MRD positivity (P = 0.007) ( Figure A) . The LFS and OS probability were not significant different in MRD-compared to MRD+ patients (58% vs 41%, P = 0.17 and 58% vs 49%, P = 0.55, respectively) ( Figure B) , likely due to the effective post-relapse treatment with TKIs and/or DLI. The cumulative incidence of non relapse mortality was similar in MRD-compared to that of MRD+ group (33% vs 20%, P = 0.22). Conclusion: Our results confirm the importance of achieving a complete molecular remission before transplant that should be considered an essential prerequisite for successful alloSCT.Introduction: The cohesin complex have been described as novel mutations occurring in 13% of acute myeloid leukemia (AML) patients, suggesting that the cohesin-complex presents an important pathway in the pathogenesis of AML. Genes that belong to the cohesion complex in somatic vertebrate cells are SMC1A, SMC3, RAD21, STAG2, and STAG1 and, these genes regulates chromosome segregation during meiosis and mitosis. However, the prognostic significance of cohesion complex mutation in normal karyotype (NK) AML is controversial in clinical setting, especially in the transplant setting. Material (or patients) and methods: A total of 413 patients were included in the present study, and all met the following eligibility criteria: 1) age ≥ 15 years; 2) a diagnosis of NK-AML confirmed by conventional cytogenetic analysis; 3) treatment with induction chemotherapy using a standard protocol (a 3-day course of anthracycline with a 7-day course of cytosine arabinoside) and, 4). NK-AML patients were diagnosed from October 1998 to May 2015 in seven participating institutes. Analysis of each mutations were performed using targeted resequencing by SureSelect platform technology. Results: The cohesion complex mutations were detected 8.5% in patients; 3.1% (13/413) in STAG2 mut , 3.6% (15/413) in SMC1A mut , 1.2% (5/413) in SMC3 mut and RAD12 mut , respectively. The 413 patients received the induction chemotherapy and 341 (82.6%) patients achieved the complete remission (CR). There was no difference in CR rate according to the status of cohesion complex mutation (wild type vs. mutated, 83.3% vs. 74.3%, P = 0.177). In the survival analysis according to the cohesion complex mutation, there was no difference in the 5year overall survival (OS) (wild type vs. mutated, 38.9% vs.29.0%, P = 0.177) and event free survival (EFS) (wild type vs. mutated, 33.8% vs.29.0%, P = 0.424). After CR achievement, 143 patients received allogeneic stem cell transplantation (allo-SCT). There was no difference in the number of patients receiving allogeneic SCT according to the cohesion complex mutation after CR achievement (wild type vs. mutated, 41.0% vs.53.8%, P = 0.200). The five-year OS and EFS in patients received allo-SCT vs. chemotherapy only after CR achievement was 57.8% vs. 38.1% ( p40.001) and, 54.9% vs. 30.0% ( p40.001), respectively. In the allo-SCT group, the five-year OS and EFS of cohesin complex wild vs. mutated was 55.7% vs. 52.6% (P = 0.874) and 52.2% vs. 50.5% (P = 0.663), respectively. In the chemotherapy only group, the five-year OS and EFS of cohesin complex wild vs. mutated was 39.1% vs. 25.0% (P = 0.095) and 30.6% vs. 25.0% (P = 0.253), respectively. We subanalyzed the cohesion complex mutated group according to the consolidative type (allo-SCT vs. chemotherapy only). In this subgroup, he five-year OS and EFS in patients received allogeneic SCT vs. chemotherapy only after CR achievement was 52.6% vs. 25.0% (P = .016) and, 53.4% vs. 25.0% (P = 0.020), respectively. Conclusion: In NK AML population, the incidence of cohesion complex mutation was 8.5%. The cohesion complex mutation did not influence the treatment response such as CR rates and OS according to chemotherapy only or allo-SCT. Each consolidative treatment (allo-SCT vs chemotherapy only) did not show the statistical difference according to cohesion complex mutation. Therefore, we speculated that allo-SCT might not be influenced by cohesion complex mutation. Disclosure of Interest: None declared. Impact of bone marrow blast count prior to allogeneic HSCT in aplasia in patients with high risk AML J. M. Middeke 1,* , F. Hönl 1 , K. Sockel 1 , F. Stölzel 1 , M. Wermke 1 , M. von Bonin 1 , R. Teipel 1 , C. Link 1 , K. Brandt 1 , C. Röllig 1 , C. Klesse 2 , G. Ehninger 1 , U. Platzbecker 1 , M. Bornhäuser 1 , J. Schetelig 1,2 1 Medizinische Klinik I, UNIKLINIKUM DRESDEN, 2 DKMS Clinical Trials Unit, dresden, Germany Introduction: Purpose. Allogeneic Hematopoietic Stem Cell Transplantation (HSCT) is the standard of care as consolidation therapy in patients (pts.) with high risk acute myeloid leukemia (AML). Several diseaserelated risk factors are known to have an impact on outcome such as cytogenetic and molecular abnormalities, remission status and treatment status at the time of transplantation. AlloHSCT in aplasia after intensive induction chemotherapy has been established as a treatment option for pts. with refractory AML. Since the impact of residual bone marrow blasts prior to HSCT in this specific treatment sequence is not clear we performed a large retrospective cohort study. Material (or patients) and methods: Patients and Methods. All pts. receiving a first alloHSCT for high risk AML at the University Hospital Dresden after conditioning with either fludarabine (30 mg/m 2 , day -6 to day -2) and melphalan (150 mg/m 2 on day -2) or clofarabine (30 mg/m 2 , day -6 to day -3) and melphalan (140 mg/m 2 on day -2) between Jan, 1, 2003 and May, 1, 2015 were included. An bone marrow blast count within 28 days prior to alloHSCT had to be available. High risk AML was defined by either high risk cytogenetic abnormalities according to the ELN classification, refractory AML, or relapsed disease. Response to last chemotherapy was classified according to the ELN criteria. Multivariate Cox regression models were fitted with information on age, sex match, CMV match, genetic risk classification, type of AML, donor type, conditioning regime and response to last chemotherapy. Results: Data from 174 pts. were evaluable. The median age was 56 years. Sixty-one percent of the pts. suffered from de novo AML, while 30% had secondary AML and 9% tMN. The conditioning regimen consisted of Fludarabine/Melphalan in 141 (81%) pts. and Clofarabine/Melphalan in 33 (19%) pts. Donors were matched siblings in 24%, fully matched unrelated donors in 48% and partial matched unrelated donors in 19% of the pts.. Eighteen pts. (10%) had a haploidentical donor. The median bone marrow blast count was 11% (range, 0-96%). At the time of transplant, 62 (35%) pts. were in CR (n = 6), CRi (n = 6) or in a morphologic leukemia-free state, 52 (30%) had a moderate response and 60 (34%) had no response or untreated relapse. The median follow-up was 4 years (range, 0.1 to 11 years). At four years, the probabilities of OS and EFS were 43% (95% CI, 35% to 51%) and 34% (95% CI, 27% to 42%), respectively. The CIR was 32% (95% CI, 25% to 40%) and the NRM 34% (95% CI, 27% to 41%). In multivariate analysis only age (HR = 1.02; p = .02) and response to last chemotherapy (HR = 1.7 for moderate response and 1.8 for no response; p = .03) had a significant influence on EFS. Notably, if elevated to 5% or more, the percentage of residual bone marrow blasts was not predictive for EFS. Conclusion: The bone marrow blast count prior to the start of the conditioning regimen is an important predictor for transplant outcome after melphalan-based reduced intensity conditioning. This result raises the question whether pts. with residual marrow blasts would benefit from an additional attempt to induce a remission with chemotherapy or fare best with alloHSCT as soon as possible. Maintenance treatment, starting early after transplantation, should be evaluated for this group of pts. Disclosure of Interest: None declared. Introduction: Allogeneic hematopoietic stem cell transplantation (SCT) is an effective treatment for various hematologic malignancies. Rarely de novo hematologic neoplasia may develop from donor cells after SCT. The leukemic transformation of otherwise healthy donor SCs provides a useful in vivo model to study the mechanisms involved in leukemogenesis. This paper describes the specific gene mutations that occur in donor cells after unrelated cord blood transplantation (UCBT) for acute lymphoblastic leukemia contributing to the development of normal-karyotype acute myeloid donor cell leukemia (DCL).Material (or patients) and methods: To identify DCLassociated mutations, whole exome sequencing (WES) was performed by Illumina HiSeq on bone marrow samples at days +98, +189, +350, +468 (DCL onset), +569 (DCL relapse) post-UCBT and on the UCB unit used SCT. Following alignment to the human reference genome (NCBI build 37/hg19) filters were applied based on coverage 410x, quality 430X, minor allele frequency o0.05 and non-synonymous variants located in coding sequences associated with leukemia. Variants meeting such criteria were retained and matched to the UCB sequence to remove those that were already in the UCB. Non shared variants were evaluated with SIFT, Polyphen and Mutation Taster softwares to predict their functional effects. Results: No mutations in well established leukemia associated genes were found in the UCB used for transplantation. Nineteen novel somatic variations were identified in the sample from day +98, 28 in the sample from day +189, 29 in the sample from day +350, 44 in the sample from day +468 and 36 in the sample from day +486. Variants appearing transiently before the development of DCL but that were absent from the sample at the onset of DCL were excluded from the analysis. In silico analysis of variants allowed the several somatic mutations to be directly related to the development of DCL ( Figure 1 ). Furthermore, mutant alleleThe Outcomes of Allogeneic Hematopoietic Stem Cell Transplantation in AML patients Registered in the Japan Adult Leukemia Study Group (JALSG) Study K. Ishiyama 1,* , S. Ohtake 2 , K. Miyamura 3 , H. Kiyoi 4 , Y. Miyazaki 5 , S. Miyawaki 6 , R. Ohno 7 , Y. Kobayashi 8 , T. Naoe 9 1 Department of Hematology, 2 Clinical Laboratory Science, Kanazawa University, Kanazawa, 3 Hematology Division, Japanese Red Cross Nagoya First Hospital, 4 Department of Hematology and Oncology, Nagoya University, Nagoya, 5 Department of Hematology and Molecular Medicine Unit, Nagasaki University, Nagasaki, 6 Division of Hematology, Tokyo Metropolitan Ohtsuka Hospital, Tokyo, 7 Aichi Cancer Center, Nagoya, 8 Department of Hematology, National Cancer Center Hospital, Tokyo, 9 National Hospital Organization Nagoya Medical Center, Nagoya, Japan Introduction: The current guidelines for acute myeloid leukemia (AML) issued by the Japan Society for Hematopoietic Cell Transplantation were updated in 2009 to reflect that intermediate-risk or unfavorable-risk AML patients who are younger than 55 years of age should be treated with allogeneic stem cell transplantation (allo-SCT) at the first complete remission (1CR). However, according to the annual report of a nationwide survey which was performed in 2014, the outcome of AML patients who receive allo-SCT at the 1CR are comparable to those who it at the second CR (2CR). In addition, most of the 'evidence' for this recommendation was based on the results of either prospective studies with randomization according to donor availability or retrospective studies, which may have led to the influence of selection bias. On the contrary, physicians sometimes have difficulty in deciding whether a patient should receive upfront allo-SCT or a final treatment with chemotherapy. We conducted a retrospective study to verify the validity of allo-SCT in AML patients at the 1CR using a dataset of patients who were registered in prospective studies. Material (or patients) and methods: We performed additional research to obtain allo-SCT information in AML patients who were registered in the AML201 studies conducted by the Japan Adult Leukemia Study Group. Furthermore, some SCT data were provided by the Japanese data center for hematopoietic cell transplantation (JDCHCT). We added this information to the original data that the CRF collected 9 years previously. This study was conducted with the National Cancer Center Research and Development Fund (26-A-24). Results: A total of 1,064 patients with de novo AML were registered in the JALSG study and 494 patients underwent allo-SCT. Patients with the favorable-and poor-risk karyotypes accounted for 20% and 9% of the study population, respectively. The numbers of patients with each disease status at the time of allo-SCT were as follows: 1CR (who achieved CR after undergoing remission induction chemotherapy once or twice), n = 134; 1 st relapse (1rel), n = 83; 2CR, n = 137; and primary induction failure (PIF), n = 66. The estimated 5-year overall survival (OS) in the patients with disease statuses of 1CR, 1rel or later (n = 251), and PIF (n = 66) were 58%, 48% and 27%, respectively. The allo-SCT outcomes in the patients with disease statuses of 1CR and 2CR were comparable, with 5-year OS rates of 60% and 65%, respectively (P = 0.83). However, the 5-year OS after allo-SCT in the 1CR patients was significantly superior to that of the patients with a disease status of 1rel or 2CR ([BQ1] 60% vs. 48%, P40.03). Conclusion: The current study reveals that the results of allo-SCT in patients at 1CR and 2CR were similar. However, we are of the opinion that the choice of upfront allo-SCT should be recommended in AML patients at the 1CR, as the outcome of allo-SCT at the 1CR is superior to that in patients after a relapse. Disclosure of Interest: None declared. Introduction: Relapse of malignancy remains the single most important cause of treatment failure in recipients of TCR alpha/beta depleted grafts in a cohort of patients with highrisk acute leukemia. Negative depletion of α/β(+) T cells keeps NK cells and gamma/delta-T cells within the graft, however anti-thymocyte antibodies currently used as part of conditioning regimen may impair the potential benefits of these lymphocyte populations and delay immune reconstitution. Interleukin 6 is one of the major cytokine drivers of acute GVHD and its inhibition (Tocilizumab) may be a potential new and simple strategy to protect from acute GvHD. In this pilot trial we evaluate the safety and effects of substituting ATG (thymoglobulin) with tocilizumab within a backbone of intensive chemotherapy-based conditioning and TCR-alpha/beta depletion in haploidentical and matched unrelated transplantation (MUD) in acute leukemia. Material (or patients) and methods: A total of 12 pediatric patients with acute leukemia, 9 with AML and 3 with ALL, 6 female/6 male, median age 7,9 years (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) , underwent allogeneic HSCT between December 2014 and September 2015, median follow-up 5 months (2, (7) (8) (9) (10) (11) . Eleven patients were transplanted from haploidentical donors and 1 from MUD. Patients were divided in 2 groups according to remission status: complete remission (CR) -3 pts (all ALL), active disease (AD) -9 pts (all AML). All patients with AD were transplanted from a haploidentical donor. TCRαβ+/CD19 +-depletion of HSCT with CliniMACS technology was implemented in all cases. Nine of the patients (AML) received Treosulfan/Melphalan/Fludarabine and 3 (ALL) Fludarabine/ 12 Gy TBI/VP. GvHD prophylaxis included rituximab 200 mg/ m 2 on day 0, tocilizumab 8 mg/kg on day -1 and posttransplant bortezomib on days +2,+5. The median dose of infused CD34+ cells was 7 x 10 6 /kg (range 4,2-17), TCR a/b -13,8 x10 3 /kg (range 2-68). Results: Primary engraftment was achieved in 12 of 12 pts., the median time to neutrophil and platelet recovery was 12 days (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) . No case of graft rejection was registered. All alive patients were in full donor chimerism on day +30. Early mortality within 100 days was 8,3%: 1 pt. with AML had acute lung injury after engraftment on day+14. There were no allergic or infusion-related adverse events associated with tocilizumab. Febrile neutropenia was registered in 6 cases (50%), mucositis grade II -in 6 pts. (50%), severe infection, Introduction: Response to first induction chemotherapy is a strong prognostic factor in patients with newly diagnosed acute myeloid leukemia (AML) eligible for intensive treatment. Primary Induction Failure (PIF) leads to an escalation of therapeutic means, the benefit of which remains controversial. Allogeneic hematopoietic stem cell transplantation (HSCT) appears to be the most appropriate consolidation treatment to lower relapse risk but its timing is still a challenge: disease-free (DFS) and overall survivals (OS) are known to be better if allogeneic HSCT is done in complete remission (CR), but earlier transplant is usually associated with a better outcome. These two conditions cannot be met in PIF AML. Material (or patients) and methods: We therefore conducted a retrospective monocentric study analyzing the outcome of all adult patients with newly diagnosed AML in Strasbourg University Hospital between 2002 and 2014 and failing to achieve remission after first induction. Results: Of 704 AML patients, 394 (56.0%) received intensive chemotherapy, and 90 (22.8%) were considered PIF. Sixtythree of these patients (70.0%) received further intensive treatment, enabling a median OS of 418 days, compared to 65 days for patients on palliative care and 253 days for patients receiving hypomethylating agents only ( p40.001). Salvage chemotherapy led to CR in 40% of patients, and those having received an allogeneic HSCT while still on CR reached a median OS of 2068 days, with a three year survival of 57.4%. Patients in CR after salvage therapy but who were not subsequently transplanted had a median OS of only 551 days. Patients receiving more than one salvage chemotherapy before being transplanted had a median OS of 418 days, compared with 467 days for patients who were transplanted while refractory without or after only one salvage therapy (P = 0.7). Among the 23 patients still refractory at transplantation, allogeneic HSCT eventually led to remission in 20 patients (87%), but with a high subsequent relapse rate. Allogeneic SCT in refractory patients leads to a median OS of 467 days, with three-year survival reaching 25.6%. On the whole, the three-year survival rate of the 38 patients who received an allogeneic HSCT was 39.5% compared with 2% for the 52 patients who did not undergo transplantation. Conclusion: Continuing intensive treatment of AML after primary induction failure is still the better therapeutic option in terms of survival for fit patients: we have observed that it can result in prolonged survival, and perhaps cure, for roughly a quarter of them, supporting this approach. However, considering our data, receiving more than one salvage chemotherapy doesn't appear to lead to a better outcome and may impair the rate of allogeneic SCT. This might eventually affect the outcome of refractory patients, since allogeneic HSCT is a crucial step in their management. Although survival is consistently lower for patients lacking CR, allogeneic SCT for refractory patients still offers prolonged survival to a significant proportion of PIF AML patients in spite of their initial dismal prognosis. Introduction: Various innovative Disease Modifying Agents (nDMAs) have been approved in the last ten years for the treatment of Multiple Sclerosis (MS). Autologous haematopoietic stem cell transplantation (aHSCT) was proposed as clinical option for aggressive MS forms unresponsive to approved therapies; the safety profile of aHSCT following the administration of new DMAs has not been assessed yet. We analyzed a series of nDMAstreated patients who underwent aHSCT in our Center; the clinical outcome was compared with other patients who were transplanted after failure of conventional treatments (cDMAs) which were used as internal controls. Material (or patients) and methods: Patients with highly active MS forms transplanted after failure of at least one line of DMAs or DMAs' withdrawal were included in this study. Washout intervals between treatment and HCST varied according to the DMA. The frequency of adverse events after aHSCT was compared in the two groups. Peripheral Blood Stem Cells (PBSCs) were mobilized by Cyclophosphamide 4 g/ sqm+G-CSF; all patients were conditioned with BEAM plus rabbit ATG. Both haematological and neurological follow-up were assessed according to standard policies. Results: Sixteen MS patients (13 RR, 3 SP) previously treated with nDMAs and 28 controls (11 RR, 16 SP, 1 PP) previously treated with others DMAs were included. The median EDSS disability score at HSCT was 4.5 for the nDMAs group and 6.0 for the controls (P = 0.069 Mann-Whitney). The median disease duration before aHSCT was respectively 11.6 years and 11.3 years (P = 0.76 Mann-Whitney). Full data are reported in table 1. Hematopoietic engraftment and neurological outcome were similar in the two groups; after stratification on the MS form. nDMAs-treated patients showed a higher incidence of infective events requiring hospitalization within twelve months after transplant (5/16 vs 0/28, p value = 0.003); in particular, 7 bacterial pneumonias were observed in nDMAs group, whilst none in the control group. A significant association to Natalizumab (NTZ) treatment was shown: all patients who developed pneumonia had previously received a median of 23 (range 2-30) NTZ administrations, with a median wash out period of 15 months (range 5-45). NTZ was variously associated to others DMAs. No differences in the incidence of CMV and EBV viral reactivations were shown. No transplant related mortality was reported. Conclusion: In this cohort of MS patients who underwent the same transplant technology, we observed a trend to a higher incidence of infection-related adverse events in patients who had received innovative DMAs, with special reference to Natalizumab. The size of the sample and clinical heterogeneity at HSCT prevent to draw any conclusions; however, these data support the need of a careful monitoring of the clinical outcome in the years following HSCT in patients who had received innovative DMAs before the transplant. Disclosure of Interest: None declared.[P333] P349 Efficacy and safety of brentuximab vedotin in combination with bendamustine as 2nd line salvage therapy and potential "bridge" to stem cell transplantation for patients with refractory Hodgkin Lymphoma E. Shaibani 1,* , P. Kaloyannidis 1 , M. Ali 1 , A. Al Shaghier 1 , K. Al Anezi 1 , H. Al Hashmi 1 1 Adult Hematology and SCT, King Fahad Specialist Hospital, Dammam, Saudi Arabia Introduction: High dose chemotherapy and autologous stem cell transplantation (autoSCT) represents the most reliable curative approach for patients (pts) with refractory or relapsed Hodgkin's lymphoma (HL) after induction therapy. Nevertheless, the outcome of pts failed post 1 st line salvage therapy is extremely poor and only 15-20% can achieve long term survival. For the majority of these pts the 2 nd salvage regimen remains ineffective; therefore there is an unmet need for more effective and less toxic approaches. Brentuximab vedotin and Bendamustine have been tested as single-agents in refractory CD30+ lymphomas, demonstrating significant activity and a favorable safety profile. However, the published data of their combination (B&B) as salvage treatment are to date limited. Material (or patients) and methods: In the present study we evaluated the efficacy and safety of the B&B combination in 6 HL-pts, aged of 26(17-34)ys and disease that had be proven to be refractory after at least 2 prior cycles of ESHAP given as 1 st salvage treatment, while 1 pt had also undergone prior autoSCT. Following a median of 12(7-31) months since initial diagnosis, the B&B-treatment was administered in an outpatient basis by i.v. infusion of 1.8 mg/kg brentuximab on day-1 and 90 mg/m 2 bendamustine on days-1 and 2, in 3-week cycles. At the time of B&B initiation, the disease stage was II, III and IV in 2, 2 and 2 pts respectively, while 2 pts also had B-symptoms. Results: After a median of 2 (2) (3) (4) (5) cycles, 4/6 (67%) pts achieved very good remission (480% metabolic response with PET/CT criteria) while in 2/6 (33%) the disease progressed. No major toxicity (WHO ≥ 3) was observed during therapy. Finally, all 4 responders underwent stem cell transplantation at a median of 1.5(1-2.5) months post B&B therapy. One out of four responders (previously treated with autoSCT) underwent alloSCT following RIC, while 2/4 responders mobilized successfully (collected CD34 +cells: 7.7 and 4.1x 10 6 /kg); in one pt the stem cell graft was collected prior to B&B treatment. All transplanted pts are alive and well at a median of 4,5(2-5) months post autoSCT. The 2 nonresponders pts currently are alive, on palliative treatment for 5 and 10 months. Conclusion: Our data, though restricted to a small number of pts, are in line with the limited published studies and support the evidence that the B&B treatment could be an efficacious, safe and promising approach for maximizing responses prior SCT in patients with refractory HL. The B&B combination merits further investigation in order to clarify its exact role as salvage treatment and a potential "bridge" to successful SCT in refractory lymphomas. PET negative splenomegaly affects engraftment kinetics but not survival in allogeneic hematopoietic cell transplant recipients with lymphoid malignancies F. Khimani 1,* , D. Introduction: It is unclear if persistent splenomegaly prior to allogeneic hematopoietic cell transplant (allo-HCT) influences post-transplant outcomes for lymphoid malignancies in the presence of a negative PET-CT scan. Material (or patients) and methods: We retrospectively reviewed records of patients who underwent allo-HCT for lymphoid malignancies (n = 152) between 2008 and 2013 at the Moffitt Cancer Center. Clinical data were captured through our research database and were supplemented with individual patient electronic medical record review. Pre-transplant CT and PET images of all patients were reviewed. Spleen volume (SV) was measured using the freehand volume segmentation tool in AW Workstation software (General Electric, Waukesha, WI, USA) on CT images. Splenic index (SI) was calculated as a product of width (W), thickness (Th) and length (L) of the spleen. Normal SV and SI was defined as SVo314.5 cm 3 and SI ≤ 480 cm 3 as described in the literature. Statistical analysis was performed using SPSS v21.0. Results: Among the study population 42.8% received an allo-HCT from an HLA-matched related donor, 36.2% from a matched unrelated donor, 12.5% from a mismatched unrelated donor, and 8.6% received a double-umbilical cord blood transplants. Most patients (61.8%) received myeloablative conditioning. Median age at transplantation was 52 (range 21-68) years. Pre-HCT spleen CT and PET images were available on 88% (n = 134) and 70.3% (n = 107) patients, respectively. Spleen volumes ranged from 90 cm 3 to 4684 cm 3 with a median of 290.5 cm 3 and a mean of 400.3 cm 3 . SI calculation showed a range from 50.3 cm 3 to 8276.4 cm 3 with a median of 582.1 cm 3 and a mean of 771.2 cm 3 . Majority of patients (83.1%) had PET negative spleen pre-HCT. The 2-year overall survival (OS) and progression-free survival (PFS) were 57.3% and 44.9%, respectively for all patients. Engraftment kinetics were significantly delayed in PET negative patients with splenomegaly with median days to neutrophil engraftment 16 vs 17 (P = 0.03) and median days to platelet engraftment 14 vs 16 (P = 0.04) but not PFS and OS as shown in fig 1. Conclusion: PET negative splenomegaly prior to allo-HCT results in delayed neutrophil and platelet engraftment but does not affect PFS or OS. Future studies using registry data or larger prospective studies are required to evaluate the impact of splenomegaly and its PET avidity on allo-HCT outcomes in lymphoid malignancies, considering the limited sample size in our study. A comparison study between allogeneic and autologous hematopoietic stem cell transplant for high-risk Peripheral T-cell Lymphomas (PTCL) H. Huang 1,* , Q. Wang 1 , T. Xu 1 , Z. Jin 1 , D. Wu 1 1 The First Affiliated Hospital of Soochow University, Soochow University, Suzhou, China Introduction: Peripheral T-cell lymphomas (PTCLs) are heterogeneous malignancies sharing common elements of chemotherapy resistance and poor outcome with standard treatments. Attempts to improve outcomes have included autologous or allogeneic hematopoietic cell transplantation (autoHCT or alloHCT). both modalities lead to durable remissions in recurrent disease settings and might be important in consolidating first remission. However, key questions remain, including relative efficacy of autologous versus allogeneic approaches, and HCT timing (first-line consolidation v relapse). Herein, we analyzed outcomes of 60 cases recieved autoHCT or alloHCT in our center. Material (or patients) and methods: From July 2007 to July 2014, Outcomes of 60 patients undergoing autologous HCT (autoHCT) or allogeneic HCT (alloHCT) were analyzed retrospectively. Primary outcomes were nonrelapse mortality (NRM), relapse/progression, progression-free survival (PFS), and overall survival (OS). Results: All 60 patients were at high risk group (carried with IPI ≥ 3), with a median age of 31 (13-55) years old. Of the 60 cases, 22 were PTCL-not otherwise specified (PTCL-NOS), 22 with ALK negative anaplastic large cell lymphoma (ALKnegative ALCL) and 16 with angioimmunoblastic T-cell lymphoma (AITL). Before receiving transplantation, 40/60 patients were in complete remission (CR) and 20/60 patients were not remission (NR). Twenty-one (21/60) received allo-HSCT and thirty-nine patients (39/60) received auto-HSCT. In the 20 NR patients before transplant, 11 patients received allo-HCT and 9 patients received auto-HCT. After a median followup of 39 (range 1-96) months, the K-M analysis showed that the 5-year PFS for auto-HSCT and allo-HSCT were 61% and 60% (P = 0.724). 1-year NRM was higher (22.5% v 7.8%) for alloHCT. The 5-year OS for auto-HCT and allo-HCT were 62% and 61% (P = 0.724). There were no statistically significant differences between the auto-HSCT and allo-HSCT. However, autoHCT recipients were more likely in complete remission (CR; 76.9% v 47.6%; Po0.01) and with chemotherapy-sensitive disease. Conclusion: Both autoHCT and alloHCT approaches can benefit patients with PTCL. We did not find a difference in PFS and OS between autoHCT and alloHCT, although NRM increased significantly in alloHCT. However, our results suggest that outcomes of alloHCT are better for refractory and relapsed patients with PTCL. Introduction: Limited data exist regarding the efficacy and toxicity of DHAP regimen compared to high-dose cyclophosphamide (HDC) in patients with NHL undergoing ASCT. The aim of this study was to investigate the effects of DHAP plus granulocyte-colony stimulating factor (G-CSF) as a stem cell mobilization regimen on efficacy of mobilization, toxicity, and transplant outcomes, compared to HDC plus G-CSF in patients with NHL, who were responsive to front-line cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) or rituximab-CHOP chemotherapies. Material (or patients) and methods: We consecutively enrolled patients with NHL treated with CHOP or rituximab-CHOP and subsequent chemomobilization using HDC (4.0 g/ m 2 ) or DHAP (cisplatin 100 mg/m 2 D1, cytarabine 4.0 g/m 2 D2, dexamethasone 40 mg D1-4) plus G-CSF regimens for up-front ASCT in 3 Korean institutions between 2004 and 2014. Results: Ninety-six patients (57 male, 39 female) with a median age of 48 years (range, 18-66) were included. Sixty-five patients (67.7%) received HDC and 31 (32.3%) received DHAP. The most common histologic type was diffuse large B-cell lymphoma (DLBCL, N = 52, 54.2%), with the second most common subtype was peripheral T-cell lymphoma, not otherwise specified (N = 27, 28.1%). Anaplastic lymphoma kinase-negative anaplastic large cell lymphoma represented 12.5% (N = 12) and angioimmunoblastic lymphoma 5.2% of the cases. 92.3% of patients with DLBCL were initially treated with R-CHOP chemotherapy. Mobilized CD34+ cells were significantly higher in DHAP group than HDC group (median, 16.1 vs. 6.1 x 10 6 /Kg, P = 0.001, respectively). Successful mobilization (mobilized CD34+ cell count 3 5.0 x 10 6 /Kg) was observed in 27 patients (87.1%) of DHAP group and 40 (61.5%) of HDC group (P = 0.011). In multivariate analysis, baseline bone marrow involvement (odds ratio [OR], 4.60 [95% CI, ]), no prior radiotherapy (OR, , WBC counts at first apheresis day41, 785/uL (OR, ]), and DHAP regimen (OR, 4.12 [95% CI, ]) were independent predictors for successful mobilization. Febrile neutropenia, none of which were fatal, developed 3 patients (9.7%) in DHAP group, which was less frequent than HDC group (21 patients [32.3%], P = 0.043). Median time to neutrophil/platelet engrafts and overall survival did not significantly different according to chemomobilization regimen. Conclusion: DHAP regimen was significantly associated with higher yields for stem cell mobilization and lower incidence of febrile neutropenia than HDC. Therefore, DHAP plus G-CSF can be an effective chemomobilization regimen in patients with NHL undergoing up-front ASCT. Disclosure of Interest: None declared. Introduction: Peripheral T-cell lymphoma (PTCL) is a neoplasm arising from mature T lymphocyte. This disorder accounts for 7% of all hematologic malignancies in adults. According to the NCCN guideline, upfront consolidative autologous stem cell transplantation (ASCT) is an option for patients with PTCL achieving complete response (CR) after induction chemotherapy. However, the evidence of ASCT is not concrete because no randomized clinical trial (RCT) has been performed so far. Thus, we tried to show the effectiveness of upfront ASCT merging previous RCT including patients with mature PTCL. Material (or patients) and methods: PubMed, EMBASE, and Cochrane database were searched to extract eligible studies, which were randomized controlled trials comparing upfront autologous stem cell transplantation (ASCT) with conventional cytotoxic chemotherapy for non-Hodgkin lymphoma (NHL) including T-cell lymphoma. The primary outcome was progression-free survival (PFS). The secondary outcomes were overall survival (OS) and complete response (CR) rate. The risk of bias was evaluated by the Cochrane tool. All measures were pooled using fixed-effects models. Conclusion: RIC allo-HCT is a feasible and effective strategy in patients with relapsed and refractory multiple myeloma. Age455 years and pre allo-HCT chemorefractory disease are important negative prognostic factors in this setting. Biologically targeted drugs are effective for relapses after allo-HCT and lenalidomide retreatment could still be beneficial probably due to its immune-modulatory effects. Disclosure of Interest: None declared. or VAD (24%). ECOG PS was ≥ 2 in 19% of patients and hematopoietic cell transplantation comorbidity index (HCT-CI) risk score was high in 70%. Melphalan 200 mg/m2 (MEL200) conditioning was given to 44%, MEL140 to 10% and MEL100 to 46% of patients, respectively. The impact of age, pretransplant disease status (DS), ECOG PS, HCT-CI score and MEL dosage on progression-free (PFS) and overall survival (OS) was analyzed. Results: CR after the first ASCT was 32% and VGPR was 39%. Grade 4 non-hematologic toxicity was observed in 9 patients (15%) (mucositis in 6, infections in 3). Transplant related mortality (TRM) was 0%; death rate was 18% and was due to progressive disease in 90% of cases. 2-year PFS was 77% and OS 90%. At univariate analysis no significant survival difference was seen between patients aged 65-69 and ≥ 70 years (3-year PFS 62% vs 35% and OS 84% vs 87%, respectively). PS 0-1 vs ≥ 2 significantly impacted on OS (3-year PFS 55% vs 64%, P = NS; OS 88% vs 69%, p 0.01, respectively). HCT-CI risk score and pretransplant DS had no impact on outcome. A significantly better outcome was seen in patients receiving MEL200, compared to lower MEL doses (3-year PFS 76% vs 40%, p 0.019; 3-years OS 100% vs 71%, p 0.0005; Figure 1 ). Considering conditioning regimen, PS, and pretransplant DS in a multivariate analysis, MEL200 independently affected PFS (p 0.02). Conclusion: ASCT is feasible and effective in MM patients aged ≥ 65 and also in selected patients aged ≥ 70 years. Neither allogeneic HCT-CI risk score nor age or DS were predictors of TRM and outcome whereas PS42 affected OS and full-dose MEL conditioning significantly improved PFS and OS. Objective methods are warranted to assess transplant eligibility and the optimal dosage of the conditioning regimen in order to optimize the use and the efficacy of ASCT in elderly MM patients. Disclosure of Interest: None declared. Introduction: High-dose chemotherapy with autologous haematopoietic stem cell transplantation (AHSCT) in multiple myeloma patients is recommended as consolidation therapy especially in younger patients that can increase both progression free survival (PFS) and overall survival (OS).The role of AHSCT in elderly pts (over 65 years) is still no clearly defined. Material (or patients) and methods: We present the treatment's outcomes of 50 pts with MM who were underwent AHSCT between July 2005 and March 2015 in our Department. There were 30 male and 20 female, with a median age of 66,5 (range 65-81 yrs). Salmon-Durie scale staging at the diagnosis was as follow: I-2%(n = 1), II-36%(n = 18), III-62% (n = 31); ISS In = 14 (28%), II-n = 31% (62%), II n = 5 (10%). 3pts renal failure was presented at the diagnosis. The induction treatment consisted of immunomodulation drugs or proteasome inhibitors. The complete response was achieved in 14 patients (28%), the very good partial respond in 17 pts (34%), the partial response in 11 patients (22%) and stable disease in 8 pts (16%). Conditioning regimens preceding AHSCT consisted of melphalan in dose 200 mg/m 2 in 47 cases, 3 pts got melphalan in dose 140 mg/m 2 due to renal insufficiency. A median number of transplanted CD34+ cells was 4,1 (2,0-17,1x10 6 /kg). Results: Eight years' disease free survival (DSF) was estimated to be 19% with a 10 years' overall survival (OS) of 29%. All patient successfully engrafted. Hematopoietic recovery was as following: WBC count41,0x10 9 /L after median of 14 days (range 8-21 days), ANC4 0,5x10 9 /L after median of 14 days (range 8-21 days) and platelet count 450x10 9 /L after median of 14 days (range 10-21 days). The major complications after AHSCT were rare and included: bacterial infections of the respiratory tract (n = 15), viral infections (n = 10), oral mucositis (n = 9). Conclusion: HDT based of melphalan followed by AHSCT seems to be effective and safe procedure for MM patients over 65 year old in everyday clinical practice. Disclosure of Interest: None declared. Hematology Oncology, Soonchunhyang University Bucheon Hospital, Bucheon, 2 Hematology Oncology, Soonchunhyang University Hospital, 3 Hematology Oncology, Kyung Hee University Hospital, 4 Hematology, The Catholic University Hospital, Seoul, Korea, Republic Of Introduction: In myeloma patients with partial response (PR) following high-dose (HD) therapy with autologous stem cell transplantation (ASCT), the available data are needed regarding optimal treatment regimens to improve the depth of response, and lead to substantial improvement of clinical outcomes. Recent data demonstrated that patients with only a PR to induction chemotherapy and ASCT derived substantial benefit from consolidation therapy. An improvement in the depth of response was seen in 26% of patients who had reached PR following ASCT. Material (or patients) and methods: We evaluated the clinical benefits according to treatment strategies, tandem ASCT or consolidation therapy with novel agents in patients with partial response to HD chemotherapy and ASCT. Response evaluation was done around D+100 after ASCT. A total of 50 patients who received tandem ASCT (n = 20) or consolidation therapy with novel agents (n = 30) were recruited. Results: Median follow-up duration was 41 months. CR rate in tandem ASCT group was higher than the group of novel agents (60% vs 30%). The rate of deep response including very good partial response (VGPR) was also higher in tandem ASCT group (85% vs 40% for novel agents group). And the time to next treatment of the tandem ASCT group was longer than the novel agents group (12.8 months vs 7.3 months, P = .05). However, no different in overall survival was seen between two groups, the estimated 3-year rates being 62.4% for the tandem ASCT group and 57.0% for the novel agents group, respectively. Patients who achieved deep response including CR and VGPR survived longer than patients with response less than VGPR (75.9% vs 47.9%, P = .03) in both groups. The efficacy of treatment regimens including novel agents as a salvage therapy was not significant different according to type of novel agents. Conclusion: In patients who achieve a partial response after HD-chemotherapy and ASCT, tandem ASCT is still an effective treatment option and achieving deeper response is associated with a better clinical outcome. Disclosure of Interest: None declared. Introduction: Although treatment of patients with multiple myeloma (MM) showed progress due to novel agents such as bortezomib, thalidomide and lenalidomide, autologous stem cell transplantation (ASCT) following melphalan high-dose chemotherapy (HDCT) is still a gold standard for younger and fit patients. Material (or patients) and methods: We conducted a retrospective analysis of 320 myeloma patients receiving either single (N = 286) or tandem (N = 34) ASCT following melphalan HDCT in the period from 1996 to 2012 at University Hospital Muenster. Main objectives were the analysis of haematological response to ASCT, overall survival (OS), progression-free survival (PFS), and transplant-related death (TRD). Adverse risk factors were calculated in multivariate Cox regression models. Additionally, the impact of induction therapy containing novel compounds on survival of the myeloma patients receiving ASCT was assessed. Results: Median age for the entire cohort at time of ASCT was 58 years (range 35-75). 51% of the patients (N = 162) presented A trend for longer OS was observed for ISS1 vs ISS3 (82% vs 62%). Conclusion: We conclude from this single centre study that high-dose therapy with ASCT is an effective and safe treatment option for patients with MM. The outcomes were independent of age ( o 55 years vs ≥ 55), time from diagnosis to ASCT, while patients with ISS 1 had a trend for a better outcome after transplantation. Our results were not significantly different from the results of other European transplantation centers. Disclosure of Interest: None declared. Introduction: Autologous hematopoietic stem cell transplantation (auto-HSCT) and development of new drugs have significantly improved the prognosis of multiple myeloma (MM) patients. However patients relapse and no definitively cure is known until now. Allogeneic hematopoietic stem cell transplant (Allo-HSCT) represents a potentially curative approach particularly for high risk patients, but its role and indication is still controversial 1 . The objective of this retrospective study was to analyzed the different approach to allo-HSCT over the decades in MM patients in our centre. Material (or patients) and methods: Between January 1990 and August 2015 51 Allo-HSCT was performed in 48 pts with MM (27 IgG, 13 IgA, 9 micromolecular) ; 31 were male and 17 were female. Median age was 50 years (range 34-65). 38 were from identical sibling donor (3 from singenic donor), 13 from matched unrelated donor (MUD). Stem cell source was PBSC in 37, BM in 12 and CB in 2. Conditioning regimen was myeloablative (MAC) in 18 and reduced intensity (RIC) in 32. MAC regimen consisted in Busulphan plus Cyclophosphamide (8) or Melphalan (6), or TBI based (4) . RIC regimen was Thiotepa-Cyclophosphamide (15), TBI-Fludarabine (9), TBI-Fludarabine-Melphalan (6), TBI-Fludarabine-Cyclophosphamide (2) . We identified three groups in this population. Group 1: 16 patients who received Allo-HSCT upfront after induction therapy; group 2: 5 patients who were transplanted after auto-HSCT (tandem Auto-Allo-HSCT) and group 3: 27 patients transplanted in relapsed after Auto-HSCT (23/27), 4/27 after chemotherapy. Results: During the period 1990-2002, before the introduction of new drugs, the upfront use of allo-HSCT (group 1) was prevalent (12/20 pts) and 50% of pts received an Allo-MAC. Between 2002 and 2010, after development of new drugs, over 20 pts, 4 were in group 1, 5 in group 2 and 11 were in group 3 (RIC 16 vs MAC 4) . Pts in group 3 (9/11) had received a reinduction therapy with proteasome inhibitors or immunomodulatory drugs before Allo-HSCT. Finally, after 2010 9 pts recipients of Allo-HSCT were all in Group 3 and they had received a reinduction therapy with proteasome inhibitors or IMIDs before Allo-HSCT. All pts received a RIC. OS was 17.6% at 15 years. 24 patients died for relapse of multiple myeloma, 15 for non relapse mortality. One patient died for complications of amyotrophic lateral sclerosis. 8 patient are alive (3 relapsed and 3 with chronic Graft versus Host Disease).Conclusion: This retrospective study shows different ways of applications of allo-HSCT for MM in our centre over almost 3 decades: particularly in the old era was prevalent the indication of allo-HSCT upfront and MAC. After a short period with a tandem auto-alloHSCT approach, the introduction of new drugs in MM scenario has shifted the allo-HSCT indication to relapsed patients at least in partial response after a new drug reinduction.The prognostic risk scores are directly obtained by adding up the relevant log-hazard ratios, which allows dividing patients into three risk groups, low, medium, high, defined by tertiles in the study population. Cumulative incidence plots of relapse for each of the three groups are shown. Conclusion: These risk scores may help to stratify patients according to their risk of relapse after stem cell transplantation which can be used for stratification in further prospective trials using posttransplant therapies at different time points after stem cell transplantation to reduce the risk of relapse. Disclosure of Interest: None declared. Mahak Cancer Children's Hospital, Tehran, Iran, Islamic Republic Of Introduction: The long-term survival probability for children with high-risk neuroblastoma was less than 15 percent. Better results have been achieved using an aggressive multimodality approach that includes chemotherapy, surgical resection, high-dose chemotherapy with hematopoietic stem-cell rescue, and radiation therapy. The prognosis of high-risk neuroblastoma after conventional chemotherapy is generally poor. The use of high-dose chemotherapy with autologous hematopoietic stem cell rescue in consolidation has resulted in improvements in survival, although further advances are still needed. Material (or patients) and methods: Twenty-seven patients age 4 to 20 years (median 4.5 years, M/F = 17/10), with high risk, relapsed or metastatic underwent ASCT in our hospital (from 2012-2015) . 25 of the 27 patients underwent a first HDCT/ASCR and 2 patients underwent a second HDCT/ASCR. 10 of the 27 patients who were evaluated for N-myc amplification were positive. Status at transplant was: second complete remission (CR2): n = 12; CR3: n = 9, (CR 43) n = 6. All patients received chemotherapy-based conditioning regimens: MEC regimen consisting of carboplatin for 3 days, VP-16 for 3 days and melphalan in 2 days was the primary cytoreductive regimen. Peripheral blood (PB) was the source of progenitor cells in 27 patients. All patients engrafted. Results: The median mononuclear cell dose was 4.3 × 10 8 / kg. The median time to absolute neutrophil count40.5 × 109/L was 14 days, and the median time to platelet count420 × 10 9 was 15days. Two patient out of 27 (3.7%) experienced transplant-related mortality. Eight patients relapsed after transplant (mean 9 m.). With a median follow-up of 19 months (3-34 months) after transplant the event-free survival (95% confidence interval) was 0.61. 20 patients remain alive. Conclusion: Longer follow-up is required to evaluate fully efficacy and long term survival of our patients. However, further studies will be needed to decrease the toxic death rate in the transplantation. Disclosure of Interest: None declared. Introduction: T cells genetically modified to express antigen specific T-cell receptors (TCR) are a novel HLA-restricted immunotherapy tool. In particular, Cancer Testis Antigens (CTAs) are relevant cytoplasmic tumor proteins, with an overall restricted expression in tumor tissues of different origin. Among all the well-known CTAs, we focused our interest on the Preferential Antigen Expressed in Melanoma (PRAME), since its expression has been described in many solid and hematologic cancers. The Leiden University has selected a new high affinity αβTCR specific for the CTA PRAME, isolated from a HLA-A2 patient affected by AML, given allo-HSCT 1 . Aim: To evaluate the anti-tumor activity of T lymphocytes genetically modify with a retroviral vector carrying PRAME αβTCR and the new suicide gene inducible caspase 9 (iC9) in the context of medulloblastoma tumor. Material (or patients) and methods: The sequence of αβ chains with high affinity for PRAME-SLL peptide have been cloned in frame by 2 A sequence in a retroviral vector carrying also iC9. Activated polyclonal T cells were transduced by iC9. SLL-PRAME-TCR retroviral supernatant. Transduced T cells were characterized by Vb staining, as well as dextramer analysis for the αβTCR pairing. iC9.SLL-PRAME-TCR T-cell functionality was tested by IFNγ ELISPOT, 51 Cr cytotoxic release and co-culture assays against tumor cell lines, either HLA-A2 matched [DAOY] or not (RS4;11, TOM-1 and D283). iC9 activation was evaluated by AnnexinV/7AAD staining after exposure to 20 nM of AP1903. Results: T cells genetically modified with iC9.SLL-PRAME-TCR showed 72 ± 18% of Vb1+ T cells (vs 4 ± 2% of Vb1+ T cells in non-transduced (NT) T cells, P40.0001). Moreover, 45 ± 12% of total CD8+ T cells showed a positive staining with SLLdextramer. The level of Vb1+ and SLL-dextramer+ T cell population seems to be stable over time in the applied culture conditions. We tested the iC9.SLL-PRAME-TCR T cell anti-tumor activity in IFN-γ ELISPOT against Prame+ HLA-A2+ DAOY cell line, derived from a patient with medulloblastoma, and several HLA-A2 negative cell lines. iC9.SLL-PRAME-TCR T cells showed 752 ± 320 IFNγ spot forming cells (IFN-SFC) when stimulated with DAOY at an E/T ratio 5:1 (vs 50 ± 16 IFN-SFC observed in the NT control condition, P40.001). Negligible IFN-SFC was observed stimulating iC9.SLL-PRAME-TCR T cells with HLA-A2 negative Prame+ D283 medulloblastoma cells. 51 Cr release and co-culture assays showed that iC9.SLL-PRAME-TCR T cells were cytotoxic and able to eliminate tumor targets. Indeed, more than 64% of specific lysis was recorded against DAOY cells when incubated in the presence of iC9.SLL-PRAME-TCR T cells at an E/T ratio of 40:1 (vs 18% of lysis in the presence of NT cells). Co-culture assays showed that iC9.SLL-PRAME-TCR T cells are able to eliminate the target even at E/T ratio of both 10:1 and 1:1, after 7 days. Addition to the colture of AP1903 eliminate iC9.SLL-PRAME-TCR T cell activity and induce apoptosis in 93% of the transduced cells. Conclusion: The novel multicistronic retroviral vector developed as a safe clinical grade TCR vector carrying the safety switch iC9 shows a robust anti-tumor activity when transferred into polyclonally activated peripheral blood T lymphocytes. This vector may have immediate applications in cancer gene therapy of patients with PRAME+ HLA-A2+ relapsed/refractory disease. Introduction: The STWP is dedicated to pre-clinical, translational and clinical studies of cell therapy for solid tumours, including autologous and allogeneic stem cell transplant, active and adoptive immunotherapy, lymphoablative therapy with expanded T cells. So far, 52,824 hematopoietic stem cell transplant (HSCT) procedures have been reported to the Solid Tumor Working Party section of the EBMT registry. 97% were autologous HSCT, and 3% allogeneic. Material (or patients) and methods: Germ cell tumours (GCT) represent the most common malignancy affecting adolescent and young adult men in Europe and the United States, and represents the main indication for autologous HSCT in adults. Conventional-dose chemotherapy (CDCT) and HSCT are both used as salvage treatment for patients with metastatic GCTs. Conflicting data exists on whether HSCT is superior to CDCT. Recent retrospective data suggest that HDCT may be superior to conventional chemotherapy. To test this hypothesis, an international phase III randomised trial is planned (TIGER study).Breast Carcinoma (BC) is the most Introduction: A graft-versus tumor (GVT) effect has been postulated after allogeneic stem cell transplantation (ASCT) for metastatic renal cell cancer. It is not known, however, which factor(s) determine the immune response to the transplant. HLA-G, a non-classical HLA molecule ligand for the inhibitory NKR ILT2/4, has recently been shown to be expressed both on tolerogenic DC and on effector T cells, suggesting an important role for this molecule in peripheral tolerance mediated by Tr1 cells. The 14 bp polymorphism of the HLA-G gene determines differential splicing with generation of an unstable form of mRNA, which in turn results in decreased expression levels of both cell surface and soluble HLA-G. The role of HLA-G polymorphism in the immune response to transplantation currently is not known. Homozygosity for the 14 bp del/del genotype has recently been associated with increased risk of acute graft versus host disease (aGvHD) grade II-IV after unrelated and related ASCT for betathalassemia. Material (or patients) and methods: We have retrospectively analyzed the impact of the 14 bp HLA-G polymorphism for GvT activity after allogeneic stem cell transplant in a series of 55 renal cell cancer patients who had undergone ASCT in different EBMT Centers (Milan HSR, Marseille, Stockholm, Clermont Ferrand) and surviving at least 30 days after transplant. Patients received an allogeneic stem cell graft from PB of matched related (N = 47) or unrelated (N = 8) donors. Conditioning regimens and aGVHD prophylaxis varied among Centers, according to Institutional protocols. Samples from fresh or frozen peripheral EDTA-blood from patients and donors were analyzed for 14 bp polymorphism within the HLA-G gene using a locally established PCR-SSP approach with two PCR reactions. The results (homozygosity for the 14 bp insertion: ins/ins, or the 14 bp deletion: del/del or heterozygosity: del/ins) were correlated with overall survival (OS) and acute GvHD grade 2-4 as primary endpoints by standard statistical methods (Cox and logistic regression, Kaplan Meier probability, in univariate and multivariate models). Results: Five-year overall survival (OS) of the entire series has been 13% at a median follow-up of 66 months. OS was analyzed as a function of 14 bp deletion: patients with homozygous deletion had an OS of 21%, patients with no deletion had an OS of 0%. Patients heterozygous (del/ins genotype) had an intermediate survival (9%). In multivariate analysis, taking into account age and donor type, HLA-G deletion was of borderline significance for OS (P = 0.062, CI.956-5947). There was no effect of HLA-G genotype on grade 2-4 aGVHD incidence (P = ns). Conclusion: These preliminary results suggest that HLA-G polymorphisms may be involved in immune response against renal cell cancer after allografting. We plan to complete the analysis of 14 bp polymorphism on further 45 samples, and to analyze 6 new polymorphisms.