key: cord-0743574-iveur6av
authors: Amadio, Jennifer M.; Rodenas-Alesina, Eduard; Superina, Stefan; Kozuszko, Stella; Tsang, Katherine; Simard, Anne; Aleksova, Natasha; Kobulnik, Jeremy; Fan, Chun-Po Steve; Wijeysundera, Harindra C.; Ross, Heather J.; McDonald, Michael A.; Duero Posada, Juan G.; Moayedi, Yasbanoo
title: Sparing the Prod: Providing an alternative to endomyocardial biopsies with non-invasive surveillance after heart transplantation during COVID-19.
date: 2022-02-13
journal: CJC Open
DOI: 10.1016/j.cjco.2022.02.002
sha: 8fc2934ee23d8f94df7161ff8799217e01ec811d
doc_id: 743574
cord_uid: iveur6av

BACKGROUND: The COVID-19 pandemic has reduced access to endomyocardial biopsy (EMBx) rejection surveillance in heart transplant (HT) recipients. This is the first Canadian study to assess the role for non-invasive rejection surveillance in personalizing titration of immunosuppression (IS) and patient satisfaction post-HT. METHODS: In this mixed methods prospective cohort study, adult HT recipients more than six months from HT had their routine EMBx replaced by non-invasive rejection surveillance with gene expression profiling (GEP) and donor-derived cell-free DNA (dd-cfDNA). Demographics, outcomes of non-invasive surveillance score, hospital admissions, patient satisfaction, and health status on Medical Outcomes Study 12-item Short Form Health Survey (SF-12) were collected and analyzed using t-tests and chi-squared tests. Thematic qualitative analysis was performed for open-ended responses. RESULTS: Among 90 patients, 31 (33%) were enrolled. 36 combined tests were performed; 22 (61%) were -GEP/-dd-cfDNA, 10 (27%) had +GEP/-dd-cfDNA, 4 (11%) had -GEP/+dd-cfDNA and 0 +GEP/+dd-cfDNA. All patients with a positive dd-cfDNA (range 0.19-0.81%) underwent EMBx with no significant cellular or antibody mediated rejection. 15 cases (42%) had IS reduction and this increased to 55% in patients with negative concordant testing. Overall, patients’ reported satisfaction was 90% and on thematic analysis they were more satisfied with less anxiety during the non-invasive testing experience. CONCLUSIONS: Non-invasive rejection surveillance was associated with the ability to lower immunosuppression, increase satisfaction, and reduce anxiety in heart transplant recipients, minimizing exposure for patients and providers during a global pandemic.

Survival after heart transplantation (HT) has significantly improved since the standardization of the monitoring and treatment of allograft acute cellular rejection (ACR).

Universally, HT programs have protocolized surveillance endomyocardial biopsies (EMBx) to monitor for allograft rejection as the "gold standard". However, there remains a continued desire to replace or reduce the number of biopsies given cost, invasiveness and potential for complications. [1] [2] [3] [4] This has been amplified during the COVID-19 pandemic, where access has been reduced for invasive care and alternative non-invasive methods to monitor for ACR are particularly relevant.

Indeed, during the COVID-19 pandemic, our institution has experienced closures in the catheterization lab and also limited procedure times due to pandemic restrictions and personnel redeployment. As such, we were unable to use routine EMBx to safely reduce immunosuppression in cardiac transplant recipients and required alternative surveillance monitoring to assist with immunosuppression titration in these patients.

Significant advances have been made in the field of novel transplant rejection biomarkers. Other centres are beginning to use non-invasive screening methods to follow heart transplant patients remotely and reduce hospital visits and procedures. 5 Gene expression profiling (GEP) (AlloMap ® , CareDx ® Inc, Brisbane, California), a non-invasive blood test for the surveillance of transplant recipients, reduces the frequency of EMBx with no difference in adverse outcomes. 2, 3 However, GEP has limitations in that it cannot be used within 55 days of HT, or in the setting of recent high dose immunosuppression as these factors can influence predictive value of GEP testing and these patients were excluded from trials. 3, 6, 7 GEP may be falsely positive in recipients with active viral infections and should be interpreted with caution in this cohort. 3, 6, 7 J o u r n a l P r e -p r o o f More recently, donor-derived cell-free DNA (dd-cfDNA) (AlloSure ® , CareDx ® Inc, Brisbane, California) from peripheral blood samples can be safely used to rule out ACR and Antibody Mediated Rejection (AMR) as early as 2 weeks after transplantation and allow for surveillance of rejection in HT recipients. 8 HeartCare ® (CareDx ® Inc, Brisbane, California) provides a comprehensive assessment of graft rejection by combining GEP and dd-cfDNA, where the tests complement each other to provide a more holistic picture of graft health. 9 Using blood samples to rule out rejection reduces the need for invasive surveillance and potential complications associated with invasive EMBx. Non-invasive surveillance reduces procedural risk to patients and leads to improved patient satisfaction and health status survey scores. 2, 3 In this study, we aimed to reduce the number of EMBx in HT recipients by using combined non-invasive screening kits. The goal was to provide surveillance monitoring for rejection and immunosuppression adjustments, improve patient quality of life, and mitigate risk during invasive testing to HT patients and healthcare workers during the COVID-19 pandemic.

This was a mixed methods study with a convergent, parallel design, collecting quantitative and qualitative data simultaneously but analyzing results separately. First, a prospective cohort study of HT patients screened by the study team for eligibility to have their EMBx replaced by non-invasive screening. Second, thematic analysis was conducted on openended survey results. The study was approved by the Quality Improvement Board with a waiver of consent from the research ethics board (QI 21-0420).

J o u r n a l P r e -p r o o f Eligible HT recipients included adult patients (>17 years) undergoing surveillance EMBx at least 6 months post-HT. Patients were excluded if they were multiorgan transplant recipients, required dialysis, were less than 6 months after HT, or at high risk for rejection (recent ISHLT ACR ≥ 2R rejection (on one of prior 2 biopsies in the last 6 months), new graft dysfunction (drop in left ventricular ejection fraction by 15% or more on echocardiogram), or patient reported symptoms of heart failure). This study was part of a quality improvement initiative, as such there was no randomization or control groups. If participants met the inclusion criteria, they were evaluated by our transplant team. If no exclusion criteria were met, participants were scheduled for non-invasive testing.

Patients were enrolled during the third wave of the COVID-19 pandemic (from May 2021 to August 2021) and had their routine EMBx replaced by non-invasive screening with combined GEP and dd-cfDNA kits. Adjustment to immunosuppression and routine care was provided to all patients based on our previously established heart transplant program protocol. A high-risk or positive AlloMap ® score was ≥ 34 for rejection but an abnormal GEP did not mandate a biopsy. A threshold dd-cfDNA level of ≤0.15% was used to rule out significant rejection. If the dd-cfDNA result was positive, then a "for cause" EMBx occurred to screen for histopathological signs of rejection (ACR and/or AMR) ( Figure 1 

Thematic analysis was performed for open-ended responses from the patient satisfaction survey. The words of participants were carefully examined, and recurring ideas were separated into different codes, consistent with the NVivo coding process described by Miles, Huberman and Saldana (2014). 11 Codes were selected and thematic analysis was used to establish themes and assertions to interpret qualitative data.

Continuous variables are presented as median (IQR) while categorical variables are presented as frequency. Comparisons between groups were done using t-tests or Mann-Whitney U test for continuous variables according to normality, and using 2 or Fisher's tests for categorical variables.

Demographics 171 patients were scheduled for EMBx over the enrollment period, 81 were excluded and 90 were eligible. Among 90 remaining patients after exclusion criteria, 31 (33%) were enrolled to have their routine EMBx replaced by non-invasive rejection screening due to the limited number of testing kits for eligible participants ( Figure 2 ). Of the 37 kits processed, 36 combined J o u r n a l P r e -p r o o f total tests were available. The median time since HT was 1.9 years (IQR 1.1,6.4 years) (Supplementary Figure 1 ). As shown in Table 1 , this cohort was predominantly male (68%), Caucasian (52%) with a primary reason for transplant related to a non-ischemic etiology (51.6%). With regards to graft function, 10% had a history of primary graft dysfunction, 32% a history of cardiac allograft vasculopathy (CAV), 53% had a history of treated acute cellular rejection (ACR) at 1 year, and 7% a history of treated antibody mediated rejection. The majority of patients remained on prednisone (84%) at the time of non-invasive testing.

There were a total of 36 combined tests, among which 22 (61%) had a negative GEP and dd-cfDNA; 10 (28%) had a negative dd-cfDNA but a positive GEP, and 4 (11%) had a negative GEP but a positive dd-cfDNA. None had a positive dd-cfDNA and GEP ( Table 2 ). All four patients with a positive dd-cfDNA (range 0.19-0.81%) underwent EMBx; 2 patients had ISHLT ACR Grade 1R and 2 patients had Grade 0 rejection. One patient with positive dd-cfDNA was within 6 months post-HT and three were ≥1 year post-HT. None of the patients had evidence of AMR (all ISHLT pAMR 0). Five patients had serial combined testing as shown in Table 3 .

Overall, non-invasive screening safely eliminated 32 (89%) endomyocardial biopsies.

Baseline immunosuppression dose and/or drug levels were determined for all enrolled HT patients and stratified by time post-transplant (Supplementary Table S1 ). In 15 cases (42%), testing results triggered a decrease in immunosuppression (IS) dosage. In patients who had a negative concordant test, IS was reduced among 55% while IS was reduced in 30% among those with a positive GEP and a negative dd-cfDNA ( Figure 3 ). Specifically, 80% of patients had a reduction in prednisone, 13% had a reduction in mycophenolic acid, and 7% had a reduction in J o u r n a l P r e -p r o o f calcineurin inhibitor.

In our cohort, 17 patients (51%) had received two COVID-19 vaccines, 1 had contracted COVID-19 6 months prior to non-invasive testing. 23.8% of the tests after vaccination had a positive GEP result compared to 33.3% of those performed in non-vaccinated patients (p=0.690).

The median time from vaccine to non-invasive testing was 51 days (IQR 17,66). No significant relationship was seen between the score positivity and time from vaccination. No difference was seen in the median GEP scores between those vaccinated and those who were not (GEP score in vaccinated 30.0 vs 33.5, p= 0.721).

All patients completed the SF-12 and a patient satisfaction survey. Patients' self-reported satisfaction was 90%, indicating that they were very satisfied with non-invasive biomarkers, testing was less painful, they had good pain control and greater preference for non-invasive surveillance testing compared to EMBx (Figure 4) . The median anxiety level score was 50 (IQR 10,71) prior to EMBx compared to 2.5 (IQR 0,7.5) prior to non-invasive testing (Figure 5a ). The median physical health score was 43 (IQR 37,53) and the median mental health score was 53 (IQR 44,58) ( Figure 5b) . A mean score of 50 points out of 100 represents the United States population average.

For qualitative data from patient satisfaction surveys, four codes ("emotions" (pain, anxiety, fear), "time", "compared to biopsy", "accuracy") were used to analyze data and uncover two common themes. Themes uncovered were "Superiority to biopsy" (Codes: emotions, time, compared to biopsy) and "Mental or Physical Stress" (Codes: emotions, time, accuracy). Patients described feeling more satisfied and less emotionally distressed with the non-invasive screening.

Reasons included that they felt this test was superior to a biopsy ("so much less invasive", "no pain or any recovery time as compared to biopsy", "much faster"). A visual representation of the qualitative open ended survey responses was created in the form of a word cloud ( Figure 6 ).

There were multiple emotions captured under the theme of Mental or Physical stress that patients were able to avoid by replacing a painful, invasive or very stressful procedure with something as simple and relatively painless as a blood draw. The non-invasive screening increased patient satisfaction with "no stress and concern" and was "much less pain, anxiety and quicker". All participants reported increased satisfaction with non-invasive testing, reflected by combined themes of superiority to biopsy and reduced mental or physical stress:

• "I am so excited that there is a new way to test rejection! [Before a biopsy] I was terrified and started weeping and shaking...each time my anxiety would skyrocket.

...Each time, I knew there was the possibility that I would die. The tiny needle prick to extract DNA info was amazing!!! I am so grateful".

• "Quick, painless and stress free. Anything is better than a heart biopsy."

Negative feedback on the non-invasive testing was patients' concerns about interpretation of these new results. Overall, patients were more satisfied, had less fear and anxiety, and were pleased with the experience of non-invasive testing.

Endomyocardial biopsies (EMBx) are used to diagnose or exclude rejection as part of a routine surveillance protocol which can allow for the reduction of higher doses of immunosuppression with its associated toxicities. 1,2 Despite resource limitations and concerns of infection during COVID-19, it is essential to continue to taper IS to the lowest tolerated levels to J o u r n a l P r e -p r o o f avoid potential complications of long-term immunosuppression use such as malignancy or early and late post-HT infections. 1, 12, 13 In our patient population, 84% were on prednisone which highlights the importance of ongoing surveillance and reduction of immunosuppression. Patients with higher prednisone doses have greater risk of infection episodes 12 and our group has shown that tapering off prednisone within the first year is associated with reduced risk of CAV. 13 Non-invasive screening with GEP and dd-cfDNA are proving to be safe, practical, and desired methods of surveillance with high patient satisfaction. 2, 3, 8, 14 Our study, a quality improvement initiative during COVID-19, allowed us to continue rejection surveillance and guide medical decisions at a time where access to biopsy procedures was limited. Through this initiative, 89% of invasive EMBx procedures were safely eliminated in our cohort. While many HT recipients could have had biopsies delayed to a later time, this initiative allowed for further reduction of immunosuppression in nearly half of the patients who underwent non-invasive surveillance. A strength of this study is that despite a global pandemic we were able to qualitatively demonstrate the patients' perspective of reduced anxiety and increased satisfaction with non-invasive screening. In immunocompromised patients, minimizing risk during waves of rising COVID-19 cases and reducing patients' anxiety surrounding testing is important to prevent nosocomial infection, reduce mental health burden, and improve quality of life. As such, we hope to continue this resource allocation strategy as the pandemic wanes.

Although the EMBx is a gold standard, there is variability in agreement of histopathological assessments resulting in unnecessary anti-rejection therapies and/or hospital admission. [15] [16] [17] Even experienced cardiac pathologists can overrate moderate rejection (ISHLT ACR ≥ Grade 2R) with low agreement when classifying significant ACR (≥ Grade 2R), which has important implications for immunosuppression and could lead to overtreatment. 15-17 A J o u r n a l P r e -p r o o f threshold GEP score of ≥ 34 for patients >6 months post-HT has a sensitivity of 51.1%, specificity of 63.4%, PPV 2.82% and NPV 98.5%. 18 In the multicentre Donor-Derived Cell-Free DNA-Outcomes GEP Registry (D-OAR) study, 740 HT recipients had biopsy samples paired with a quantified dd-cfDNA which showed a negative predictive value (NPV) of 97.1% using a threshold of 0.20%. 8, 19 Therefore, non-invasive surveillance provides a more objective way to exclude rejection to further reduce IS. 2, 3, 8, 14, 20 In our cohort, 4 patients underwent EMBx for discordant non-invasive screening with positive dd-cfDNA testing. None of these patients had clinically significant rejection (either ACR or AMR) requiring immunosuppression change, hospital admission, or further intervention.

While these are "false-positive", the increased levels of dd-cfDNA may be the result of other graft or endothelial injury that we are unable to detect in this study due to the limited follow-up time. 8, 14, 20, 21 Other studies using combination of non-invasive surveillance methods in patients over 6 months post-transplant were able to avoid the need for EMBx in many more patient using combination dd-cfDNA and GEP testing than isolated GEP screening (83.9% vs. 72.2%) 22 .

In this study, nearly a third of tests were discordant with a positive GEP and negative dd-cfDNA. The gene expression profiling test has several limitations that can be associated with a false positive test leading to a limited positive predictive value (PPV) for cellular rejection. We suspect that several false positive GEP results were associated with viral infection, including cytomegalovirus (CMV) which can cause elevated GEP with normal dd-cfDNA. 6, 7, 12 In our cohort, 5 patients had prior CMV infection and none of the patients had active CMV. Our group has shown that risk of CMV infection is higher at increased prednisone dose and for female recipients. 12 Patients with viral infections were more likely to have elevated GEP scores, although these scores remained below the threshold for rejection screening. 12 A similar sized, J o u r n a l P r e -p r o o f single-centre trial of 37 HT recipients greater than one year post-HT had six patients require EMBx after elevation in either GEP or dd-cfDNA testing. 23 In this study, 61% of patients remained on prednisone therapy after 1 year. In the patients with elevated dd-cfDNA levels, two patients had CAV and one had CMV infection, indicating the importance of screening for alternative conditions that may be causing graft injury while using dd-cfDNA as a non-invasive surveillance method in patients at a later time-point post-transplant. We did not find any significant association between the COVID-19 vaccination and GEP scores, which we predicted may be associated with elevated GEP scores due the immune response caused by mRNA vaccines. However, our cohort only included 17 patients vaccinated at variable times before the non-invasive testing.

The cost effectiveness of biomarkers assessing for acute rejection has been previously assessed in a simulation model comparing GEP to EMBx in the United States. 24 The model accounted for the probability of complication rates and incorporated published utility estimates and direct medical costs. This analysis suggested that GEP was an economically dominant strategy compared to EMBx with an average per patient cost-saving of $27,244 and QALY gain of 0.046 over the first 5 years. 24 While these cost-savings may be overestimated in a socialized healthcare system, our patient satisfaction survey reinforces the benefits by highlighting the reduced anxiety and fear related to avoiding a biopsy with a routine blood test. In previous studies, GEP can improve patient satisfaction without increasing the risk of adverse outcomes. 2, 3 Our patients had SF-12 PCS scores that were slightly below average and MCS scores slightly above average compared to the US population, and were similar when compared to SF-12 scores found in GEP studies for rejection surveillance. The most important limitation in this study is the lack of long-term follow up to ensure the safety of the reduction of IS based on non-invasive testing. Nevertheless, several studies have shown longer-term safety with a dd-cfDNA threshold of ≤ 0.15% and we do not anticipate harm. 8, 14, 19 Additionally, none of the patients managed by non-invasive testing had clinically significant rejection, allograft dysfunction, or hemodynamic deterioration within a median follow up time of 3 months. Since completion of the dd-cfDNA testing for this study, all patients have undergone further surveillance testing with repeat standard testing (biopsy or GEP). We are confident none of the patients have had any evidence of significant rejection (ACR >1R or any evidence of AMR). We also limited our patient population to those at least six months post-HT since this is our current local practice for non-invasive testing, however, GEP has validity evidence for use as early as 55 days post-HT and dd-cfDNA levels stabilize after 28 days post-HT. 2, 3, 8, 14 Future studies should include patients at all eligible time points post-transplant. This study was part of a quality improvement initiative, as such there was no randomization or control groups. In this type of design (open label, prospective time series), the goal was not to demonstrate that a given intervention increases or reduces risk of rejection, rather we were aiming to show that despite the limitations in access to surveillance endomyocardial biopsy we J o u r n a l P r e -p r o o f were able to adhere to our institutional surveillance protocol without experiencing excess shortterm morbidity.

Other limitations of this study include the small sample size and low incidence of allograft rejection. None of our patients developed significant ACR or AMR during the study period, so we were not able to interpret these results or follow trends in non-invasive testing in assessment of patients with rejection. Based on our institutional transplant protocol, we require surveillance biopsies in all transplant recipients who undergo initiation of mTOR inhibitor or a change in target levels. In our cohort, 45% of patients required a surveillance biopsy for significant changes in immunosuppression beyond 2 years after transplant, where a biopsy may not have been warranted. While it is unlikely that all EMBx will be fully eliminated in the shortterm, we showed that it is possible to use non-invasive surveillance to safely reduce the total burden of this procedure on local resources without increasing short-term adverse events during a global pandemic. Further prospective research to assess applicability of non-invasive rejection surveillance in a post-pandemic era is warranted.

In conclusion, this is the first Canadian study to assess the role for non-invasive rejection surveillance in personalizing titration of immunosuppression post-HT. Overall, non-invasive rejection surveillance was associated with increased satisfaction and reduced anxiety in HT recipients, while minimizing exposure for patients and providers during a global pandemic.

This study was supported by CareDx ® Inc, Brisbane, California with in-kind donations of HeartCare ® testing kits to provide dd-cfDNA and GEP testing for patients. No other disclosures. Note: Data is presented as median (interquartile range) and count (frequency) for continuous and categorical data, respectively. Graft dysfunction was defined as any history of decline in left ventricular ejection fraction of 15% or more from baseline. Abbreviations: dd-cfDNAdonor-derived cell-free DNA; GEPgene expression profiling. Abbreviations: dd-cfDNAdonor-derived cell-free DNA; EMBxendomyocardial biopsy; GEPgene expression profiling. 

The International Society of Heart and Lung Transplantation Guidelines for the care of heart transplant recipients

Randomized pilot trial of gene expression profiling versus heart biopsy in the first year after heart transplant: early invasive monitoring attenuation through gene expression trial

Gene-expression profiling for rejection surveillance after cardiac transplantation

Complications of endomyocardial biopsy in heart transplant patients

Remote Monitoring of Heart Transplant Recipients during the COVID-19 Pandemic

Impact of cytomegalovirus infection on gene expression profile in heart transplant recipients

Correlation of Longitudinal Gene-Expression Profiling (GEP) Score to Cytomegalovirus (CMV) Infection: Results from the Outcomes Allomap® Registry

Noninvasive detection of graft injury after heart transplant using donor-derived cell-free DNA: A prospective multicenter study

Combining Donor Derived Cell-Free DNA and Gene Expression Profiling for Non-Invasive Surveillance after Heart Transplantation

A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity

Qualitative Data Analysis: a methods sourcebook

Infectious complications after heart transplantation in patients screened with gene expression profiling

Risk factors for early development of cardiac allograft vasculopathy by intravascular ultrasound

Cell-Free DNA to Detect Heart Allograft Acute Rejection

Nodular Endocardial Infiltrates (Quilty Lesions) Cause Significant Variability in Diagnosis of ISHLT Grade 2 and 3A Rejection in Cardiac Allograft Recipients

Concordance among pathologists in the second Cardiac Allograft Rejection Gene Expression Observational Study (CARGO II)

Clinical usefulness of gene-expression profile to rule out acute rejection after heart transplantation: CARGO II

Risk evaluation using gene expression screening to monitor for acute cellular rejection in heart transplant recipients

Clinical utility of donor-derived cell-free DNA testing in cardiac transplantation

Donor fraction cell-free DNA and rejection in adult and pediatric heart transplantation

Donor-derived cell-free DNA is associated with cardiac allograft vasculopathy

Single-center utilization of donor-derived cell-free DNA testing in the management of heart transplant patients

Initiation of noninvasive surveillance for allograft rejection in heart transplant patients > 1 year after transplant

Cost-Effectiveness of a Blood-Based Biomarker Compared to Endomyocardial Biopsy for the Diagnosis of Acute Allograft Rejection