key: cord-1031428-pyrqo1oa authors: FILLATRE, P.; DUFOUR, M. J.; BEHILLIL, S.; VATAN, R.; REUSSE, P.; GABELLEC, A.; VELMANS, N.; MONTAGNE, C.; GEFFROY, S.; DROUMAGUET, E.; MEROUR, V.; Enouf, V.; BUZELE, R.; VALENCE, M.; GUILLOTEL, E.; GAGNIERE, B.; BAIDALUK, A.; ZHUKOVA, A.; TOURDJMAN, M.; THIBAULT, V.; GROLHIER, C.; PRONIER, C.; LESCURE, X.; SIMON-LORIERE, E.; COSTAGLIOLA, D.; VAN DER WERF, S.; TATTEVIN, P.; MASSART, N. title: A new SARS-CoV-2 variant poorly detected by RT-PCR on nasopharyngeal samples, with high lethality date: 2021-05-10 journal: nan DOI: 10.1101/2021.05.05.21256690 sha: ca55a6fa0fd83908f155d64849e2d4813d1ac7fe doc_id: 1031428 cord_uid: pyrqo1oa Background: In early January 2021, an outbreak of nosocomial cases of COVID 19 emerged in Western France, with RT PCR tests repeatedly negative on nasopharyngeal samples but positive on lower respiratory tract samples. Whole genome sequencing (WGS) revealed a new variant, currently defining a novel SARS CoV 2 lineage: B.1.616. In March, WHO classified this variant as "under investigation" (VUI). We analyzed the characteristics and outcomes of COVID 19 cases related to this new variant. Methods: Clinical, virological, and radiological data were retrospectively collected from medical charts in the two hospitals involved. We enrolled patients with at least one of the following: i) positive SARS CoV 2 RT PCR on a respiratory sample; ii) seroconversion with anti SARS CoV 2 IgG/IgM; iii) suggestive symptoms and typical features of COVID 19 on chest CT scan. Cases were categorized as either: i) B.1.616; ii) variant of concern (VOC); iii) unknown. Findings: From January 1st to March 24th, 2021, 114 patients fulfilled the inclusion criteria: B.1.616 (n=34), VOC (n=32), and unknown (n=48). B.1.616 related cases were older than VOC related cases (81 years [73-88], vs 73 years [67-82], P<0.05) and their first RT PCR tests were less often positive (5/34, 15% vs 31/32, 97%, P<0.05). The B.1.616 variant was independently associated with severe disease (multivariable Cox model HR 4.2 [1.3 , 13.5], P=0.018), and increased lethality (logrank test P=0.01): 28day mortality 15/34 (44%) with B.1.616, vs. 5/32 (16%) for VOC, P=0.036. Interpretation: We report a nosocomial outbreak of COVID-19 cases related to a new variant, B.1.616, poorly detected by RT PCR on nasopharyngeal samples, with high lethality. Funding: None SW lab is funded by Institut Pasteur, CNRS, Université de Paris, Santé publique France, Labex IBEID (ANR-10-LABX-62-IBEID), REACTing (Research & Action Emerging Infectious Diseases) , and by the H2020 project 101003589 (RECOVER). Conflicts of interest: SW has patents issued and pending for SARS-associated coronavirus diagnostics. All other authors declare no competing interests Ethics approval: This study respects the French reference method MR004. A complete de-identified patient dataset, accompanied by the original study protocol will be available to European researchers on request. Individuals wishing to access the data should send a request to pierre.fillatre@armorsante.bzh. Evidence before this study Among the numerous SARS-CoV-2 variants described worldwide, only 3 are currently classified as Variant of Concern (VOC) by the WHO, since they are associated with either an increased risk in transmissibility, severity, or significant reduction in neutralization by antibodies: Our observational study, conducted from January 1 Of note, the diagnostic performance of RT-PCR tests was satisfactory on lower respiratory tract samples, suggesting that failure to detect B.1.616 on NP samples would be due to a viral load below the limit of detection in the upper respiratory tract, rather than to genomic mismatches between routine RT-PCR targets and this variant. In our cohort, B.1.616 was independently associated with worse clinical outcome, with high 28-day mortality (44%). Diagnosis of B.1.616-related COVID-19 cases should not rely on RT-PCR tests on NP samples. In the epidemic area, strict infection control measures must be maintained as long as COVID-19 diagnosis is not ruled out, in order to limit nosocomial clusters and case fatality. Further studies are needed to confirm and investigate the association between genomic characteristics of B.1.616, and i) poor detection by RT-PCR tests on NP samples; ii) prognosis. Since the emergence of COVID-19, numerous SARS-CoV-2 variants have been described 1 . At the end of 2020, novel concerns were raised with the detection of rapidly spreading variants of concern (VOCs) in distinct regions of the world. We conducted a retrospective observational study in two hospitals of the We screened medical files of all patients who were diagnosed with SARS-CoV-2 infection at the Lannion and Saint-Brieuc hospitals from January 1 st to March 24 th , 2021. SARS-CoV-2 infection was defined by at least one of the All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 following: i) positive SARS-CoV-2 RT-PCR on a respiratory sample; ii) seroconversion based on paired sera tested for anti-SARS-CoV-2 IgG/IgM; iii) suggestive symptoms and typical features of COVID-19 on chest CT scan. COVID-19 cases were categorized in one of the three following groups: i) B. The VOC group included all COVID-19 cases due to VOCs B.1.1.7, B.1.351 and P.1. Routine screening for these VOCs has been performed by a prescreening RT-PCR in all patients with positive RT-PCR since February 10 th , 2021 in the participating hospitals. COVID-19 patients not fulfilling the criteria for the first two groups were categorized in the 'unknown' group. These patients were included in the study for clinical, biological, virological, and radiological description. However, since it was not possible to assign them to either the B.1.616 group, or the VOC group, they were not included in the primary analysis. Cases were categorized as nosocomial COVID-19 if symptoms appeared at least 2 days after hospital admission 7 . All patients with COVID-19 confirmation or suspicion, based on clinical and/or radiological findings, were isolated with droplet transmission precautions for at least 10 days, including 48 h after clinical improvement and apyrexia. Patients who required ICU management were isolated during 14 days, and immunocompromised patients were isolated during 24 days, including 48 h after clinical improvement and apyrexia. Infection control measures included admission to a single room, wearing gown and gloves, protection glasses, surgical mask (if treated with O 2 flow <6 L/min) or FFP2 mask (if treated with O 2 flow >6 L/min) and reinforced hand hygiene. All COVID-19 patients received prophylactic therapy with heparin, according to national guidelines, and those who required treatment with O 2 received dexamethasone, 6 mg daily for 10 days or until they no longer required O 2. Tocilizumab could be prescribed at the discretion of the physicians in charge. Different commercially available RT-PCR assay were used at Lannion and Saint-Brieuc hospitals: Xpert® Xpress SARS-CoV-2, GeneXpert Cepheid®, BD SARS-CoV-2 reagents for BD Max, BD®, Viasure SARS-CoV-2 Real Time PCR Detection, Certest®. Screening for VOCs was performed on positive RT-PCR samples using ID SARS-CoV-2/UK/SA All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.05.21256690 doi: medRxiv preprint Variant Triplex, ID Solutions®. When B.1.616-related COVID-19 was suspected, samples were sent to the National Reference Center (Institut Pasteur, Paris) for WGS. Whole Genome Sequencing RNA extracted from clinical samples was used for SARS-CoV-2 WGS using a highly multiplexed PCR amplicon approach 8 with the ARTIC Network multiplex PCR primers set v3 (https://artic.network/ncov-2019). Synthesized cDNA was used as template and amplicons were generated using two pooled primer mixtures for 35 rounds of amplification. Libraries were prepared using the Nextera XT DNA Library Prep Kit (Illumina) and sequenced on an Illumina NextSeq500 (2x150 cycles) on the Mutualized Platform for Microbiology (P2M) at Institut Pasteur. Illumina adaptors, low quality reads, primer sequences were trimmed with Trimmomatic v0.36 9 and reads were assembled with Megahit 10 as well as mapped onto reference genome Wuhan/Hu1/2019 (NCBI Nucleotide -NC_045512, GenBank -MN908947) using the CLC Genomics Suite v5.1.0 (QIAGEN). SAMtools v1.3 11 was used to sort the aligned bam files and generate alignment statistics. Aligned reads were manually inspected using Geneious prime v2020.1.2 (2020) (https://www.geneious.com/), and consensus sequences were generated (> 5X read-depth coverage for a base call). SARS-CoV-2 sequences generated during this study (n=6) were combined to a global dataset of sequences available on GISAID to produce a subsampled phylogenetic tree (n=3238) with augur and auspice as implemented in the Nextstrain pipeline version (April, 2021) (https://github.com/nextstrain/ncov). Primary outcome was clinical severity defined as a score >5 in the WHO clinical progression scale 12 . Secondary outcomes were ICU admission and 28 day-mortality. Statistical analysis was performed with R 3.4.3. Categorical variables were expressed as numbers (percent) (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.05.21256690 doi: medRxiv preprint using the Cox proportional-hazard regression model. Variables associated with outcome with a P-value <0.2 in univariate analysis or those considered clinically relevant were included in the multivariate analysis. All tests were two-sided, and a P-value ≤ 0.05 was considered statistically significant. Patients or closest relatives were informed of the retrospective collection of data and could refuse to participate. The French infectious diseases society ethics committee (Comité d'Éthique de la Recherche en Maladies Infectieuses et Tropicales, CER-MIT) approved the study (N° COVID 2021-06). Written informed consent was waived. Phylogenetic analysis revealed an original variant carrying a unique constellation of mutations, which received the B.1.616 Pango lineage designation 1 (Figure 1 ). It is characterized by 9 amino acid changes and one deletion in the S protein in comparison with the original Wuhan strain, several unique amino-acid changes in the E, M, and N proteins, in ORF1ab and ORF3 as well as by a deletion and frameshift in ORF6 and replacement of the stop codon of ORF7a resulting in a 5 amino acids extension at its C-terminus (Table 1) . Interestingly, some mutations (Y144-, H655Y) in the S protein have been observed for previously described VOCs, i.e B.1.1.7 and P.1, respectively. The V483A is located in the receptor binding motif next to residue 484, for which the E484K change found in several VOCs has been associated with reduced neutralization by post-infection and post-vaccination antibodies [13] [14] [15] . Also of note are the mutations in ORF6 and ORF7a, two proteins that antagonize various steps of type I interferon (IFN-I) production and signaling [16] [17] [18] . cases. In addition, 21 patients developed COVID-19 not related to a VOC during or within 14 days of close contact with at least one patient with active COVID-19 classified as a confirmed B.1.616 case. Although no sample was available for All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; WGS or PCR screening for these patients, they were classified as probable B.1.616 cases, according to our predefined criteria. Therefore, these 34 patients (13 confirmed and 21 probable) constituted the B.1.616 group. During the study period, 108 patients were hospitalized in a ward where at least one case of B.1.616 COVID-19 was admitted, for a total of 780 patients-days at risk. Of these 108 exposed patients, 32 (30%) developed COVID-19 symptoms after they were admitted for another reason and were categorized as nosocomial B.1.616 cases (11 confirmed B.1.616, and 21 probable). Therefore, the B.1.616 incidence rate in these wards was estimated at 41/1000 patient-days at risk. In addition, 52/86 patients living in the epidemic area were hospitalized because of COVID-19, in the absence of any contact with a B.1.616 case, among which 4 were infected by a VOC (B. Table 1 ). In addition, the median cycle threshold (Ct) in nasopharyngeal sample was higher in the B. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.05.21256690 doi: medRxiv preprint VOC cases (logrank test P=0.01), Figure 3 , although this was not significant, after age adjustment ], P=0.057). The low rate of SARS-CoV-2 detection by RT-PCR tests on nasopharyngeal samples within a large cluster of COVID-19 cases led to the identification of a novel variant, B. COVID-19 cases with at least one positive RT-PCR tested negative for RT-PCR on their first nasopharyngeal sample. Failure to detect B.1.616-related COVID-19 with the gold-standard diagnostic test most likely contributed to the emergence of several clusters, since implementation of specific infection control measures mostly relied on virological confirmation during the study period. Retrospective review of medical charts found that, in several cases, the diagnosis All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.05.21256690 doi: medRxiv preprint was suspected early in the COVID-19 course, but specific infection control measures were relieved once RT-PCR tests returned negative. Failure to detect B.1.616 on nasopharyngeal samples is even more problematic for the screening of contacts. Indeed, the incidence of nosocomial COVID-19 was much higher than reported 7 during the early phase of the outbreak, before we realized the low proportion of nasopharyngeal samples from which this specific circulating variant could be detected. Of note, although 'diagnostic detection failures' is one of the criteria to define VOC, none of the three major VOCs nor any of the current VUI meet this classification criteria. As RT-PCR assays used in France target at least two different viral genomic regions, it is unlikely that the B.1.616 strain would not be detected due to specific mismatches. Identification of this new lineage in lower respiratory tract samples confirmed the correct detection of the viral genome by commercially available assays used during the study period. Repeated failures to detect B.1.616 on nasopharyngeal samples were therefore most likely due to SARS-CoV-2 viral loads below detection threshold in this site, rather than sub-optimal sensitivity of routine RT-PCR tests. In our study, among patients with a positive RT-PCR assay in the B.1.616 group, the sensitivity of one, two, three, and four tests on nasopharyngeal samples were, respectively, 5/34 (15%), 13/34 (38%), 14/34 (45%), and 17/34 (55%). RT-All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.05.21256690 doi: medRxiv preprint PCR tests on sputum, or broncho-alveolar lavage (BAL), were positive in 8/34 (24%) B.1.616-related COVID-19 cases with previous negative nasopharyngeal RT-PCR tests. As samples from the lower respiratory tract are more difficult to obtain in frail patients, the real extent of the B.1.616-related COVID-19 outbreak in our institution has probably been underestimated. A large surveillance study, with sequencing of a representative sample of 15% of all RT-PCR-positive COVID-19 cases during the study period found no community-acquired B.1.616-related COVID-19 (Flash study#5, SpF, Paris, France, unpublished data), but the low detection in standard sampling may have contributed to this result. Our study has limitations. First, the small sample size and the retrospective design both limit the statistical power. However, due to the fast pace of the pandemic, early communication on the characteristics of new variants is warranted, and this would be the first case series of B.1.616-related COVID-19 cases. Second, B.1.616 confirmed cases where those in whom a deep respiratory sample was obtained, mostly motivated by disease severity, hence constituting a selection bias. . Finally, the selection of controls may be an additional limitation: VOC-related cases were selected as controls mostly because these cases could be reliably classified as 'non-B.1.616': indeed, other COVID-19 cases managed during the study period were no longer available for virological characterization, hence we could not rule out that they could be B.1.616. Of note, our VOC control group was mostly constituted by the B.1.1.7 variant, which is largely dominant in Western Europe, and has been associated with an increased mortality 4 . Hence, as B.1.616 cases were independently associated with worse outcomes as compared to VOC, this would probably be even more pronounced if compared with COVID-19 cases non-related to VOC. In addition, all consecutive cases with available VOC screening were enrolled in the control group, which limits selection bias. We report a nosocomial outbreak of COVID-19 cases related to a new variant, B.1.616, characterized by poor detection by RT-PCR tests on nasopharyngeal samples despite typical clinical, radiological, and biological features of COVID-19. We also noted high case fatality rate in our sampled population. The novel variant reported here adds to the diversity of emerging SARS-CoV-2 variants with impact on early diagnosis and control. Further investigations are required to confirm those observations. This work also highlights the difficulties to manage nosocomial cases when the gold-standard test fails to confirm the diagnosis. With constantly emerging new variants, one should remain attentive to any unusual clinical situation that could be linked to such emergence. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. i c k m a n H M , R a m p l i n g T , S h a w K , e t a l . N o s o c o m i a l T r a n s m i s s i o n o f C o r o n a v i r u s D i s e a s e 2 0 1 9 : A R e t r o s p e c t i v e S t u d y o f 6 6 H o s p i t a l -a c q u i r e d C a s e s i n a L o n d o n T e a c h i n All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.05.21256690 doi: medRxiv preprint N T325I § changes relative to the reference Wuhan-Hu-1 strain (NC_045512) with changes to the most recent common ancestor (MRCA) highlighted in bold. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Note. WHO, world health organization; VOC, variant of concern All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; Note. VOC, variant of concern All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.05.21256690 doi: medRxiv preprint Figure 1 : Phylogenetic analysis of the B.1.616 lineage and characteristic non-synonymous substitutions. A. Subsam global phylogenetic maximum likelihood tree of SARS-CoV-2 with annotated Nextstrain clades next to the correspo nodes and tips highlighted only for sequences from the Pangolin B.1.616 lineage. B. Detailed view of the B1.616 lin A-B. Branch lengths correspond to the number of nucleotide substitutions (shown below the tree) from the reference Wuhan-Hu-1 strain (NC_045512). C. Nucleotide and amino acid substitutions from the reference Wuhan-Hu-1 strain shared among the sequences from Lannion, France are represented as ticks along the SARS-CoV-2 genome and are annotated with text if non-synonymous. Light grey text annotated amino acid substitutions are not unique to the Lan (B1.616) lineage. mpled ponding lineage. ce rain re annion All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Note. VOC, variant of concern All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted May 10, 2021. ; https://doi.org/10.1101/2021.05.05.21256690 doi: medRxiv preprint Serum C-reactive protein, mg/mL 81 Acknowledgments:We thank Olivia DA CONCEICAO and Cylia IMEKHLAF for their help in collecting data from patients charts.We would like to thank all of the healthcare workers, public health employees, and scientists involved in the COVID-19 response and this outbreak.We acknowledge the authors, originating and submitting laboratories of the sequences from GISAID (Table S2) . We avoided any direct analysis of genomic data not submitted as part of this paper and used this genomic data only as background. This work used the computational and storage services (Maestro cluster) provided by the IT department at Institut Pasteur, Paris.