key: cord-0867475-llli25t2 authors: Murchu, Eamon O.; O'Neill, Sinead; Byrne, Paula; De Gascun, Cillian; O'Neill, Michelle; Ryan, Máirín; Harrington, Patricia title: Comparative genomic analysis demonstrates that true reinfection following SARS-CoV-2 infection is possible date: 2021-05-04 journal: Journal of clinical virology plus DOI: 10.1016/j.jcvp.2021.100015 sha: afb996603e5676ce1fa50d4dbeb85889f0a1cbf1 doc_id: 867475 cord_uid: llli25t2 Background: In recent months, multiple cases of confirmed SARS-CoV-2 reinfection have been reported. However, accurate epidemiological and virological data, including genomic analysis where possible, are required to differentiate cases of prolonged viral RNA shedding (i.e. intermittent detection) from true reinfection. The objective of this review was to systematically identify and summarise all cases of SARS-CoV-2 reinfection confirmed by comparative genomic analysis. Methods: A protocol based on Cochrane rapid review methodology was employed. Databases and pre-print servers were searched until 9/11/2020. Results: Ten studies, representing 17 patients, were identified (mean age=40; 71% male). The time interval between primary infection and reinfection ranged from 13 to 142 days (median: 60). Comparative whole genome sequencing confirmed reinfection in 14 patients (the primary and secondary infections were caused by different viruses). A further three cases had strong, but not confirmed evidence of reinfection, as only partial genomes were retrieved on primary infection. Across 12 studies that reported the number of single nucleotide polymorphisms (SNPs) comparing the first and second genomes, between 8 and 24 SNPs were discovered. With an average SARS-CoV-2 mutation acquisition rate of 1-2 per month, in all cases it is likely that the secondary infection was caused by a different SARS-CoV-2 virus, rather than prolonged shedding of viral RNA from the primary infection. In five reinfection cases, the primary and secondary infections were caused by different SARS-CoV-2 lineages/clades, strongly indicating that infections were caused by different viruses. Conclusion: Comparative genomic analyses from 14 patients confirm that SARS-CoV-2 reinfection can occur. In recent months, multiple cases of confirmed SARS-CoV-2 reinfection have been reported. However, accurate epidemiological and virological data, including genomic analysis where possible, are required to differentiate cases of prolonged viral RNA shedding (i.e. intermittent detection) from true reinfection. The objective of this review was to systematically identify and summarise all cases of SARS-CoV-2 reinfection confirmed by comparative genomic analysis. A protocol based on Cochrane rapid review methodology was employed. Databases and pre-print servers were searched until 9/11/2020. Ten studies, representing 17 patients, were identified (mean age=40; 71% male). The time interval between primary infection and reinfection ranged from 13 to 142 days (median: 60). Comparative whole genome sequencing confirmed reinfection in 14 patients (the primary and secondary infections were caused by different viruses). A further three cases had strong, but not confirmed evidence of reinfection, as only partial genomes were retrieved on primary infection. Across 12 studies that reported the number of single nucleotide polymorphisms (SNPs) comparing the first and second genomes, between 8 and 24 SNPs were discovered. With an average SARS-CoV-2 mutation acquisition rate of 1-2 per month, in all cases it is likely that the secondary infection was caused by a different SARS-CoV-2 virus, rather than prolonged shedding of viral RNA from the primary infection. In five reinfection cases, the primary and secondary infections were caused by different SARS-CoV-2 lineages/clades, strongly indicating that infections were caused by different viruses. Comparative genomic analyses from 14 patients confirm that SARS-CoV-2 reinfection can occur. Background Accurate epidemiological and virological data, including genomic analysis where possible, are required to differentiate cases of prolonged viral RNA shedding from true SARS-CoV-2 reinfection. In previous reviews conducted by our team at the Health Information and Quality Authority (HIQA), no true cases of reinfection were identified 1 . In recent months, however, multiple cases of confirmed reinfection have been reported. We therefore conducted a review of the literature to characterise all true cases of reinfection based on comparative genomic analysis. A standardised protocol was employed 2 , based on Cochrane rapid review methodology guidance. Electronic databases (PubMed, EMBASE and EuropePMC) and pre-print servers (medRxiv, bioRxiv) were searched until 9/11/2020. Database searches retrieved 3,272 unique citations. Following screening and full text review (in duplicate), 10 studies, representing 17 patients, met our inclusion criteria (Table 1) Whole genome sequencing (WGS) confirmed that primary and secondary infections were caused by different viruses in 14 cases, although the degree of separation between infections varied (Table 1) . Three apparent reinfections could not be confirmed by sequencing due to insufficient genetic material extracted from the primary infection (partial genomes). However, genetic evidence consistent with reinfection was still present. With the exception of one study 9 , all confirmed cases included a quantification of the number of single nucleotide polymorphisms (SNPs) comparing primary and secondary infections (range: eight 11 to 24 7 differences, Table 1 ). In addition, five studies presented stronger evidence of reinfection through phylogenetic analysis; in each case the primary and secondary infections belonged to different SARS-CoV-2 lineages or clades 5, 7, 8, 10, 11 . Mutations that result in the D614G amino acid change in the spike protein were present on reinfection in four studies 7, 9, 10, 12 . Additionally, rare mutations leading to an amino acid (AA) change in a single viral lineage were identified in some studies, such as NSP6 and L142F 7 . In terms of antibody testing, only two studies performed IgG testing at both infection events. In the first case, anti-SARS-CoV-2 IgG was not detected at primary infection, or at four days post-symptom onset; however, IgG was detected at reinfection and 30 days post-symptom onset 5 . The timing of testing may have impacted the findings, however, as the first sample may have been taken prior to seroconversion taking place. In the second case, anti-SARS-CoV-2 IgG was not detected at the initial infection, or at 10 days post-symptom onset 7 . However, seroconversion occurred following the reinfection event, with IgG not detected at serial testing on days 1-3 post-hospitalisation, but detected on day five post-hospitalisation. No study investigated neutralising antibody profiles or cellmediated immunity. Only one study assessed the rate of reinfection 9 . In this study, potential reinfection cases were identified among a larger cohort of 133,266 laboratory-confirmed SARS-CoV-2 infections. The authors estimated the risk of reinfection to be 0.01% (95% CI: 0.01-0.02%). While clinical and epidemiological factors are important in the assessment of possible SARS-CoV-2 reinfections, comparative genomic analysis provides the best evidence. This review identified 14 individual patients with confirmed SARS-CoV-2 reinfection. Strong evidence of reinfection exists if sequences recovered from the two infectious events belong to different genetic clades or lineages 13 . However, even if viral strains are from the same clade or lineage, differences in the number of SNPs may indicate different viruses. The virus is known to mutate by 1-2 nucleotides per month 14 . Therefore, if infecting viruses differ by more mutations than this over a given timeframe, it increases the likelihood that the infections are from genuinely different origins, rather than a virus that has evolved in the setting of a persistent infection. All reinfections in this review recorded a greater number of mutations than would have been expected to occur through natural viral evolution. The analysis of specific genetic mutations provides additional insight into reinfection events, such as mutations that result in the D614G AA change in spike protein. The D614G AA change was found on reinfection in four studies 7, 9, 10, 12 . This AA change defines the SARS-CoV-2 variant with greater replicative fitness 15 and is now present in most circulating SARS-CoV-2 lineages 7 . Another genetic variation, 22882T>G (S:N440K) within the receptor-binding domain of the spike protein which possibly confers resistance to neutralising antibodies, was detected in one study 3 . The presence of rare mutations, based on published sequence data, strengthens the case for reinfection. In the study by To et al. 7 , the secondary genome also contained the mutation NSP6 L142F, which was only rarely reported (only 0.009% of genomes deposited into GISAID contained this mutation on 20/8/2020 16 ). One of the primary limitations of included studies was their inability to definitively exclude falsepositive RT-PCR results, either at primary or secondary infection events. While RT-PCR testing is the operational gold standard for SARS-CoV-2 diagnosis 17 , technical problems at some point during the testing process may result in false-positives. Indeed, the problem of contamination can be a real concern; for example, the CDC in the US had to withdraw SARS-CoV-2 RT-PCR testing kits in March 2020 after a high rate of false-positives due to reagent contamination 18 . Nonetheless, the fact that SARS-CoV-2 WGS requires a given sample to be processed -typically in separate pathways -from the primary sample at least twice, the risk of false positive results, if not contamination, is significantly reduced if not eliminated. Methods to exclude false-positive RT-PCR results across studies included excluding RT-PCR results with high Ct values and/or using cut-offs for the interval between infection events. All patients had a minimum interval of 45 days, except one patient who developed symptoms just 13 days after initial recovery 11 . Study authors suggested that this patient's primary infection may have occurred days to weeks before his first positive RT-PCR test, as he was asymptomatic. However, the risk of contamination must be considered, especially when clinical or epidemiological data suggest otherwise. Another limitation of our findings is the inability to calculate a population-level reinfection rate due to the extremely low number of confirmed reinfections identified. As confirmation of reinfection necessitates WGS of both events, our findings may represent a significant underestimation of all reinfections. Nonetheless, these data suggest that reinfections can occur, but are a rare phenomenon, suggesting strong protective immunity following primary infection. The phenomenon of SARS-CoV-2 reinfection has significant policy implications and suggests that immunity following primary SARS-CoV-2 infection is not universal. Although there were no documented cases of onward transmission from these reinfected cases, knowledge is evolving. Infection prevention and control, isolation and contact tracing considerations are not likely to differ for the reinfections compared with the primary infection. This research was funded in part by the Health Research Board under grant no. HRB-CICER-2016-1871. ☒ The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Immune response following SARS-CoV-2 infection Protocol: Evidence summary of the duration of immunity and reinfection following SARS-CoV-2 infection Asymptomatic reinfection in two healthcare workers from India with genetically distinct SARS-CoV-2. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America Reinfection of SARS-CoV-2 in an immunocompromised patient: a case report COVID-19 Re-Infection by a Phylogenetically Distinct SARS-CoV-2 Variant, First Confirmed Event in South America Genomic Evidence for a Case of Reinfection with SARS-CoV-2 COVID-19 re-infection by a phylogenetically distinct SARScoronavirus-2 strain confirmed by whole genome sequencing. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America Clinical infectious diseases : an official publication of the Infectious Diseases Society of America Assessment of the Risk of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Reinfection in an Intense Reexposure Setting Reinfection with SARS-CoV-2 and Failure of Humoral Immunity: a case report Whole Genome Sequencing Confirmed SARS-CoV-2 Reinfections Among Healthcare Workers in India with Increased Severity in the Second Episode A Case of Early Re-infection with SARS-CoV-2. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America A dynamic nomenclature proposal for SARS-CoV-2 lineages to assist genomic epidemiology. Nature Microbiology. 2020. 14. New and Emerging Respiratory Virus Threats Advisory Group (NERVTAG) The Impact of Mutations in SARS-CoV-2 Spike on Viral Infectivity and Antigenicity False-positive COVID-19 results: hidden problems and costs. The Lancet Respiratory Medicine Labs Were Contaminated, Delaying Coronavirus Testing, Officials Say. Sheila Kaplan The authors would like to thank Executive Assistant Debra Spillane (HIQA). None.