Evidence of SARS-CoV-2 re-infection with a different genotype.

Dear Editor,

We read with great interest in this journal the description by Tomassini et al. of six possible cases of re-infection with SARS-CoV-2 in England. The characterization and extent of such re-infections are currently increasingly investigated, and their implications are a growing concern. Indeed, the emergence of SARS-CoV-2 in December 2019 in China was followed by the worldwide spread of the virus and its circulation for several months (https://coronavirus.jhu.edu/map.html). In several European countries, including France, the outbreak almost ended during spring, but a second COVID-19 outbreak occurred in late summer (https://covid19-country-overviews.ecdc.europa.eu/). We observed such an evolution of SARS-CoV-2 diagnoses at the Méditerranée Infection Institute in Marseille, France, where we have performed more than 300,000 SARS-CoV-2 qPCR since end of January 2020 and have detected the first infection at the end of February (https://www.mediterranee-infection.com/covid-19/). As the SARS-CoV-2 pandemic is still on-going, a current major issue is whether or not and how long immune responses to the virus are protective. In this regard, it is important to prove the cases of reinfection, which were first reported in August in Hong Kong. Tomassini et al. defined re-infection as qPCR positivity at least 28 days after a previous qPCR-positive Covid-19 episode that was followed by clinical recovery and at least one negative qPCR. We report here a patient with two infections at a 105 days interval despite seroconversion. In addition to Tomassini et al.’ criteria for re-infection, we demonstrated by genotypic analyses that the two successive infections involved distinct viral variants and that samples tested were collected from the same individual.

The patient is a 70-year-old immunocompetent man living in a retirement home due to behavioral and memory disorders. On April 22nd, 2020, he developed fever and cough. His oxygen saturation was 95%. SARS-CoV-2 PCR performed on a nasopharyngeal swab was positive (cycle threshold value (Ct)= 27). A low-dose chest CT-scan highlighted minimal ground glass images in both lungs. The patient subsequently fully recovered and further nasopharyngeal samples, collected on May 8th, 14th and 18th, were PCR-negative. Serological testing performed by a chemiluminescent immunoassay (CLIA) on a Liaison DiaSorin XL instrument (DiaSorin Inc., Saluggia, Italy) showed IgG seroconversion. Indeed, a serum sample collected on May 5th, two weeks after the onset of clinical symptoms and PCR diagnosis was IgG-negative, whereas a serum sample collected two weeks later, on May 18th, was IgG-positive (signal= 21; positivity threshold= 15). On August 19th, the patient was tested PCR-positive again (Ct= 18), when sampled during a systematic screening performed in his retirement home while he was asymptomatic.

Next-generation sequencing (NGS) of SARS-CoV-2 genomes was carried out using Illumina (San Diego, CA, USA) technology as previously described. Genome consensus sequences were generated with the CLC Genomics workbench v.7 (https://digitalinsights.qiagen.com/) by mapping NGS reads on the Wuhan-Hu-1 SARS-CoV-2 genome (GenBank accession no. NC_045512) with 0.8 and 0.9 as coverage and similarity thresholds, respectively. The genome sequence (20,879 non-contiguous nucleotides; IHU-3844/2020) from April 22nd was most closely related to those from strains of Nextrain clade 20A that circulated during the first outbreak in our geographical area (Fig. 1 ). The SARS-CoV-2 genome (deposited in the GISAID database (https://www.gisaid.org/) with no. France/PAC-IHU-1347/2020) from August 19th belonged to the Marseille 4 lineage that emerged in our geographical area during the second outbreak (Fig. 1), and 11 mutations that are hallmarks of the Marseille 4 lineage (C4543U, G5629U, G9526U, C11497U, G13993U, G15766U, A16889G, G17019U, G22992A, G28975C, G29399A) were absent from the genome obtained from the first sample. In contrast, 2 mutations (C2416U, G8371U) that are hallmarks of the genotype identified in the first sample were absent in the second genome (Supplementary Table S1). In order to prove that samples were from the same patient, we confirmed genetically that each of 24 independent short tandem repeat markers analyzed (Supplementary Material) identified identical alleles.

Fig. 1

Phylogeny reconstruction based on SARS-CoV-2 genomes recovered during the first and second infections. Phylogenetic tree was reconstructed using the MEGA X software (https://www.megasoftware.net/) based on SARS-CoV-2 genome sequences, with a total of 29,703 positions in the final dataset. This analysis incorporated the genome sequences the most similar through BLASTn searches (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch) to the two genome sequences recovered from the case-patient in April and August 2020 (indicated by a black background and a white bold font and a framed white background and a black bold font, respectively) among those obtained in our center from respiratory samples collected since end of February 2020 until end of September (indicated by a gray bold font) and those from the GISAID database (https://www.gisaid.org/) (indicated by a black bold font). Among top hit sequences from the GISAID database, a single one was kept by country. The evolutionary history was inferred by using the Maximum Likelihood method and Kimura 2-parameter model. The percentage of trees in which the associated taxa clustered together is shown next to the branches. Bootstrap greater than 50% are indicated in the tree. The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. All nucleotide positions with less than 80% site coverage were discarded (partial deletion option). Prior nucleotide sequence alignment was performed using Muscle.

Here, we demonstrate that the same patient was infected in April, cleared the virus, seroconverted, but was re-infected four months later with a new viral variant. The two infections reflect the circulating strains in Marseille at the same time. It is the most comprehensive studied as it documented seroconversion following the first infection, showed drastically different viral genomes with 34 nucleotide differences, and ruled out errors of samples by techniques commonly used for forensic identifications. The present case adds to 13 previously reported cases of re-infection with a different SARS-CoV-2 strain that occurred in China, Belgium, the Netherlands, India, Ecuador and the USA , 5, 6, 7, 8 – (Supplementary Table S2) documented with varying degrees of robustness (Supplementary Table S3). Mean age (± standard deviation) of the cases was 40±20 years (range, 24–89), and patients were mostly immunocompetent individuals (in 12 cases (86%)). The 14 reports involved men in 9 of 13 documented cases (69%). The mean delay between the two diagnoses was 81±36 days (19–142). The symptomatology of the first and second infections was much variable. In eight cases, symptoms were reported in both infections, re-infection being less severe in two cases and more severe in four, including one death. In two cases, both infections were asymptomatic, in two only re-infection was asymptomatic, and in two only re-infection was symptomatic. Serology was performed in three cases following the first infection and was positive. Serology was performed in 11 cases following the second infection and was negative in four and positive in seven.

Such early re-infections with SARS-CoV-2 is surprising, as we are used with a majority of respiratory viruses to observe a single, annual epidemic episode. This atypical epidemiological pattern is particularly relevant in our geographical area where the second outbreak that started during the summer was linked to multiple distinct variants having accumulated mutations that differed from viral mutants that circulated during the first outbreak. This deserves conducting further studies to figure out whether or not this would make sense to include several viral variants in future vaccines.

Declaration of Competing Interest

The authors have no conflict of interest to declare. Funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.

Ethics

The study was approved by the ethical committee of the University Hospital Institute Méditerranée Infection (N°: 2020-029). Access to the patients’ biological and registry data issued from the hospital information system was approved by the data protection committee of Assistance Publique-Hôpitaux de Marseille (APHM) and was recorded in the European General Data Protection Regulation registry under number RGPD/APHM 2019-73.