Skip to main content

Genetic Evidence and Host Immune Response in Persons Reinfected with SARS-CoV-2, Brazil

Our take —

This study assessed a small cluster of 4 individuals in Brazil who tested positive for COVID-19 during 2 distinct periods in March and May 2020. Plasma, serum, and nasopharyngeal swab samples were collected at baseline and then biweekly or at longer intervals based on the patient availability. In March, Patients A, B, and D reported mild symptoms and tested positive for SARS-CoV-2, and all patients (A, B, C, D) tested negative in April. Anti–SARS-CoV-2 neutralizing antibody response was not detected after the first episode in late March. In May, all four patients reported signs and symptoms, tested positive for SARS-CoV-2, and had higher viral loads than during the first infection. Whole genome sequences were generated from both time points from patients B and C. Patient B was infected by two distinct SARS-CoV-2 lineages, suggesting reinfection. The authors suggest that Patient C was also a reinfection, but without more detailed information on the reference sequences used in the phylogenetic tree analysis, the evidence they provide is not conclusive. This suggests that at least in one instance, humoral immunity was not developed at the first infection as neutralizing anti–SARS-CoV-2 titers were not detected following the first episode, nor at the baseline of the second episode. The results suggest that patients were still vulnerable after the primary infection and that they may have experienced reinfection with SARS-CoV-2 in a short timeframe.

Study design

Prospective Cohort

Study population and setting

This study screened 30 individuals weekly for COVID-19 from March to December 2020 in Brazil for recruitment into a longitudinal study. Individuals were tested for SARS-CoV-2 using RT-PCR in nasopharyngeal swab specimens, and if they tested positive, they were invited to participate in the study. Plasma, serum, and nasopharyngeal swab samples were collected at baseline and then biweekly or at longer intervals based on the patient availability. From this study, a cluster of 4 individuals were identified who tested positive for COVID-19 during 2 distinct periods in March and May 2020. Five age matched controls (negative for SARS-Cov-2) were selected from those screened for the study, and from the same city as the cases (Rio de Janeiro) for comparison to the 4 identified patients. Recruited patients were tested for presence and neutralization of SARS-CoV-2 antibodies as well as for markers indicating the induction of the host immune response to infection. When possible, whole genome sequences were generated from both March and May specimens.

Summary of Main Findings

Four screened individuals tested positive for COVID-19 during two distinct periods in March and May 2020. Patients A and B were close contacts, and Patients C and D shared a household with Patient B. All patients tested positive for SARS-CoV-2 in late March (despite an initial negative test from Patient D) and showed increased indicators of immune response compared to negative controls. The cytokine responses among patients were consistent with a resolved infection, however neutralizing antibody responses were not detected. All patients tested negative for SARS-CoV-2 via RT-PCR in April. In May, all 4 patients reported signs and symptoms and had higher viral loads than observed during the first infection. In Patients A and B, SARS-CoV-2 immunoglobulins were observed one week after the second infection, however, there was low to no neutralizing activity. Full genome sequence were obtained from patients B and C in March and indicated infections with viruses belonging to clades 19A and 20B, respectively. Viral sequences were obtained from all participants in May, and all sequences were closely related members of clade 20B. In July, tests for all four patients continuously showed upregulated pro-inflammatory markers.

Study Strengths

This study leveraged many serological, molecular, and genomic tests to evaluate a cluster of individuals who experienced a suspected second SARS-CoV-2 infection in a short timeframe. Given the investigators were conducting regular follow up visits with individuals who had experienced a SARS-Cov-2 infection, they were able to detect and evaluate a second infection among those individuals. This study provides evidence for reinfection of one individual in the study (Patient B).


Although the primary purpose of this study is to understand potential reinfection among 4 individuals identified through a larger study, little information is provided on the process for recruitment of individuals. Participants were followed for regular visits after the first SARS-Cov-2 identified infection through repeat testing. Therefore, it is possible that individuals were tested at different time points in the course of their infection between episodes 1 and 2. The comparison of viral load levels between episode 1 and 2 for patients may be subject to bias if these were not taken at the same time on the course of infection. This article did not discuss other potential options which may influence limited antibody response after the first infection, such as if this may be related the specific strain or SARS-CoV-2 or related to genetic factors of the individuals within the cluster studied. Lastly, the authors do not provide much detail about how other sequences in the phylogenetic tree were selected or discuss whether this selection is truly representative of SARS-CoV-2 in Brazil, which could affect interpretation of the relationship between the two samples collected from Patient C. Given this, it is currently unclear exactly how different or similar are the two sequences from patient C.

Value added

This study provides serological and immunological evidence of SARS-CoV-2 reinfection with no neutralizing antibody response and a very short duration of time in between infections.

This review was posted on: 19 June 2021