Prevalence and Dynamics of SARS-CoV-2 Antibodies in the Population of St. Petersburg, Russia

Abstract

Background: The aim of the study was to assess the prevalence of seropositive status for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2)-IgA, -IgM, and -IgG; its dynamics in connection with restrictive measures during the coronavirus disease (COVID-19) pandemic; and the quantitative dynamics of antibody levels in the population of St. Petersburg, Russia.

Methods: From May to November 2020, a retrospective analysis of Saint Petersburg State University Hospital laboratory database was performed. The database included 158,283 test results of 87,067 patients for SARS-CoV-2 detection by polymerase chain reaction (PCR) and antibody detection of SARS-CoV-2-IgA, -IgM, and -IgG. The dynamics of antibody level was assessed using R v.3.6.3.

Results: The introduction of a universal lockdown was effective in containing the spread of COVID-19. The proportion of seropositive patients gradually decreased; approximately 50% of these patients remained seropositive for IgM after 3-4 weeks; for IgG, by follow-up week 22; and for IgA, by week 12. The maximum decrease in IgG and IgA was observed 3-4 months and 2 months after the detection of the seropositive status, respectively.

Conclusions: The epidemiological study of post-infection immunity to COVID-19 demonstrates significant differences in the dynamics of IgA, IgM, and IgG seropositivity and in PCR test results over time, which is linked to the introduction of restrictive measures. Both the proportion of seropositive patients and the level of all antibodies decreased in terms of the dynamics, and only approximately half of these patients remained IgG-positive 6 months post-infection.

Keywords: Antibody; COVID-19; Quarantine; SARS-CoV-2; Seropositivity; Seroprevalence.

Fig. 1

Prevalence of positive SARS-CoV-2 IgA, IgM, IgG, and PCR patients for whom each test result was available separately. If more than one result was available for a patient in a calendar month, the "worst" status (most positive) was considered. Exact estimates of positive, borderline, and negative test results are given in Appendix 1 (Tables S2-S5). PCR, polymerase chain reaction

Fig. 2

Dynamics of SARS-CoV-2 IgA seropositivity since the first positive test result. If more than one result was available for a patient during each week, the “worst” status was considered (in order: positive, doubtful, negative)

Fig. 3

Dynamics of SARS-CoV-2 IgM seropositivity since the first positive test result. If more than one result was available for a patient during each week, the “worst” status was considered (in order: positive, doubtful, negative)

Fig. 4

Dynamics of SARS-CoV-2 IgG seropositivity since the first positive test result. If more than one result was available for a patient during each week, the “worst” status was considered (in order: positive, doubtful, negative)

Fig. 5

SARS-CoV-2 IgA level plotted over time, starting from the first positive test result. Native (non-transformed) data, medians, first and third quartiles are given, the shape of the background figures reflects the distribution. If more than one result was available for the patient during the month, the “worst” status (the highest test value) was considered. The dotted red line indicates the assay cut-off value (1.1). The box indicates the statistical significance of the fixed effect (“month”) in the omnibus test, with solid lines indicating statistically significant pairwise comparisons (if p value estimate not given, it is less than 0.0001)

Fig. 6

SARS-CoV-2 IgG level plotted over time, starting from the first positive test result. Native (non-transformed) data, medians, first and third quartiles are given, the shape of the background figures reflects the distribution. If more than one result was available for the patient during the month, the "worst" status (the highest test value) was considered. The dotted red line indicates the assay cut-off value (1.1). The box indicates the statistical significance of the fixed effect (“month”) in the omnibus test, with solid lines indicating statistically significant pairwise comparisons (if p value estimate not given, it is less than 0.0001)