High SARS-CoV-2 seroprevalence in rural Peru, 2021; a cross-sectional population-based study

Abstract Latin America has been severely affected by the COVID-19 pandemic. The COVID-19 burden in rural settings in Latin America is unclear. We performed a cross-sectional, population-based, random-selection SARS-CoV-2 serological study during March 2021 in the rural population of San Martin region, northern Peru. The study enrolled 563 persons from 288 houses across 10 provinces, reaching 0.19% of the total rural population of San Martin. Screening for SARS-CoV-2 IgG antibodies was done using a chemiluminescence immunoassay (CLIA) and reactive sera were confirmed using a SARS-CoV-2 surrogate virus neutralization test (sVNT). Validation using pre-pandemic sera from two regions of Peru showed false-positive results in the CLIA (23/84 sera; 27%), but not in the sVNT, highlighting the pitfalls of SARS-CoV-2 antibody testing in tropical regions and the high specificity of the two-step testing algorithm. An overall 59.0% seroprevalence (95% CI: 55-63%) corroborated intense SARS-CoV-2 spread in San Martin. Seroprevalence rates between the 10 provinces varied from 41.3-74.0% (95% CI: 30-84). Higher seroprevalence was neither associated with population size, population density, surface area, mean altitude or poverty index in spearman correlations. Seroprevalence and reported incidence diverged substantially between provinces, suggesting regional biases of COVID-19 surveillance data. Potentially, limited healthcare access due to environmental, geographic, economic, and cultural factors, might lead to undetected infections in rural populations. Additionally, test avoidance to evade mandatory quarantine might affect rural regions more than urban regions. Serologic diagnostics should be pursued in resource-limited settings to inform country-level surveillance, vaccination strategies and support control measures for COVID-19. Importance Latin America is a global hotspot of the COVID-19 pandemic. Serological studies in Latin America have been mostly performed in urban settings. Rural populations comprise 20% of the total Latin American population. Nevertheless, information of COVID-19 spread and transmission in rural settings is scarce. Using a representative population-based seroprevalence study, we detected a high seroprevalence in rural populations in San Martin, northern Peru in 2021, reaching 41 to 74 %. However, seroprevalence and reported incidence diverged substantially between regions, suggesting either limited healthcare access or test avoidance due to mandatory quarantine. Our results suggest that rural populations are highly affected by SARS-CoV-2 even though they are socio-demographically distinct from urban populations, and that highly specific serological diagnostics should be performed in resource-limited settings to support public-health strategies of COVID-19 surveillance and control.