Cross-Reactivity of SARS-CoV-2 Laboratory Diagnostics to Endemic Diseases in Africa: A Diagnostic Accuracy Study (preprint)

ABSTRACT

Background: Serology is a great tool to assess the level of immunity against SARS-CoV-2 in settings with limited access to molecular diagnostics. However, African populations displays a particular immunological profile with massive circulation of infectious agents from different aetiologies that can affect assays performance.Methods: We evaluated the OMEGA Diagnostics COVID-19 ELISA-IgG and the ID Screen® SARS-CoV-2-N IgG Indirect in Senegal using a panel of 636 blood samples covering several African-endemic diseases and healthy donors to determine test sensitivity and specificity. The sensitivity panel of sera includes 461 serum samples collected from 91 patients hospitalized for COVID-19 disease. COVID-19 cases were confirmed by qRT-PCR and samples were collected on an interval of three days until viral clearance. In addition, 272 sera obtained from COVID-19 negative individuals were selected from a well-documented biobank of sera collected before the COVID-19 outbreak.Finding: High-cross reactivity have been found in individuals with a history of exposure to Chikungunya, HIV, malaria (Plasmodium falciparum), rheumatoid factor as well as healthy donors with respective specificities of 55%, 41.8%, 70%, 70% and 75%. ELISA experiments with commercial assays targeting either SARS-CoV-2 Nucleocapsid protein and Spike 2 protein or nucleocapsid protein only suggest that cross-reactivity might be directed against Spike 2 protein and not Nucleocapsid protein. Further samples characterisation reveals that anti-malaria IgG is the leading cause of such poor specificities, but exposure to other diseases contributed as well.Interpretation: We anticipate that COVID-19 seroprevalence can be biased if assays are not contextualized. Since malaria is endemic in African settings, we propose that a particular attention must be given in serological surveillance of COVID-19 or anti-SARS-CoV-2 antibodies quantification as vaccines are being rolled out.FundingUK Foreign, Commonwealth and Development Office/Wellcome Trust Joint Initiative for Research in Epidemic Preparedness and Response (JIREP grant number 220764/Z/20/Z).Funding Information UK Foreign, Commonwealth and Development Office/Wellcome Trust Joint Initiative for Research in Epidemic Preparedness and Response (JIREP grant number 220764/Z/20/Z).Declaration of Interests JRAF was an employee of Mologic Ltd, which was the development partner of one of the ELISAs adopted in this study. The remaining authors declare no competing interest.Ethics Approval Statement Pre-COVID-19 samples for malaria (PCR), dengue, yellow fever, Zika, Chikungunya, Influenza A/B, HIV, rheumatoid arthritis and samples tested negative for the same diseases and also Crimean Congo haemorrhagic fever, West Nile fever encephalitis and Rift Valley fever were part of national public health surveillance program of the Senegalese Ministry of Social Action and Health performed in collaboration with Institut Pasteur de Dakar. Therefore, consultation with ethics committee was not required. Pre-COVID-19 samples for malaria endemic areas were from a longitudinal cohort survey performed in Dielmo village and approved by the Senegalese National Ethics Committee for Research in Health (reference number 00000007/MSAS/CNERS/Sec 26 January 2021). Samples from COVID-19 RT-PCR positive patients were obtained from a multicentre cohort survey approved by the Senegalese National Ethics Committee for Research in Health (reference number 00000068/MSAS/CNERS/Sec, 10 April 2020).