Background In low-income countries, like Malawi, important public health measures including social distancing or a lockdown, have been challenging to implement owing to socioeconomic constraints, leading to predictions that the COVID-19 pandemic would progress rapidly. However, due to limited capacity to test for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection, there are no reliable estimates of the true burden of infection and death. We, therefore, conducted a SARS-CoV-2 serosurvey amongst health care workers (HCW) in Blantyre city to estimate the cumulative incidence of SARS-CoV-2 infection in urban Malawi. Methods Five hundred otherwise asymptomatic HCWs were recruited from Blantyre City (Malawi) from 22nd May 2020 to 19th June 2020 and serum samples were collected all participants. A commercial ELISA was used to measure SARS-CoV-2 IgG antibodies in serum. We run local negative samples (2018 - 2019) to verify the specificity of the assay. To estimate the seroprevalence of SARS CoV-2 antibodies, we adjusted the proportion of positive results based on local specificity of the assay. Results Eighty-four participants tested positive for SARS-CoV-2 antibodies. The HCW with a positive SARS-CoV-2 antibody result came from different parts of the city. The adjusted seroprevalence of SARS-CoV-2 antibodies was 12.3% [CI 9.0-15.7]. Using age-stratified infection fatality estimates reported from elsewhere, we found that at the observed adjusted seroprevalence, the number of predicted deaths was 8 times the number of reported deaths. Conclusion The high seroprevalence of SARS-CoV-2 antibodies among HCW and the discrepancy in the predicted versus reported deaths, suggests that there was early exposure but slow progression of COVID-19 epidemic in urban Malawi. This highlights the urgent need for development of locally parameterised mathematical models to more accurately predict the trajectory of the epidemic in sub-Saharan Africa for better evidence-based policy decisions and public health response planning.

No abstract

Although the COVID-19 pandemic has left no country untouched there has been limited research to understand clinical and immunological responses in African populations. Here we characterise patients hospitalised with suspected (PCR-negative/IgG-positive) or confirmed (PCR-positive) COVID-19, and healthy community controls (PCR-negative/IgG-negative). PCR-positive COVID-19 participants were more likely to receive dexamethasone and a beta-lactam antibiotic, and survive to hospital discharge than PCR-negative/IgG-positive and PCR-negative/IgG-negative participants. PCR-negative/IgG-positive participants exhibited a nasal and systemic cytokine signature analogous to PCR-positive COVID-19 participants, predominated by chemokines and neutrophils and distinct from PCR-negative/IgG-negative participants. PCR-negative/IgG-positive participants had increased propensity for Staphylococcus aureus and Streptococcus pneumoniae colonisation. PCR-negative/IgG-positive individuals with high COVID-19 clinical suspicion had inflammatory profiles analogous to PCR-confirmed disease and potentially represent a target population for COVID-19 treatment strategies.

Background Persistent carriage of pneumococcal vaccine serotypes has occurred after introduction of PCV13 vaccination in Africa but the mechanisms are unclear. We tested the feasibility of using a human pneumococcal challenge model in Malawi to understand immune correlates of protection against carriage and to trial alternative vaccine candidates. We aimed to identify a dose of Streptococcus pneumoniae serotype 6B sufficient to establish nasopharyngeal carriage in 40% of those nasally inoculated and evaluate nasal mucosal immunity before and after experimental inoculation. Methods Healthy student volunteers were recruited and inoculated with saline, 20,000 CFU/naris or 80,000 CFU/naris of Streptococcus pneumoniae serotype 6B Post inoculation carriage was determined by nasal sampling for bacterial culture and lytA PCR. Immunological responses were measured in serum and nasal mucosal biopsies before and after bacterial inoculation. Findings Twenty-four subjects completed the feasibility protocol with minimal side effects. pneumococcal carriage was established in 0/6, 3/9 and 4/9 subjects in the saline, 20,000 CFU/naris and 80,000 CFU/naris groups, respectively. Incidental (natural) serotype carriage was common (7/24 participants carried non-6B strains, 29.2%. Experimentally induced type 6B pneumococcal carriage was associated with pro-inflammatory nasal mucosal responses prior to inoculation and altered mucosal recruitment of immune cells post bacterial challenge. There was no association with serum anti-capsular antibody. Interpretation The serotype 6B experimental human pneumococcal carriage model is feasible in Malawi and can now be used to determine the immunological correlates of protection against carriage and vaccine efficacy in this population.