Background: In low- and middle-income countries where SARS-CoV-2 testing is limited, seroprevalence studies can characterise the scale and determinants of the pandemic, as well as elucidate protection conferred by prior exposure.Methods: We conducted repeated cross-sectional serosurveys (July 2020 - November 2021) using residual plasma from routine convenient blood samples from patients with non-COVID-19 conditions from Cape Town, South Africa. SARS-CoV-2 anti-nucleocapsid antibodies and linked clinical information were used to investigate: (1) seroprevalence over time and risk factors associated with seropositivity, (2) ecological comparison of seroprevalence between subdistricts, (3) case ascertainment rates, and (4) the relative protection against COVID-19 associated with seropositivity and vaccination statuses.Findings: Among the subset sampled, seroprevalence of SARS-CoV-2 in Cape Town increased from 39.2% in July 2020 to 67.8% in November 2021. Poorer communities had both higher seroprevalence and COVID-19 mortality. Only 10% of seropositive individuals had a recorded positive SARS-CoV-2 test. Antibody positivity before the start of the Omicron BA.1 wave (28 November 2021) was strongly protective for severe disease (adjusted odds ratio [aOR] 0.15; 95%CI 0.05-0.46), with additional benefit in those who were also vaccinated (aOR 0.07, 95%CI 0.01-0.35).Interpretation: The high population seroprevalence in Cape Town was attained at the cost of substantial COVID-19 mortality. At the individual level, seropositivity was highly protective against subsequent infections and severe COVID-19.Funding Information: The study is funded by National Health Laboratory service, Western Cape Department of Health, Wellcome Trust, and in part by the Division of Intramural Research, NIAID, NIH. RJW is supported by the Francis Crick Institute which receives funding from Cancer Research UK (FC0010218), Medical Research Council (FC0010218), and Wellcome (FC0010218). He also received funding from Wellcome (203135,222754).Declaration of Interests: No competing interests.Ethics Approval Statement: The study was approved by the Human Research Ethics Committees of the University of Cape Town (HREC REF 449/2020) and Stellenbosch University (N20/08/051). Institutional approval was obtained from the National Health Laboratory Service and the Western Cape Department of Health.},