Estimated protection against COVID-19 based on predicted neutralisation titres from multiple antibody measurements in a longitudinal cohort, France, April 2020 to November 2021
Tom Woudenberg, Laurie Pinaud, Laura García, Laura Tondeur, Stéphane Pelleau, Alix De Thoisy, Françoise Donnadieu, Marija Backović, Mikaël Attia, Nathanaël Hozé, Cécile Duru, Aymar Davy Koffi, Sandrine Castelain, Marie‐Noëlle Ungeheuer, Sandrine Fernandes Pellerin, Delphine Planas, Timothée Bruel, Simon Cauchemez, Olivier Schwartz, Arnaud Fontanet, Michael White
Abstract
BackgroundThe risk of SARS-CoV-2 (re-)infection remains present given waning of vaccine-induced and infection-acquired immunity, and ongoing circulation of new variants.AimTo develop a method that predicts virus neutralisation and disease protection based on variant-specific antibody measurements to SARS-CoV-2 antigens.MethodsTo correlate antibody and neutralisation titres, we collected 304 serum samples from individuals with either vaccine-induced or infection-acquired SARS-CoV-2 immunity. Using the association between antibody and neutralisation titres, we developed a prediction model for SARS-CoV-2-specific neutralisation titres. From predicted neutralising titres, we inferred protection estimates to symptomatic and severe COVID-19 using previously described relationships between neutralisation titres and protection estimates. We estimated population immunity in a French longitudinal cohort of 905 individuals followed from April 2020 to November 2021.ResultsWe demonstrated a strong correlation between anti-SARS-CoV-2 antibodies measured using a low cost high-throughput assay and antibody response capacity to neutralise live virus. Participants with a single vaccination or immunity caused by infection were especially vulnerable to symptomatic or severe COVID-19. While the median reduced risk of COVID-19 from Delta variant infection in participants with three vaccinations was 96% (IQR: 94-98), median reduced risk among participants with infection-acquired immunity was only 42% (IQR: 22-66).ConclusionOur results are consistent with data from vaccine effectiveness studies, indicating the robustness of our approach. Our multiplex serological assay can be readily adapted to study new variants and provides a framework for development of an assay that would include protection estimates.