COVID-19 Pandemic
Vaccination with Strain-Targeted COVID-19 Vaccines May Induce Selection Pressure for Less Genetically Related SARS-CoV-2 Strains Keeley Morris* Keeley Morris Gillian Tarr Scott Seys Cody Schardin Jake Garfin Miriam Muscoplat Kathryn Como-Sabetti Stephanie Meyer
On 9/1/2022, bivalent COVID-19 vaccines targeting the Omicron BA.4/5 lineages were released. Little is known about the role targeted COVID vaccines might play in exerting selection pressure that advantages lineages not targeted by the vaccine (i.e., BQ.1 and XBB). This study examined whether COVID cases who received bivalent vaccines were at increased risk of infection by non-targeted lineages relative to cases who had not.
We linked COVID case and immunization data from the Minnesota Department of Health (MDH) to whole genome sequencing results from the MDH Public Health Lab for all sequenced specimens collected between 10/1/2022 and 3/1/2023. We performed three separate log-binomial regressions to model the risk that a case was infected with BA.4/5, BQ.1, or XBB lineages as a function of vaccination status (unvaccinated, monovalent vaccinated, bivalent vaccinated). All models were adjusted for age, sex, race/ethnicity, and prior infection.
Bivalent cases were older, more likely to be white, and less likely to have a prior COVID infection than monovalent and unvaccinated cases. Bivalent cases were less likely to be infected with BA.4/5 and more likely to be infected with BQ.1 or XBB [Figure]. After adjustment, bivalent cases were 0.55 (95%CI: 0.49-0.60), 1.35 (95%CI: 1.24-1.48), and 1.68 (95%CI: 1.48-1.91) times as likely as monovalent cases to be infected with BA.4/5, BQ.1, or XBB lineages, respectively. Results were similar when comparing to unvaccinated cases.
A limitation when comparing bivalent to monovalent vaccines is the lack of temporal overlap in vaccine administration, resulting in greater waning immunity among monovalent cases by the end of our study when XBB was more prominent. Therefore, these results represent a conservative estimate of selection pressure for non-targeted lineages. Our results suggest that bivalent vaccines may induce selection pressure, increasing the risk of infection with non-targeted lineages among bivalent vaccinated cases.
