Risk of BA.5 infection in individuals exposed to prior SARS-CoV-2 variants

Abstract

The SARS-CoV-2 omicron BA.5 subvariant is progressively displacing earlier subvariants, BA.1 and BA.2, in many countries. One possible explanation is the ability of BA.5 to evade immune responses elicited by prior BA.1 and BA.2 infections. The impact of BA.1 infection on the risk of reinfection with BA.5 is a critical issue because adapted vaccines under current clinical development are based on BA.1. We used the national Portuguese COVID-19 registry to analyze the risk of BA.5 infection in individuals without a documented infection or previously infected during periods of distinct variants' predominance (Wuhan-Hu-1, alpha, delta, BA.1/BA.2). National predominance periods were established according to the national SARS-CoV-2 genetic surveillance data (when one variant represented >90% of the sample isolates). We found that prior SARS-CoV-2 infection reduced the risk for BA.5 infection. The protection effectiveness, related to the uninfected group, for a first infection with Wuhan-Hu-1 was 52.9% (95% CI, 51.9 - 53.9%), for Alpha 54.9% (51.2 - 58.3%), for Delta 62.3% (61.4 - 63.3%), and for BA.1/BA.2 80.0% (79.7 - 80.2%). The results ought to be interpreted in the context of breakthrough infections within a population with a very high vaccine coverage (>98% of the study population completed the primary vaccination series). In conclusion, infection with BA.1/BA.2 reduces the risk for breakthrough infections with BA.5 in a highly vaccinated population. This finding is critical to appraise the current epidemiological situation and the development of adapted vaccines.

Rapid waning of protection induced by prior BA.1/BA.2 infection against BA.5 infection

“…We further showed that the impact of 20% or 40% of unreported infections within the "uninfected" group did not lead to major changes in the calculated risks 1 . We used the national SARS-CoV-2 genetic surveillance database 7 to identify periods when one variant represented >90% of the sample isolates, as also defined and used in other studies 1,3 .…”

“…The number of unreported cases was estimated by the National Serological Panel (from November 2021, assaying the presence of antibodies against the SARS-CoV-2 N protein to exclude vaccination seropositivity) to infer that, at that time, there were 29.2% more people who had been infected with SARS-CoV-2 than officially reported 19 . We further showed that the impact of 20% or 40% of unreported infections within the "uninfected" group did not lead to major changes in the calculated risks 1 . We used the national SARS-CoV-2 genetic surveillance database 7 to identify periods when one variant represented >90% of the sample isolates, as also defined and used in other studies 1,3 .…”

“…The population included in the study comprises: It is anticipated that the population we classified as "uninfected" included individuals with a prior unnoticed infection. In a previous publication we have shown that the calculated risk of BA.5 infection does not significantly change even if we assume a percentage of 29.2% unreported infections in each period 1 . The number of unreported cases was estimated by the National Serological Panel (from November 2021, assaying the presence of antibodies against the SARS-CoV-2 N protein to exclude vaccination seropositivity) to infer that, at that time, there were 29.2% more people who had been infected with SARS-CoV-2 than officially reported 19 .…”

“…The SARS-CoV-2 omicron BA.5 subvariant has been progressively displacing earlier BA.1 and BA.2 subvariants. We and others have recently reported that natural infection with BA.1/BA.2 offers significant protection against BA.5 reinfection [1][2][3] . However, the duration of the protective effect is a concern, given the poor ability of BA.1/BA.2 infection to achieve robust neutralizing antibodies against BA.…”

“…Portugal was one of the first countries affected by omicron BA.1 and, more recently, omicron BA.5. We used national SARS-CoV-2 genetic surveillance data to identify periods when different variants represented >90% of the isolates 1,7 . Then, we used the Portuguese COVID-19 registry ( SINAVE ), which includes all notified cases of infection in the country based on a positive molecular or professional rapid antigen test for SARS-CoV-2 and irrespective of clinical presentation, to identify all individuals infected during those periods.…”