In a recent study published on the Preprints server with The Lancet*, researchers estimated the vaccine efficacy (VE) of primary and booster coronavirus disease 2019 (COVID-19) vaccines against reinfection by Omicron lineages BA.5 and BA.2. and severe outcomes after infection.
Study: Comparative effectiveness of COVID-19 vaccines in preventing SARS-CoV-2 Omicron lineages BA.5 and Ba.2 infections, hospitalizations and deaths: a case-cohort study using electronic health records in Portugal . Image credit: Adao/Shutterstock
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) Omicron BA.5 subvariant was reclassified as a variant of concern (VOC) by the European Center for Disease Prevention and Control (ECDC) on 12 May of 2022. Its prevalence represented about 37% of the cases of COVID-19 in Portugal in May 2022. The National Institute of Health of Portugal estimated a growth advantage of 13% per day for BA.5 , which, together with a doubling time of about six days, would have made it the dominant lineage in Portugal.
background
There is conflict between results from animal models, neutralization assays, cell cultures, and early vaccination risk assessment comparing BA.5 with other Omicron lineages. Due to the lack of a negative control group, any study design that quantifies the relative effect of COVID-19 vaccination against Omicron reinfections among highly vaccinated populations could be useful.
Study design
In the present study, the investigators compared the VE of primary and booster vaccination between the Omicron BA.5 and BA.2 lineages. They evaluated how vaccination prevented breakthrough infection of Omicron lineages and estimated lineage-specific post-infection VE against COVID-19-related hospitalization and mortality.
They implemented two different approaches. In the first approach, they compared case-by-case odds of the COVID-19 vaccine among individuals with breakthrough Omicron BA.5 and BA.2 infection. In the second approach, they followed BA.2 and BA.5-infected groups to assess lineage-specific post-infection VE (VEp) against COVID-19-related hospitalization and death. They compared VEp only between vaccinated and unvaccinated infected individuals to compare the risk of serious outcomes.
The study population consisted of individuals diagnosed with COVID-19 from mainland Portugal between April 25 and June 10, 2022, by reverse transcriptase-polymerase chain reaction (RT-PCR), with samples positives subjected to S. SGTF gene target failure) or whole genome sequencing (WGS).
The team considered only samples with positive nucleocapsid (N) and open reading frame 1a (ORF1a) signals and cycle threshold (Ct) values ≤30 for SGTF-based classification. According to vaccination status, the study population comprised unvaccinated individuals and those who had received complete primary and booster vaccinations.
The researchers obtained hospitalization data related to COVID-19 from the registry of the Integrated Hospital Information System (SONHO), which collects all information from Portugal’s National Health Service (NHS) hospitals. Likewise, the National Death Certificate Information System (SICO) issues a death certificate for each person who dies in Portugal with COVID-19 as the main cause of death for which the International Classification of Diseases (ICD) -10 code U.071 according to the World Health Organization classification is used.
In addition, researchers performed data extraction and deterministic linkage of electronic health records with laboratory data on July 12, 2022. In addition, they collected age, sex, residence, and week of collection of swabs through the national surveillance system SINAVE. In addition, the researchers used a logistic regression model adjusted for sex, age group, region of residence, and week of swab collection to estimate the odds of vaccination.
An odds ratio (OR) estimate greater than one and one suggested that VE was lower for the BA.5 lineage than BA.2 and the same for the two Omicron lineages, respectively. They interpreted ORs for prior infection similarly. Finally, the team used penalized logistic regression to reduce bias caused by hospitalization or deaths for VEp estimates against severe outcomes in BA.5 and BA.2 cases. The OR for the interaction of lineage and vaccination status measured the relationship between relative VE to prevent serious outcomes among those infected with BA.5 compared with BA.2.
Results of the study
Between April 25 and June 10, 2022, there were 15,396 BA.2 cases and 12,306 BA.5 cases. Compared to BA.2 cases, BA.5 cases were slightly younger and frequently resided in the Alentejo and Center regions. In addition, the proportion of cases with a previous infection with COVID-19 was higher in BA.5 than in BA.2 (10% vs. 5.6%). Regarding vaccination status, both groups had between 4 and 5% of unvaccinated cases, but BA.5 had a higher proportion of individuals who had received primary vaccination series (20.6% vs . 15.8%), and BA.2 had more individuals who had received a booster dose (80.1% vs. 74.7%). In addition, there were 106 hospitalizations for COVID-19 and 42 deaths.
The odds of complete primary vaccination or booster dose (aOR=1.07 vs. 0.96) among BA.5 cases were similar to BA.2 cases, suggesting no relevant differences in VE vs. infection for the BA.5 lineage compared to BA. 2. The authors observed higher odds of previous infection in BA.5 cases compared to BA.2 (aOR=1.44). Combining primary vaccination and previous infection status, the estimated aOR for BA.5 was 1.70 times higher than for a re-infection with BA.2. For booster vaccination, a greater reduction in the risk of hospitalization was observed for both BA.2 and BA.5, representing a post-infection VE of 93% and 77%, respectively.
The interaction term allowing the comparison between BA.5/BA.2 lineage VEs was 3.36, suggesting reduced post-infection protection induced by the booster dose against BA hospitalization. 5 compared to BA.2. The ratio of complete primary VE after infection in BA.5 vs BA.2 cases was 2.06 with a wide confidence interval (CI). For the booster dose, the aOR of death suggested a greater risk reduction than unvaccinated for BA.2 (VEp=94%) than for BA.5 (VEp=88%), with overlapping ICs. For the death outcome, the interaction term allowing comparison of post-infection VE between BA.5/BA.2 lineages was 0.43 for full vaccination and 1.98 for booster dose , both with very imprecise IC.
Conclusions
Overall, VE against BA.5 and BA.2 infection was similar. However, the SARS-CoV-2 Omicron BA.5 lineage was associated with higher odds of prior infection compared to BA.2, suggesting a reduction in the protection conferred by prior infection against BA.5 in comparison with BA.2. Also, compared to hospitalization outcomes, VE was a lower post-BA.5 infection-boosting dose than the post-BA.2 infection-boosting dose. Furthermore, among patients infected with BA.5, the protective effect of the first boost on reducing the odds of hospitalization outcome was greater than that of primary vaccination (VEp=77% vs. 22%). For the outcome of death, the authors estimated a high post-infection VE of the booster dose for both BA.5 and BA.2 cases.
In summary, the vaccines currently used in Portugal were less effective in reducing the risk of severe outcomes for BA.5 than BA.2. Differences observed in primary vaccination and post-infection VE booster dose versus severe outcomes observed for BA.5 and BA.2 lineages highlighted the importance of higher vaccination coverage to prevent COVID -19 serious
*Important news
Preprints with The Lancet publish preliminary scientific reports that are not peer-reviewed and therefore should not be considered conclusive, guide clinical practice or health-related behavior, or be treated as established information.