Revising the influenza vaccine selection timeline in order to improve the antigenic match between vaccine and circulating viruses could yield significant epidemiological benefits, according to study findings published in The Journal of Infectious Diseases.
The occurrence of antigenic drift during the 5-month gap between vaccine virus selection and vaccine distribution can lead to less effective influenza vaccines. The emergence of non-egg-based vaccine production technologies could potentially shorten the gap between virus selection and vaccine distribution by 1 to 3 months. Investigators therefore sought to evaluate the possible benefit of delaying seasonal influenza vaccine virus selection decisions.
The investigators conducted a retrospective modeling study in the US, noting the increased prevalence of the season-dominant A(H3N2) subclade (2014-2015) and the season-dominant B/Victoria clade (2019-2020) after vaccine decisions were already made for the Northern Hemisphere. A SEIR model of seasonal influenza (which divides the population into 4 groups: susceptible [S], incubating [E], infected [I], and recovered [R]), was used to estimate the US influenza burden that might have been avoided if vaccine decisions had been delayed or updated to ensure a more antigenically-similar match between vaccines and viruses.
The investigators used data from the Global Initiative on Sharing All Influenza Data (GISAID; available every month for A[H3N2], A[H1N1]pdm09, and B/Victoria, and every 6 months for B/Yamagata) to evaluate the level of viral activity and antigenic characteristics of emergent viruses. Season-dominant clades or subclades were defined as those prevailing early (September-January) and late (February-May) in the season.
“
With rapid vaccine production, revising current timelines for vaccine selection could result in substantial epidemiological benefits, particularly when additional data could help improve the antigenic match between vaccine and circulating viruses.
Changing viral activities were accounted for over 3 intervals: (1) prior to vaccine virus decision (5 months before February); (2) between the vaccine decision and season start (February-August); and (3) after the season start. During the 8 seasons analyzed, 18 clades or subclades were predominant, of which 6 were antigenically distinct from the selected vaccine virus (dominant clades present with less than 50% frequency at the time of the vaccine decision).
Sharp increases in activity after vaccine decision (February) but before start of influenza season (September) were noted in the antigenically distinct season-dominant H3N2 subclade in 2014-2015 and the B/Victoria clade in 2019-2020, suggesting the vaccine decision might have been different if the decision had been made later than February.
In modeling, the investigators found 5000 to 65,000 influenza hospitalizations in the US in the 2014-2015 season could have been averted with an updated A(H3N2) vaccine. In modeling of the 2019-2020 season, updating the B/Victoria vaccine did little to change the influenza burden.
Study limitations include unaccounted-for factors such as prior vaccination status, age, and vaccine type and the inability to generalize the estimated impact of vaccination changes beyond US influenza epidemics. In addition, the investigators also noted that the real-world transition away from the egg-based vaccine platform may require significant time.
“With rapid vaccine production, revising current timelines for vaccine selection could result in substantial epidemiological benefits, particularly when additional data could help improve the antigenic match between vaccine and circulating viruses,” the investigators concluded. “The uncertainty in timing of antigenic variant emergence and the risk of delaying vaccine decision along with the harm of transitioning away from egg-based vaccines should be carefully examined,” they added.
Disclosure: Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.
