Influenza A(H3N2) Subclade K During the 2025–26 Season and Differential Humoral Responses in Vaccinated Individuals Versus Vaccinated Convalescents in Israel

Abstract

Background/Objectives: During the 2025–26 influenza season, the antigenically distinct A(H3N2) subclade K emerged and spread globally, raising significant concerns regarding seasonal vaccine effectiveness. This study aimed to map antigenic and structural divergence of this novel strain compared to the 2025/26 A(H3N2) vaccine component. Furthermore, we investigated how this vaccine mismatch impacts functional humoral immunity by directly comparing antibody neutralization profiles between individuals with vaccine-induced protection and those with infection-acquired immunity. Methods: Phylogenetic and structural modeling analyses were performed on 2025–26 Israeli influenza samples to identify genetic distances and mutations in the hemagglutinin protein compared to the 2025/26 A(H3N2) vaccine component. Hemagglutination inhibition (HI) and microneutralization assays were conducted using 2025–26 winter serum samples from a cohort of vaccinated healthcare workers and vaccinated convalescents to quantify antibody responses against both vaccine-like and subclade K strains. Results: Phylogenetic analysis revealed a clear dominance of the subclade K strain in Israel during the winter of 2025–26. Significant vaccine mismatch was defined by eight newly identified hemagglutinin mutations in the circulating subclade K strains. Serological testing demonstrated that while vaccination alone produced a modest two-to-three-fold increase in antibody titers, baseline protection against subclade K remained critically low. Conversely, natural infection in vaccinated convalescents elicited a significantly stronger, broader immune response, demonstrating a 6-fold and 13-fold increase in neutralizing antibodies against the vaccine-like and subclade K strains, respectively. Conclusions: Natural infection provides a more pronounced increase in both binding and neutralizing antibodies compared to vaccination alone, which might contribute to broader cross-protection against antigenically drifted variants. These findings highlight the primary limitations of current influenza strain prediction and emphasize the urgent need for faster, more adaptable vaccine manufacturing platforms to better match actual circulating strains.