How do antibody responses to the 2025-2026 H3N2 vaccine compare to the 2024-2025 season?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
Laboratory and surveillance reports show that antibody responses elicited by the 2024–25 H3N2 vaccines were antigenically weaker against some circulating viruses that season, while the 2025–26 vaccine strains were updated because a genetically and antigenically drifting subclade — “subclade K” — rose rapidly in 2025 (CDC/WHO/ECDC reporting) [1] [2]. Studies of H3N2 immune responses indicate vaccination often raises HAI/GMTs to potentially protective levels (e.g., post‑vac GMT ≈95 and HAI ≥80 considered a strong response on average) but infection or hybrid immunity produced higher A/H3N2 antibody titres than vaccination alone in some cohorts [3] [4].
1. Vaccine strain change and why it matters
Global authorities revised the H3N2 vaccine composition for 2025–26 because circulating viruses diverged from the prior vaccine references; WHO candidate viruses and national regulators document cell‑ and egg‑derived 2024–25 vaccine viruses, and the 2025 recommendation replaced those with A/District of Columbia/27/2023‑like (cell/recombinant) and A/Croatia/10136RV/2023‑like (egg) references to better match emergent viruses [5] [6] [1]. The European Centre for Disease Prevention and Control (ECDC) flagged subclade K as substantially divergent in HA with multiple amino‑acid changes compared with the vaccine and with viruses that circulated in 2024–25, a genomic shift that can reduce antibody recognition [2].
2. Laboratory measures: reduced reactivity for H3N2 in 2024–25
U.S. CDC antigenic testing during 2024–25 found that antibodies raised to the 2024–25 vaccine antigens for A(H3N2) “exhibited reduced reactivity but still reacted well with the majority of the circulating viruses tested,” indicating measurable antigenic mismatch for some viruses but not wholesale loss of binding across samples [1]. Independent lab reports and commentary noted ferret antisera and vaccinated‑ferret antibodies sometimes recognized a minority of patient H3N2 viruses in that season (examples cited: ~42% recognition reported in one analysis), which illustrates how animal antisera can flag antigenic drift even when human population protection is more complex [7] [8].
3. Real‑world protection versus lab reactivity: mixed signals
Gavi and public health agencies emphasize that laboratory antibody reactivity is only part of the picture; early UK evidence reported that the 2025–26 vaccine still provided meaningful clinical protection (e.g., estimates of 70–75% effectiveness against hospital attendance in children and 30–40% in adults in one report), even though ferret antibody tests showed reduced recognition of subclade K [8]. CDC likewise noted reduced antigenic reactivity for A(H3N2) sera in lab assays yet described that most circulating viruses were still recognized reasonably well [1]. Thus laboratory antigenic drift does not automatically equal no clinical protection; observational effectiveness can remain substantial in some age groups.
4. What serology studies tell us about magnitude of responses
Aggregated analyses of prior H3N2 vaccine studies found that one‑month post‑vaccination hemagglutination‑inhibition (HAI) responses follow a log‑normal distribution with a reported geometric mean titer (GMT) ~95, and authors consider HAI ≥80 to be a “strong response,” noting about two‑thirds of sera reached that threshold on average in historical data [3]. Conversely, cohort research in healthcare workers found that infection or hybrid immunity produced higher A/H3N2 antibody responses than vaccination alone, and repeated vaccination without strain updates could blunt responses — findings that informed decisions to update vaccine strains for 2025–26 [4].
5. Limitations, uncertainties and competing interpretations
Available sources show consistent genomic divergence of subclade K and reduced laboratory antibody recognition, but they also report retained reactivity for many viruses and early real‑world protection estimates; these are competing but co‑existing findings [2] [1] [8]. Animal ferret data can overstate antigenic drift relative to human immunity — several sources caution that ferret antisera are a standard lab tool but not a perfect surrogate for human vaccine performance [8] [7]. Large‑scale predictive models that use pre‑vaccination titres can forecast post‑vac responses, but these rely on historical datasets and do not directly measure 2025–26 outcomes [9] [10].
6. Takeaway for clinicians and the public
Public‑health agencies updated the 2025–26 H3N2 vaccine because of measurable genetic and antigenic drift; lab tests showed reduced antibody reactivity for some H3N2 viruses in 2024–25, but both lab and early field data indicate the updated vaccine aims to restore better match and protection [2] [1] [8]. Serologic work shows vaccination produces substantial HAI rises for many people (GMTs around 95 in pooled data; HAI ≥80 often considered protective), while infection or hybrid immunity tends to produce still higher A/H3N2 titres — an important nuance when interpreting vaccine‑induced antibody levels [3] [4].
Limitations: available sources do not provide direct head‑to‑head human antibody titer comparisons between the 2024–25 and 2025–26 vaccines measured in the same cohorts; real‑world VE estimates for 2025–26 remain preliminary in public reporting (not found in current reporting).