Near real-time data on the human neutralizing antibody landscape to influenza virus as of early 2026 to inform vaccine-strain selection

1. Problem Statement

Seasonal influenza vaccines are updated biannually to match circulating viral strains. However, the rapid evolution of influenza A viruses (H3N2 and H1N1) often outpaces these updates.

• The Challenge: By the time vaccine-strain selection meetings occur (February for the Northern Hemisphere), the dominant circulating strains may have already shifted antigenically.
• The Gap: Traditional serological methods (e.g., hemagglutination inhibition in ferrets) are slow and may not fully capture the complexity of human immune responses. There is a critical need for near real-time, high-throughput data using human sera to accurately assess the antigenic distance between current vaccine candidates and emerging dominant subclades before the next selection meeting.

2. Methodology

The authors employed a high-throughput sequencing-based neutralization assay to generate a comprehensive map of human neutralizing antibody titers.

• Viral Library Construction:

  • Strain Selection: Based on HA sequences circulating in late 2025, the team selected 53 recent H3N2 strains and 30 recent H1N1 strains. Selection criteria included high frequency, rapid growth rates, and mutations at known antigenic sites or receptor-binding regions.
  • Subclades Covered: The library specifically targeted the newly dominant H3N2 subclade K (derived from J.2.4) and H1N1 subclade D.3.1.1 (derived from D.3.1), as well as historical vaccine strains (2018–2026).
  • Barcoding: 91 unique Hemagglutinin (HA) proteins were expressed in a library of 186 unique barcoded viral variants (using the lab-adapted A/WSN/1933 backbone). This allowed for multiplexed neutralization assays where multiple strains are tested simultaneously in a single well.

• Human Sera Panel:

  • Cohort: 302 human serum/plasma samples collected primarily in October–November 2025.
  • Demographics: Participants ranged from 0 to 103 years old across five geographic locations (USA, Japan, Hong Kong).
  • Subsets: Included pre- and post-vaccination sera from the 2025–2026 Northern Hemisphere vaccine (Flulaval Trivalent), pediatric sera, and blood donor samples.

• Assay Protocol:

  • Sera were treated with receptor-destroying enzyme (RDE) and heat-inactivated.
  • Serial dilutions of sera were mixed with the pooled barcoded viral library.
  • After incubation with MDCK-SIAT1 cells, viral RNA was extracted, and barcodes were sequenced.
  • Metric: Neutralization titers (IC50) were calculated as the reciprocal serum dilution required to reduce infectivity by 50%.
  • Scale: The study generated 27,409 individual titers (17,142 for H3N2 and 10,267 for H1N1).

3. Key Results

A. H3N2 Neutralization Landscape

• Dominance of Subclade K: The newly dominant H3N2 subclade K showed significantly lower neutralization titers (median ~3-fold lower) compared to the current vaccine strain (J.2 clade) and the strain selected for the 2026 Southern Hemisphere vaccine (J.2.4).
• Antigenic Erosion within Subclade K: Not all K strains were equal. Subclade K variants with additional mutations in antigenic regions D and E (sites 96, 207, 223, and 261) exhibited the lowest titers, suggesting these specific mutations further erode immunity.
• Age Stratification: Children (SCH cohort) showed a more pronounced drop in titers against recent subclades (J.2.3, J.2.4, K) compared to adults, indicating a lack of prior immunological imprinting against these newer variants.

B. H1N1 Neutralization Landscape

• Dominance of Subclade D.3.1.1: Similar to H3N2, the dominant H1N1 subclade D.3.1.1 (mutations R113K, A139D, E283K, K302E) was poorly neutralized compared to the current vaccine strain (C.1.1) and the Southern Hemisphere candidate (D.3.1).
• Age-Specific Vulnerability: Adolescents and young adults showed significantly reduced titers against strains with mutations at sites 155 and 157.
• Reversion Effects: Children showed reduced titers to strains with the N130K mutation, likely because N130 was present in strains they were previously exposed to, whereas adults had been exposed to the K130 variant, making the reversion to N130 less novel for them.

C. Vaccine Impact (Post-Vaccination Analysis)

• H3N2: Vaccination with the 2025–2026 egg-based vaccine (J.2 component) induced a moderate increase in titers (~1.5–2.5 fold). However, the response was weaker against the most recent subclades (K, J.2.4), particularly those with D/E region mutations.
• H1N1: Post-vaccination titer increases were modest (~1.4–2 fold). The vaccine provided little additional protection against the emerging D.3.1.1 subclade compared to pre-vaccination levels.

4. Key Contributions

1. Near Real-Time Antigenic Mapping: Provided the most comprehensive, high-resolution view of human neutralizing antibody landscapes against influenza as of early 2026, capturing the shift to subclades K and D.3.1.1.
2. Identification of "Escape" Mutations: pinpointed specific mutations within the dominant subclades (specifically in H3N2 antigenic regions D/E and H1N1 sites 155/157) that drive further immune escape.
3. Human-Centric Data: Moved beyond ferret models to utilize a diverse panel of 302 human sera, revealing age-specific immune vulnerabilities that animal models might miss.
4. Open Science: All data, code, and interactive visualizations (including Nextstrain trees colored by titer) were made publicly available to facilitate immediate use by vaccine manufacturers and public health officials.

5. Significance

• Vaccine Strain Selection: The data strongly suggests that the strains selected for the 2026 Southern Hemisphere vaccine (J.2.4 for H3N2 and D.3.1 for H1N1) may be antigenically mismatched with the strains dominating in late 2025/early 2026. The authors' data supports updating the 2026–2027 Northern Hemisphere vaccine to include strains from H3N2 subclade K (specifically those without D/E mutations, or potentially a variant thereof) and H1N1 subclade D.3.1.1.
• Evolutionary Forecasting: The identification of low-titer subvariants within dominant clades (e.g., K strains with D/E mutations) serves as an early warning system for potential future dominant strains.
• Public Health Strategy: The findings highlight the need for more frequent surveillance and potentially more rapid vaccine update mechanisms to keep pace with the accelerated evolution of influenza post-pandemic.

In summary, this study provides a critical, data-driven foundation for updating the 2026–2027 influenza vaccine, demonstrating that the current dominant strains have already drifted significantly from the vaccine components, necessitating an immediate strain update to maintain population immunity.