Near real-time data on the human neutralizing antibody landscape to influenza virus as of the summer of 2026 to inform vaccine-strain selection (Summary results)¶

Overview¶

This report summarizes the results of a study that uses sequencing-based neutralization assays to measure titers to influenza viruses with HAs from human seasonal H3N2 and H1N1 strains representative of those circulating in mid-2026 against human sera collected in early to mid 2026.

For background about sequencing-based neutralization assays, see:

• Loes et al (2024), Journal of Virology
• Kikawa et al (2026), eLife
• Detailed experimental protocol on protocols.io

The full data and computer code are at https://github.com/jbloomlab/flu-seqneut-2026. This report summarizes the key results; a detailed report of all experimental results at a per-plate and per-serum level is available here.

Download titer data¶

To download the processed data, go to this subdirectory. Specifically:

• human_titers.csv: QC-ed titers for each virus/serum pair.
• human_viruses.csv: detailed information about the viruses for which titers were measured.
• human_sera.csv: detailed information about sera for which titers were measured; each serum is listed once in this file. See also human_sera_multicohort.csv for another file that provides additional sera assignments to more fine-grained cohorts (each serum may be listed multiple times in this file).
• human_titers_summarized_by_virus.csv: median and geometric mean titers against each virus.
• human_summary.txt: report on number of sera and viruses analyzed with detailed breakdowns per sera set.

Viral HAs tested¶

We chose a set of naturally occurring HAs to represent the diversity of H3N2 and H1N1 influenza in humans as of the summer of 2026, also including recent vaccine strains for both subtypes. Overall we included 148 HAs (82 H3N2, 66 H1N1) as listed here. For the assays, we used viruses with HA ectodomains from these strains and the rest of the genes from the lab-adapted A/WSN/1933 strain.

The HAs we chose largely covered the current diversity of human seasonal influenza, as quantified by the fact (shown in the two figures below) that the HA1 proteins of most H3N2 and H1N1 strains sequenced over the last year are either identical or within one amino-acid mutation of an HA in our library:

For interactive Nextstrain phylogenetic trees showing the strains included in the library colored by their subclade identities, see the following links:

• H3N2
• H1N1

Human sera tested¶

The sera come from a variety of sources as summarized in the table below; see here for the full details on the individual sera.

H3N2 results¶

Median and per-serum titers¶

The simplest overview of the results is in the interactive figure below, which shows the median (points) and interquartile range (shaded area) titers for all sera against all strains. This figure has many interactive options described in its legend; please use them, and for extensive exploration open the plot in a new tab with the link at the end of the legend.

The above plot shows that several variants of subclade K have reduced titers, most notably variants with mutations at site 223 (eg, V223I) or 222 (eg, R222K). Both of these mutations are in antigenic region D, see Liu et al (2026), medRxiv for discussion of the reason that they may have such a substantial impact in subclade K.

Another view of the same data is in the figure below, which now shows individual lines for each serum rather than the interquartile range. This figure is much busier, but if you interactively mouse over the lines you can trace the titers for individual sera and visualize the remarkable variability in both overall and strain-specific titers.

Titers projected on phylogenetic tree¶

A helpful way to examine the titers is on an interactive Nextstrain phylogenetic tree.

The figure below shows a Nextstrain tree (available as a standalone link here) of the HA proteins in the library colored by the median titer against all sera. The sidebar provides numerous options to change the coloring, including coloring by HA genotype or other properties. You can also click on strains for more details, and use the Measurements panel below the tree to break down the titers in various ways.

Note that you can also use the Scatter option to the left of the tree to stratify titers against strains with specific mutations, such as in this view showing the median titers against strains with different amino-acid identities at site 223.

Pre- and post-vaccination titers¶

For the VIDRL cohort, we have pre- and post-vaccination titers for adults who received an egg-based (Fluzone) or cell-based (Flucelvax) vaccine, elderly individuals who received an egg-based (Fluad) vaccine, and children who received an egg-based (Fluzone) vaccine. The days after vaccination at which the post-vaccination sera were collected are in the table above. The table also shows the fraction vaccinated in the prior year for each group; these fractions differ among groups so comparisons of absolute vaccine responses among groups who received different vaccines are potentially confounded since prior-year vaccination can influence current-year vaccine responses (see Cowling et al (2024)).

The figure below shows the pre- and post-vaccination titers for each group to all of the strains:

Here is another figure that shows the fold change in titer against each strain after vaccination:

For plots that show the same data but at the level of individual sera rather than interquartile ranges, see here and here.

Key mutations on HA structure¶

The titer data above show an effect of mutations at sites 223 and 222 on the titers to subclade K variants. The interactive figure below shows those sites on an interactive HA structure in purple alongside other mutations in subclade K relative to the 2025-2026 vaccine strain in red. You can use the dropdown below the figure to also color the HA by antigenic region (you will see that 223 and 222 are in antigenic region D whereas most prior mutations in subclade K were in regions A and B); you can also further interact with the structure using all of the options in the Mol* viewer documented here.

H1N1 results¶

Median and per-serum titers¶

The interactive figure below shows the median and interquartile range titers for all sera against all strains.

The above figure shows that most D.3.1.1 strains have reduced titers relative to D.3.1 (the cell-based strain in the 2026-2027 Northern Hemisphere vaccine), with titers especially reduced for strains with mutations at site 155 (eg, G155E) or to a lesser extent 157 (eg, S157L).

Notably, the impact of G155E is greatly accentuated for individuals 15-25 years of age compared to other individuals; you can see this by using the sliders below the plot to subset just on sera from individuals in that age range.

Below is a busier figure showing every individual serum as a line; by mousing over lines you can again see that G155E most affects a subset of sera.

Sera stratified by impact of G155E¶

As noted in the section above, the effect of mutations like G155E and S157L is especially profound for sera from a subset of individuals, disproportionately those aged ~15 to 25 years.

To visualize that more clearly, the interactive figure below stratifies sera by their relative titers to two different strains. The initial view separates sera where the titer to strain D.3.1.1:G155E is 2-fold or more lower than the titer to D.3.1.1 (green) versus sera where it is not (orange). As can be seen, G155E has a very strong effect for a subset of sera, and these sera are enriched for individuals aged ~15 to 25 years. You can use the interactive options to also stratify sera by relative titers to other pairs of strains.

Note the comparable plot for H3N2 does not show similarly dramatic stratification for the largest-antigenic-effect mutations.

Titers projected on phylogenetic tree¶

The figure below shows a Nextstrain tree (available as a standalone link here) of the HA proteins in the library colored by the median titer against all sera.

Note that you can also use the Scatter option to the left of the tree to stratify titers against strains with specific mutations, such as in this view showing the median titers against strains with different amino-acid identities at site 155.

Pre- and post-vaccination titers¶

The figure below shows the pre- and post-vaccination titers for each group to all of the strains:

Here is another figure that shows the fold change in titer against each strain after vaccination:

For plots that show the same data but at the level of individual sera rather than interquartile ranges, see here and here.

Key mutations on HA structure¶

The titer data above show an effect of mutations at site 155, and to a lesser extent 157, on the titers. The interactive figure below shows those sites on an interactive HA structure in indigo alongside other mutations in subclade D.3.1.1 relative to the 2026-2027 vaccine strain in red. You can use the dropdown below the figure to also color the HA by antigenic region (you will see that 155 and 157 are in antigenic region Sa).

Titers to older and egg-based vaccine strains¶

The above plots show only the titers to recently circulating strains and recent cell-based vaccine strains. However, we also measured neutralization titers against some older vaccine strains including egg-based ones. The reason that they are not shown in the above plots is that they sometimes "blow out" the scale as adults can have very high titers to older strains, and we consistently measure higher titers to egg-based vaccine strains (which tend to have non-natural egg adaptation mutations) than cell-based vaccine strains as described in Kikawa et al (2026).

We therefore have created separate plots showing the titers against just the vaccine strains, including older strains and egg-based vaccine strains. The overall titers against these H3N2 and H1N1 strains are shown below:

Other plots for the vaccine strains are at the following links:

• H3N2:

  • Titers for individual sera against all tested egg- and cell-based vaccine strains
  • Pre- and post-vaccination titers to all tested egg-and cell-based vaccine strains, interquartile range
  • Pre- and post-vaccination titers to all tested egg- and cell-based vaccine strains, individual sera
  • Fold change post-vaccination to all tested egg- and cell-based vaccine strains, interquartile range
  • Fold change post-vaccination to all tested egg- and cell-based vaccine strains, individual sera

• H1N1:

  • Titers for individual sera against all tested egg- and cell-based vaccine strains
  • Pre- and post-vaccination titers to all tested egg-and cell-based vaccine strains, interquartile range
  • Pre- and post-vaccination titers to all tested egg- and cell-based vaccine strains, individual sera
  • Fold change post-vaccination to all tested egg- and cell-based vaccine strains, interquartile range

Contributors¶

This study was led by Caroline Kikawa, Andrew Butler, John Huddleston, and Jesse Bloom.

Contributors include Heidi Peck and Ian Barr (Doherty Institute, Australia) and Shidan Tosif (The Royal Children's Hospital, Australia); Janet Englund and Kirsten Lacombe (Seattle Children's Hospital); Alex Greninger (University of Washington); Michael Busch, Marion Lanteri, Mars Stone, and Bryan Spencer (Vitalant Research Institute and the American Red Cross); Sam Turner and Derek Smith (University of Cambridge); and Scott Hensley (University of Pennsylvania).