Simon de Jong1, Laura van Groeningen1, Nikoloz Nemanichvili1, Karen de Haan1, Sarah van Leeuwen1, Haye Nijhuis1, Coco Fuhri Snethlage2, Dirk Eggink1,3, Alvin Han1, Max Nieuwdorp2, Nordin Hanssen2, Colin Russell1, Katina Hulme1
1Department of Medical Microbiology and Infection Prevention, Amsterdam University Medical Center, University of Amsterdam, Amsterdam, The Netherlands
2Department of Vascular Medicine, Amsterdam UMC, Amsterdam, The Netherlands
3Centre for Infectious Disease Control, National Institute for Public Health and the Environment, Bilthoven, The Netherlands
Presenting author: Katina Hulme | email: k.d.hulme@amsterdamumc.nl
Background
Annual influenza vaccination is recommended in type 1 diabetes (T1D), yet whether T1D impairs vaccine responses remains unclear, with reports ranging from adequate protection to high rates of non-response. Vaccine responses are also shaped by prior influenza immunity, which differs by birth cohort. Here, we examined how vaccine-associated antibody binding varies with birth cohort in T1D.
Methods
In a cross-sectional study, a Luminex panel of 28 haemagglutinin ectodomains spanning A/H3N2 (2000–2022), A/H1N1 (2000–2009) and A/H1N1pdm09 (2009–2022) measured HA-binding titers in 500 individuals with T1D. Titers were compared between individuals vaccinated in the previous season and unvaccinated individuals, stratified by birth-cohort decade. Vaccination effects were estimated as regression coefficients representing age-adjusted differences in log2 titres between groups.
Results
HA-binding titers were generally higher in vaccinated than unvaccinated individuals, but the difference depended on birth cohort and subtype. For strains circulating from 2000-2022, the estimated effect of prior season vaccination on log2 binding titer, pooled across strains, increased from 0.61 (0.38-0.84, 95% CI) and 0.56 (0.11-1.02) for the 1960–1969 cohort to 1.23 (1.00-1.45) and 2.82 (2.52-3.12) in the 2000–2009 cohort, for A/H3N2 and A/H1N1pdm09, respectively. Restricting to strains contemporaneous with vaccination (2019–2022), the estimated difference in log2 titer by vaccination status for A/H3N2 fell to 0.16 (-0.24-0.57) in the 1960–1969 cohort but persisted at 1.14 (0.69-1.58) in the 2000–2009 cohort.
Conclusions
These data may help explain conflicting reports: vaccine-associated binding in T1D depends on birth cohort and on the antigenic distance between the vaccine and previously circulating strains. Strikingly, older individuals showed little-to-no vaccine-associated boost to contemporary strains despite ample assay headroom, contrary to the general observation that vaccination most strongly boosts contemporary-strain titres. Whether this is a T1D effect, an age-related deficit, or reflects differing prior-vaccination histories will require longitudinal study with a non-diabetic comparator.