Predicting influenza H3N2 vaccine efficacy from evolution of the dominant epitope

Abstract

We predict vaccine efficacy with a measure of antigenic distance between influenza A(H3N2) and candidate vaccine viruses based on amino acid substitutions in the dominant epitopes. In 2016-2017, our model predicts 19% efficacy compared to 20% observed. This tool assists candidate vaccine selection by predicting human protection against circulating strains.\n\n40-word summary of main pointOur pepitope model predicts the ability of the influenza vaccine to reduce the A(H3N2) disease attack rate, with an r^2=0.77. This fast, sequence-based method compliments strain-to-strain antigenic comparisons from ferret models and provides antigenic comparisons for all circulating sequences.