Genomic epidemiology of meningococcal vaccine antigens in the United Kingdom

Vaccination against Neisseria meningitidis remains the optimal means to prevent invasive meningococcal disease (IMD), an important cause of septicaemia and meningitis disproportionately affecting infants and young children. Subcapsular protein-based vaccines were developed to offer broad protection against genetic lineages of meningococci responsible for endemic capsular group B IMD. As meningococci solely colonise the human oropharynx, the inherent variation in surface proteins potentially limits their efficacy as vaccine antigens. With increasing use of these vaccines worldwide, high-throughput genomic methods were required to analyse meningococcal protein diversity. The aims of this study were to determine the diversity and distribution of proteins in subcapsular protein-based vaccines Bexsero® and Trumenba® amongst meningococcal whole genome sequences (WGS) from disease and carriage isolates from the UK, 2010-2017. Using the principles underlying multilocus sequence typing, two novel genomic typing schemes were established on PubMLST.org Neisseria website. The Bexsero® Antigen Sequence Typing (BAST) and Outer Membrane Vesicle peptide Typing (OMVT) schemes were used to analyse ~3500 disease and ~650 carriage meningococcal WGS. Using this comprehensive, national dataset, non-random distribution of BASTs and Bexsero® antigens was demonstrated, with a few variants occurring at high frequency. Strong, non-overlapping associations between OMVT clusters, BASTs, cc and genogroups were identified in disease and carriage meningococci, likely driven by host immune and metabolic selection. Understanding the population structure of subcapsular and outer membrane proteins has important implications when selecting antigens and variants for inclusion in existing or novel protein-based vaccines, for both meningococci and other Neisseria species, including gonococci. The development of BAST typing coincided with Bexsero® vaccine introduction into the UK infant immunisation programme and detailed pre-vaccine genomic surveillance provided estimations of vaccine coverage (58.3%-60.3%) and identified low incidence of exact antigenic variants. Both typing methods are rapid, scalable and open-source, with the potential to optimise vaccine development, reformulation and implementation worldwide.