Machine learning-driven identification of serotype-independent pneumococcal vaccine candidates using samples from human infection challenge studies

Identifying conserved, immunogenic proteins that confer protection against Streptococcus pneumoniae colonisation could enable development of serotype-independent vaccines. We analysed baseline samples from 86 healthy adults experimentally challenged with pneumococcal serotypes 6B or 15B to investigate whether immune responses to 75 universally expressed pneumococcal proteins associated with protection against colonisation. We measured serum IgG using a novel Luminex assay and cytokine responses from peripheral blood mononuclear cells following antigen stimulation. No individual IgG or cytokine marker correlated significantly with protection in univariate analysis. However, machine learning identified IgG responses to PdB, SP1069, and SP0899 as predictive of protection. MCP-1 responses to SP1069 and SP0899, and IL-17 production in response to SP0648-3 also correlated with protection. Elevated baseline IFN-γ, RANTES, and anti-protein IgG correlated with lower colonisation density. We highlight SP1069 and SP0899 as potential serotype-independent vaccine candidates and demonstrate the utility of machine learning to identify immune correlates of protection.