Predicting immunogenic tumour mutations by combining mass spectrometry and exome sequencing

A combination of genome-wide exome and transcriptome analysis, mass spectrometry and computational structural modelling are used here to identify immunogenic neo-antigens in two mouse tumour cancer cell lines; mice vaccinated with predicted immunogenic peptides yielded therapeutically useful cytotoxic T-lymphocyte responses. Immunogenic tumour mutations predicted Mahesh Yadav et al . use a combination of genome-wide exome and transcriptome analysis, mass spectrometry and computational structural modelling to identify immunogenic neo-antigens in two mouse tumour cell lines. Mice vaccinated with predicted immunogenic peptides yielded therapeutically useful cytotoxic T-lymphocyte responses. Human tumours typically harbour a remarkable number of somatic mutations 1 . If presented on major histocompatibility complex class I molecules (MHCI), peptides containing these mutations could potentially be immunogenic as they should be recognized as ‘non-self’ neo-antigens by the adaptive immune system. Recent work has confirmed that mutant peptides can serve as T-cell epitopes 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 . However, few mutant epitopes have been described because their discovery required the laborious screening of patient tumour-infiltrating lymphocytes for their ability to recognize antigen libraries constructed following tumour exome sequencing. We sought to simplify the discovery of immunogenic mutant peptides by characterizing their general properties. We developed an approach that combines whole-exome and transcriptome sequencing analysis with mass spectrometry to identify neo-epitopes in two widely used murine tumour models. Of the >1,300 amino acid changes identified, ∼13% were predicted to bind MHCI, a small fraction of which were confirmed by mass spectrometry. The peptides were then structurally modelled bound to MHCI. Mutations that were solvent-exposed and therefore accessible to T-cell antigen receptors were predicted to be immunogenic. Vaccination of mice confirmed the approach, with each predicted immunogenic peptide yielding therapeutically active T-cell responses. The predictions also enabled the generation of peptide–MHCI dextramers that could be used to monitor the kinetics and distribution of the anti-tumour T-cell response before and after vaccination. These findings indicate that a suitable prediction algorithm may provide an approach for the pharmacodynamic monitoring of T-cell responses as well as for the development of personalized vaccines in