Identification of somatic mutations and neoantigens to predict development of autoimmune adverse events to immune therapy in melanoma.

Abstract

Listen

Translate article

19 Background: Immunotherapy has quickly altered the treatment paradigm for advanced melanoma. Prior studies using whole exome sequencing (WES) in immunotherapy have shown limited correlation with clinical response to immunotherapy, but not an evaluation of autoimmune side effects. Understanding mechanisms of immune-related adverse events (irAE) at a genomic level may lead to improved rational treatment design and better patient selection. Methods: DNA was extracted from 87 tumor samples from 49 patients with metastatic melanoma treated with ipilimumab (anti-CTLA-4) or with nivolumab or pembrolizumab (anti-PD-1) immunotherapy. WES was performed with exome capture on an Illumina HiSeq sequencer. Statistical analysis was performed with Chi-squared testing. Computational neoantigen analysis has been performed on the sequencing data. Results: 27 patients with advanced or metastatic melanoma were treated with only ipilimumab, 11 with only anti-PD1, and 11 patients were treated with both anti-CLTA-4 and anti-PD1 either sequentially or in combination. irAE evaluated included cutaneous, colitis, endocrinopathies, hepatitis, and pneumonitis. In 10 patients who developed colitis, 14 mutated genes were enriched in their tumor samples (p=0.01-0.046) compared with patients without colitis. Analyzing all patients with any irAE versus no irAE treated with anti-CTLA-4, anti-PD1, or both (either sequential or combination) revealed 7 mutated genes enriched in patient samples who developed irAE (p=0.02-0.05). Preliminary neoantigen analysis has identified possible irAE associated neoantigens and further analysis of additional samples are ongoing. Overall mutational burden of the tumor samples did not correlate with development of irAE. Conclusions: Autoimmune side effects are frequent and often serious side effects of immunotherapy for melanoma. Genetic analysis may provide insight in prediction of these side effects and help direct therapeutic decisions.