The new identified biomarkers determine sensitivity to immune check-point blockade therapies in melanoma.

Immune checkpoint blockade (ICB) therapy has achieved remarkable clinical benefit in melanoma. However, our understanding of biomarkers that predict response to ICB remained obscure. Here we systematically analyzed the association between somatic mutations profile and clinicopathologic information from 336 melanoma patients treated by ICB (CTLA-4/PD-1). We identified eight new significantly mutated genes including COL5A1, SEMA3E, COL28A1, DGKG, RAPGEF5, GLDN, NCF2 and RCAN2. A mutational signature featured by enrichment of T > C mutations was identified to be associated with immune resistance (logistic regression model, OR, 2.59 [95%CI, 1.07 to 7.00], P = .043). High neoantigen quality was associated with prolonged immunotherapy survival (log-rank test, P = .009). This association remained significant after controlling for age, gender, stage and hypermutation (Cox proportional hazards model, HR, 0.56 [95%CI, 0.38 to 0.82], P = .003). Our findings shed new insights on biomarkers that are useful to predict melanoma patients who may benefit from ICB treatment; however, these biomarkers need to be validated in future studies.