Editors' ChoiceCancer

The more (mutations), the better

See allHide authors and affiliations

Science Translational Medicine  30 Jan 2019:
Vol. 11, Issue 477, eaaw5320
DOI: 10.1126/scitranslmed.aaw5320

Abstract

High somatic tumor mutational burden predicts response to immune checkpoint inhibitors across multiple cancers.

All tumor cells harbor somatic mutations that contribute to their malignant phenotype. These somatic mutations can lead to the production of mutated proteins with aberrant or diminished activity. At the same time, mutated proteins can be recognized as “nonself” neo-antigens by the adaptive immune system. Although the accumulation of mutations in tumor cells can provide a selective advantage by increasing genetic diversity and adaptability, it also poses a tremendous risk for the tumor cells, as they become more easily recognizable by the immune system. The tumor mutational burden (TMB) has been associated with better responses to immune checkpoint inhibitors (ICI) in melanoma and non–small cell lung cancer (NSCLC), two tumor types with high prevalence of mutations. In a recent study, Samstein et al. established that higher somatic TMB also predicts clinical benefit across diverse human cancers with distinct average TMBs.

In this study, the authors profiled DNA from cancer patients using the U.S. Food and Drug Administration–approved Integrated Mutation Profiling of Actionable Cancer Targets (IMPACT) assay. The analysis included 1662 patients treated with at least one dose of ICI therapy [anti–cytotoxic T lymphocyte antigen 4 (CTLA-4), anti–programmed cell death protein 1/programmed cell death ligand 1 (PD-1/PD-L1), or combination of anti–CTLA-4 and anti–PD-1/PD-L1], the largest cohort of patients treated with ICI studied to date. Since TMB is variable across different cancers, each tumor type was stratified by TMB decile, which revealed an association between a higher number of mutations and improved overall survival. Similar trends were observed in individual cancers, with the exception of glioma. As expected, the TMB cutoff that could predict response to ICI for each tumor type was variable, suggesting that there is no universal value. The association between high TMB and survival was still statistically significant after removing the melanoma and NSCLC cohorts, demonstrating that the association was not due solely to these cancers. Furthermore, there was no association between high TMB and survival in patients not treated with ICI, highlighting the predictive value of high TMB for ICI therapies. This study takes a giant step toward defining biomarkers that can predict immunotherapy response. Additional studies integrating TMB with other parameters shown to also associate with response to ICI will further refine predictions for improved patient selection. Lastly, studies in additional cohorts of patients treated with ICI will help define cancer-specific biomarkers.

Highlighted Article

View Abstract

Navigate This Article