In the literature: December 2018

Abstract

The current development of immune checkpoint modulatory treatments has shown durable responses in the treatment of multiple cancer types.1 However, predictive biomarkers beyond PD-1 ligand 1 (PD-L1) expression and high microsatellite instability (MSI-H) to stratify patients and identify those who could benefit of these therapies are needed. In this sense, a recent study published in Science by Cristescu et al 2 describes the potential usefulness of combining the tumour mutational burden (TMB) and the T cell-inflamed gene expression profile (GEP) to jointly predict clinical response to pembrolizumab. Both PD-L1 and the GEP represent a T cell-inflamed tumour microenvironment (TME), whereas TMB and MSI-H are indirect measures of tumour antigenicity generated by somatic tumour mutations. In the mentioned study, TMB was evaluated by whole-exome sequencing (WES) of germline and tumour DNA, and the T cell-inflamed GEP was analysed by targeted gene expression profiling of tumour RNA (with the NanoString platform) from formalin fixed, paraffin-embedded pretreatment samples. To assess the individual and joint clinical utility of TMB and GEP, patients were stratified in four biomarker-defined clinical response groups based on predefined cut-offs for TMB and GEP. Objective response rates were strongest in patients with GEPhi TMBhi (37%–57%), moderate in those with GEPhi TMBlo (12% to 35%) and GEPlo TMBhi (11%–42%) and reduced or absent in those with GEPlo TMBlo (0%–9%). Additionally, longer PFS were seen in patients with higher levels of both TMB and GEP. Findings were comparable when TMB and PD-L1 expression were jointly assessed. Interestingly, more than 300 patient samples with advanced solid tumours and melanoma across 22 tumour types from four KEYNOTE clinical trials were included and analysed. Samples were assessed separately in four cohorts: one discovery, one pan-tumour validation and …