TMB is promising as a biomarker for the treatment with ICIs. However, because it is difficult to apply whole exon-based TMB analysis into the current practice, clinically applicable simple methods for estimating TMB are needed. The aim of this study was to evaluate TMB based on a small-sized targeted NGS as a biomarker for ICIs.
Since March 2015 to April 2017, a total of 2243 NSCLC patients (pts) from 244 institutions were enrolled in our large-scale nationwide genome screening network (LC-SCRUM-Japan). Tumor samples were subjected to a 260 kb-sized NGS assay, Oncomine Comprehensive Assay (OCA) ver.1, targeting 143 cancer-related genes. TMB, number of somatic mutations/Mb, was assessed from the results of OCA ver.1 assay, and a cut-off point of TMB to predict response to ICIs was determined by ROC curve. Clinico-genomic database of LC-SCRUM-Japan was utilized for this analysis.
470 NSCLCs, consisting of 359 adenocarcinomas, 73 squamous cell carcinomas and 38 others, were evaluated in this study. The median number of mutations was 11.5/Mb (range, 0-130.8/Mb). We defined number of mutations ≧15.4/Mb as high TMB and <15.4/Mb as low TMB. High TMB was observed 34.3% (123/359) in adenocarcinoma and 41.1% (30/73) in squamous cell carcinoma. The response rate was higher in pts with high TMB than in those with low TMB (13.3% [23/173] vs. 5.7% [17/297], p = 0.0059). The durable clinical benefit (DCB; complete response, partial response or stable disease that lasted >6 months) rate also tended to be higher in pts with high TMB than in those with low TMB (17.3% [30/173] vs. 11.8% [35/297], p = 0.0980). The progression-free survival (PFS) was not significantly different between the high and low TMB pts (median PFS, 4.4 vs. 3.3 months, p = 0.1401).
TMB estimated by OCA ver.1 seemed to be correlated with response rate and DCB but not PFS in NSCLC pts treated with ICIs, suggesting a limitation of TMB estimation by this small-sized targeted NGS as a biomarker of ICIs. Optimal TMB estimation to predict the efficacy of ICIs are warranted.
Clinical trial identification
Legal entity responsible for the study
National Cancer Center Hospital East.
Has not received any funding.
All authors have declared no conflicts of interest.