Purpose: Tumor mutational burden (TMB) is emerging as a promising biomarker in immune checkpoint inhibitor therapy. Although whole-exome sequencing (WES) is the gold standard to quantify TMB, most TMB measurements in routine diagnostic applications were done by gene panels due to their lower cost and technical accessibility. Therefore, data is lacking for WES derived TMB in several major tumors. However, panel derived TMB can be misleading because research found its correlation with WES derived TMB was compromised when a subset of samples with high TMB values were excluded from the dataset. In this study, we aim to establish the landscape of WES derived TMB in six tumor types in Chinese patients, test the association between WES-TMB and patient survival in immunotherapy, and compare the performance between WES-TMB and panel-TMB.

Methods: We recruited 357 patients who had received immune checkpoint inhibition (ICI) therapy, representing six types of cancers (see Results for the list). We analyzed the tumors and patient-matched blood samples using WES and called somatic mutations of the patients. We defined our WES-TMB as the total number of somatic nonsynonymous mutations in the tumor exome. In each of the tumor types tested, we assigned the patients with the upper tertile TMB values as the high-TMB group and the remaining two thirds of patients as the low-TMB group.

Results: The distribution of WES-TMB medians are highly variable among different tumors, ranging from 2.71 (cholangiocarcinoma), 2.97 (nervous system tumor), 3.69 (gastric cancer), 4.31 (hepatocellular carcinoma), 4.64 (colorectal cancer), to 5.67 (lung cancer) mutations/Mb. This trend is almost the same as other TMB studies done with panel sequencing, despite the values derived from WES tend to be lower. Additionally, TMB was significantly associated with overall survival in colorectal patients receiving ICI. The high-TMB group also showed better overall survival in hepatocellular carcinoma and gastric cancer.

Conclusion: TMBs derived from panel are overestimated compared to that derived from WES although that doesn't affect the rank of TMBs in different cancer types. It is rational to define TMB subgroups by percentile within different cancer types. These results support TMB as a biomarker to predict clinical benefit to ICI therapy in colorectal cancer.

Citation Format: Fei Xu, Jinwang Wei, Xiaochen Zhang, Jun Liu, Chun Dai, Guan Wang, Qiang Xu. Tumor mutational burden estimated by whole-exome sequencing across multiple cancer types [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 398.