Extended Data Fig. 2: Biomarker analysis of pretreatment tumor tissue in patients with microsatellite stable metastatic colorectal cancer (n=38).

(a) Clinical benefit by tumor mutational burden (TMB) at baseline (pretreatment). TMB was assessed by whole-exome sequencing (WES) using Personalis ImmunoID NeXT™ platform v1.0. TMB was defined as tumor-specific nonsynonymous somatic variants (and insertion-deletions)/megabase (Mb). Complete response (CR) or partial response (PR) and stable disease (SD) for at least 12 weeks was defined as per Response Evaluation Criteria in Solid Tumors version 1.1. Statistical analysis was performed using Kruskal-Wallis (ANOVA) nonparametric one-way test (p=0.7290, Kruskal-Wallis H=0.6321), followed by post-hoc Dunn test for adjusted multiple comparisons (CR/PR vs SD: mean rank difference= 1.111, z=0.2449, adjusted p=1; CR/PR vs PD: mean rank difference= 3.747, z=0.7503, adjusted p-value=1; SD vs PD: mean rank difference= 2.636, z=0.6199, adjusted p-value=1). Multiplicity adjusted p-values of 1 are indicated as p>0.9999 on the graph. Mut/Mb, mutations per megabase; PD, progressive disease.