Genomic signatures contributing to high tumour mutational burden (TMB-H) independent from mismatch-repair deficiency (dMMR) or microsatellite instability-high (MSI-H) status are not well studied. We aimed to characterise molecular features of microsatellite stable (MSS) TMB-H gastrointestinal tumours.

Methods

Molecular alterations of 48 606 gastrointestinal tumours from Caris Life Sciences (CARIS) identified with next-generation sequencing were compared among MSS–TMB-H, dMMR/MSI-H, and MSS–TMB-low (L) tumours, using χ² or Fisher’s exact tests. Antitumour immune response within the tumour environment was predicted by analysing the infiltration of immune cells and immune signatures using The Cancer Genome Atlas database. The Kaplan-Meier method and the log-rank test were used to evaluate the impact of gene alterations on the efficacy of immune checkpoint inhibitors in MSS gastrointestinal cancers from the CARIS database, a Memorial Sloan Kettering Cancer Center cohort, and a Peking University Cancer Hospital cohort.

Findings

MSS–TMB-H was observed in 1600 (3·29%) of 48 606 tumours, dMMR/MSI-H in 2272 (4·67%), and MSS–TMB-L in 44 734 (92·03%). Gene mutations in SMAD2, MTOR, NFE2L2, RB1, KEAP1, TERT, and RASA1 might impair antitumour immune response despite TMB-H, while mutations in 16 other genes (CDC73, CTNNA1, ERBB4, EZH2, JAK2, MAP2K1, MAP2K4, PIK3R1, POLE, PPP2R1A, PPP2R2A, PTPN11, RAF1, RUNX1, STAG2, and XPO1) were related to TMB-H with enhanced antitumour immune response independent of dMMR/MSI-H, constructing a predictive model (modified TMB [mTMB]) for immune checkpoint inhibitor efficacy. Patients with any mutation in the mTMB gene signature, in comparison with patients with mTMB wildtype tumours, showed a superior survival benefit from immune checkpoint inhibitors in MSS gastrointestinal cancers in the CARIS cohort (n=95, median overall survival 18·77 months [95% CI 17·30–20·23] vs 7·03 months [5·73–8·34]; hazard ratio 0·55 [95% CI 0·31–0·99], p=0·044). In addition, copy number amplification in chromosome 11q13 (eg, CCND1, FGF genes) was more prevalent in MSS–TMB-H tumours than in the dMMR/MSI-H or MSS–TMB-L subgroups.

Interpretation

Not all mutations related to TMB-H can enhance antitumour immune response. More composite biomarkers should be investigated (eg, mTMB signature) to tailor treatment with immune checkpoint inhibitors. Our data also provide novel insights for the combination of immune checkpoint inhibitors and drugs targeting cyclin D1 or FGFs.