Landscape of homologous recombination-related (HRR) genes mutations in colon cancer

e15525 Background: Homologous recombination deficiency (HRD) might serve as a biomarker of immunotherapy and chemotherapy response. In addtion, HRD is associated with sensitivity to PARP inhibitor (PARPi) therapy. The HRD phenotype has been defined as the presence of a non-silent somatic mutation in homologous recombination-related (HRR) genes. Here, we focused on mutational landscape of HRR genes and the relationship between HRR-related genes and clinical outcomes in colon cancer. Methods: We prospectively sequenced 784 Chinese patients with colon cancer using next-generation sequencing techniques with 808 cancer-related genes. 290 patients with somatic mutation from the TCGA project were included. The 15 HRR genes were selected on clinical evidence of sensitivity to PARP inhibitors and immunotherapy. TCGA datasets were performed to identify the correlations between HRR gene mutations and clinical outcomes. Results: In the Chinese cohort, 11.5% (90/784) patients exhibited deleterious somatic mutation in HRR genes, including ATM (3.3%), BRCA2 (2.4%), CDK12 (2.4%), BRCA1 (1.0%), BRIP1 (1.0%), CHEK2 (0.5%), BARD1 (0.4%), RAD54L (0.4%), PALB2 (0.3%), and RAD51B (0.3%). In TCGA cohort, 28.6% (83/290) patients had at least one deleterious somatic mutation in HRR genes, including ATM (14.8%), BRCA2 (9.0%), CDK12 (7.9%), BRIP1 (3.8%), BRCA1 (3.4%), FANCL (3.1%), CHEK2 (2.4%), PALB2 (2.1%), PPP2R2A (2.1%), RAD54L (2.1%), CHEK1 (1.7%), BARD1 (1.4%), RAD51C (1.4%), RAD51B (1.0%), and RAD51D (0.3%). Overall, the mutational frequency of HRR genes in the Chinese cohort was lower than that in the TCGA cohort (p < 0.001). In the TCGA cohort, colon patients with HRR-mut had significantly higher tumor mutational burden and incidence of microsatellite instability high and enhanced immune activity than patients with HRR-wt. Furthermore, In the MSK-IMPACT cohort treated by immune checkpoint inhibitor, patients with HRR-mut had significantly better OS (p = 0.009) and increased immune activities, such as infiltration of cytotoxic cells (p < 0.05) and exhausted CD8+ T cells (p < 0.01), and increased the IFN-γ scores (p < 0.05) than patients with HRR-wt. Conclusions: Our data suggest that detection of somatic mutations in HRR genes might contribute to identify patients who might benefit from immune checkpoint blockade therapy and PARP inhibitors. In addition, the molecular features of HRR provide new opportunities to predict the tumor response to multiple treatments. Exploring other biomarkers of HRD to predict the response to PARPi and immunotherapeutic in colon cancer is necessary.