Abstract 1122: Pan-cancer clinical and molecular landscape of MTAP deletion in nationwide and international comprehensive genomic data

Abstract Methylthioadenosine phosphorylase (MTAP) deletion in cancer cells induces vulnerability to protein arginine methyltransferase 5 (PRMT5) inhibition, offering a synthetic lethal target. Early-phase clinical trials have shown promising efficacy of PRMT5 inhibitors in MTAP-deleted tumors, but the full spectrum of patients with MTAP-deleted tumors and their molecular characteristics remain unclear. We report a pan-cancer landscape of MTAP-deleted tumors using comprehensive genomic profiling (CGP) data of 51, 828 cases in the Center for Cancer Genomics and Advanced Therapeutics (C-CAT), covering 99.7% patients who underwent CGP in Japan. Findings were validated using TCGA (n = 9, 896) and GENIE (n = 178, 034) data. MTAP deletion was observed in 9.6% of the C-CAT cohort, 8.9% of TCGA, and 6.6% of GENIE. A machine learning prediction model using pre-CGP clinical data, combined with SHapley Additive exPlanations (SHAP) analysis, identified clinical factors associated with MTAP deletion: sex, age, cancer type, and time intervals from cancer diagnosis to CGP. In C-CAT, MTAP-deleted patients were predominantly male (56.0% vs 47.8%, p < 0.0001) and slightly older (mean: 62.4 vs 59.8 years, p < 0.0001). MTAP deletion rates were high in pleura (33.1%), bladder/urinary tract (23.8%), brain (19.0%), pancreas (18.4%), biliary tract (15.6%), and lung (14.3%) cancers, but low in prostate (1.2%), bowel (1.7%), cervix (3.6%), uterus (4.0%), ovary/fallopian tube (4.3%), and breast (4.9%) cancers. The TCGA and GENIE cohorts showed similar trends, indicating the broad applicability of therapeutic strategies targeting MTAP deletion across different demographic groups. MTAP-deleted tumors demonstrated a shorter time to standard therapy failure in pan-cancer C-CAT data (median: 380.0 vs 567.0 days, HR: 1.34 [1.30-1.38], p < 0.0001) and worse overall survival in TCGA (median: 26.7 vs 93.0 months, HR: 2.36 [2.14-2.61], p < 0.0001). Comprehensive molecular analyses revealed that CDKN2A/B loss was an almost necessary, but not sufficient, condition for MTAP deletion. MTAP deletion was more frequent in KRAS-mutant pancreatic cancer (19.8% vs 8.9%), suggesting a cooperative role of MTAP and KRAS and synergistic effects of PRMT5 and KRAS inhibitors. MTAP deletion was less common in RB1-mutated tumors across multiple cancer types, including pancreatic (7.6% vs 18.7%), lung (2.5% vs 16.1%), and biliary (5.4% vs 16.0%) cancers. In lung adenocarcinoma, MTAP deletion was found in 22.8% of EGFR-mutated tumors, 25.0% of ALK-translocated tumors, and 20.8% of ROS1-translocated tumors, indicating that MTAP deletion is not mutually exclusive with druggable mutations and that PRMT5 inhibitors might benefit patients with known driver mutations. Our findings define the characteristics of MTAP-deleted cancers, providing a basis for refinement and expansion of synthetic lethal strategies in precision oncology. Citation Format: Hiroaki Ikushima, Kousuke Watanabe, Aya Shinozaki-Ushiku, Katsutoshi Oda, Hidenori Kage. Pan-cancer clinical and molecular landscape of MTAP deletion in nationwide and international comprehensive genomic data [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1122.