Molecular landscape and therapeutic vulnerability of RRAS- and RRAS2-mutant solid tumors.

3065 Background: The RRAS subfamily of small GTPases shares considerable sequence homology with the canonical RAS oncoproteins KRAS , NRAS , and HRAS. Mutations in RRAS and RRAS2 that are homologous to KRAS hotspot mutations promote transformation in vitro . Most diagnostic sequencing panels do not profile the RRAS subfamily, and therefore, the prevalence and clinical relevance of these mutations and their treatment have not been fully established. Methods: Based on a clinical targeted DNA sequencing assay (MSK-IMPACT), an institutional cohort of 51,040 solid tumor cases prospectively sequenced between 2016-2024 was analyzed to identify RRAS / RRAS2 mutations. Lung cancers were excluded and analyzed in a separate study. Hotspot mutations in RRAS/RRAS2 were determined based on homology to KRAS hotspot mutations and/or previous literature demonstrating potential oncogenicity (hotspot RRAS : G38, G39 and Q87; hotspot RRAS2 : G23, G24, A70T, Q72 and in-frame insertions in G23/G24). The sensitivity of cells harboring RRAS and RRAS2 -mutations to the clinically active pan-RAS inhibitor RMC6236 was examined in vitro and in vivo . Western blotting was utilized to determine changes in protein expression and activation. Results: Among the 51,040 cases analyzed, hotspot RRAS and RRAS2 mutations were detected in 6 (0.01%) and 270 (0.5%) patients, respectively. Hotspot RRAS mutations were seen in various cancer types, and most had other mitogenic drivers (4/6). Amongst tumors with hotspot RRAS2 mutations, the most common cancer types included endometrial (n = 172), ovarian (n = 27), and germ cell tumors (n = 26); these variants were seen in 5%, 0.9%, and 5% of these respective patient populations. Most endometrial, ovarian, and germ cell tumors lacked other K/H/NRAS mutations (83%, 89%, 85%, respectively). Other tumor types with hotspot RRAS2 mutations included esophagogastric cancers (n = 9), cholangiocarcinomas (n = 5), and breast cancers (n = 5, 2 of which were triple negative). Treatment with RMC-6236 reduced ERK and P90 RSK phosphorylation in CAL-51 (human triple-negative breast cancer line) and A2780 cells (human ovarian carcinoma cell line), both harboring the RRAS2 Q72L mutation. Treatment of mice bearing CAL-51 xenograft tumors with RMC-6236 (50 mg/kg, once daily) significantly reduced tumor growth. Growth of cells harboring RRAS and RRAS2 mutations was also blocked by MEK1/2 and ERK1/2 inhibitors. Conclusions: Hotspot RRAS2 mutations are rare but recurrently found in endometrial, ovarian, and germ cell tumors. These mutations are predominantly mutually exclusive with other canonical RAS mutations, although co-mutations with RAS do occur in a subset. RRAS2 Q72L -mutant cancer cells are sensitive to inhibition of the MAPK pathway including pan-RAS inhibition both in vitro and in vivo . These preliminary findings may inform future therapeutic strategies for patients with RRAS2 -mutated solid tumors.