Abstract A012: Emergence of MEK1 mutations in RAF or RAS-mutant solid tumors following treatment with a combination of selective type II RAF inhibitor belvarafenib and MEK inhibitor cobimetinib

Abstract

Abstract Acquisition of ARAF mutations upon treatment with the potent and selective RAF dimer (type II) inhibitor belvarafenib has been previously observed, but whether these alterations or others are observed in patients treated with a combination of belvarafenib and the MEK inhibitor cobimetinib is unknown. The mutational profiles from circulating tumor DNA (ctDNA) were determined using FoundationOne liquid ctDNA assay from paired baseline (BL) and end of treatment (EOT) peripheral blood samples from 79 patients with locally advanced or metastatic RAF or RAS-mutant solid tumors treated with belvarafenib and cobimetinib in the HM-RAFI-103 trial (NCT03284502). The proportion of mutations were compared between BL and EOT using a McNemar’s test (unadjusted) to identify mutations significantly enriched or lost following treatment. ARAF mutations were not observed to significantly increase following treatment with belvarafenib + cobimetinib (4% BL vs 4% EOT in all solid tumors, p = 1; 6% BL vs 6% EOT in melanoma), and in patients with pre-existing ARAF mutations, the variant allele frequency did not consistently increase following treatment. The MEK1 gene had the greatest increase in mutations following treatment, present in 5% (4/79) patients at BL vs 16% (13/79) at EOT (p = 0.02) across all solid tumors, which included 44 colorectal cancer (CRC) cases (56%), and 18 melanoma (23%). While CRC comprised the largest fraction of all solid tumors in this study, it did not contribute the greatest gain in MEK1 mutations at EOT, with 7% (3/44) of CRC patients having MEK1 mutations at BL vs 14% (6/44) at EOT (p = 0.2). The indication with the greatest number of patients developing previously undetected MEK1 mutations at EOT was melanoma, where MEK1 mutations were present in 0% (0/18) patients at BL vs 28% (5/18) at EOT (p = 0.07). The MEK1 mutations gained in melanoma, which included Q58_E62del, P124L, and C121S, occurred in the N-terminal or catalytic core domains of MEK1, were known to be of functional impact, and all but one patient developing them had ≥ 1 known to be gain of function via the Clinical Knowledge Database; one patient gained three distinct MEK1 mutations following treatment. The MEK1 C121S mutation was previously observed to appear in a melanoma patient who developed resistance to the BRAF V600 inhibitor vemurafenib, and in vitro studies demonstrated it also conferred resistance to MEK inhibition (Wagle et al., 2011). Melanoma patients in the HM-RAFI-103 trial gaining MEK1 mutations at EOT were more commonly BRAF-mutant(mt) or NRAS non-mt melanoma at BL, but sample size is small and results were non-significant (p = 0.3 and 0.3, for association with BRAFmt and NRASmt, respectively, Fisher’s exact test). All melanoma patients who developed MEK1 mutations post treatment had either a partial response (1/5) or stable disease (4/5) as best overall response. Together, these data suggest the acquisition of gain of function MEK1 mutations following treatment of solid tumors, including melanoma, with combined type II RAF dimer and MEK inhibition. Citation Format: Stephanie Hilz, Marissa Chen, Malgorzata Nowicka, Harini Chakravarthy, Maryam Moshref, Luca Gerosa, Jennifer Eng-Wong, Cassie Chou, Young S. Noh, Yoon-hee Hong, Yibing Yan. Emergence of MEK1 mutations in RAF or RAS-mutant solid tumors following treatment with a combination of selective type II RAF inhibitor belvarafenib and MEK inhibitor cobimetinib [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Translating Cancer Evolution and Data Science: The Next Frontier; 2023 Dec 3-6; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(3 Suppl_2):Abstract nr A012.