Abstract 2: Vertical MAPK pathway targeting in novel genetically engineered mouse and cell line models of NF1-altered melanoma: the mSK-Mel murine cohort

Large scale clinical genomic sequencing efforts have revealed inactivating mutations in the RAS-GTPase Neurofibromin 1 (NF1) in a significant subset of melanomas. To date, immunotherapy and MAPK pathway-directed targeted therapies have been largely inactive in this molecularly defined cohort and immunogenic models that reflect the distinct co-mutation patterns found in NF1-mutant melanoma patients are lacking. Leveraging a population-scale tumor genomic profiling initiative, we identified TP53 as a gene significantly co-altered with NF1 in melanoma. We thus generated and molecularly characterized a cohort of genetically engineered mice with targeted deletion of NF1 in melanocytes. Melanocyte-specific, homozygous knockout of NF1 induced hyperpigmentation yet was insufficient for tumorigenesis. Addition of TP53 knockout and/or conditional activating mutation in BRAF (BRAFVE), resulted in melanoma formation with variable and high penetrance, respectively, along with histologic features consistent with human melanomas. Tumor latency and overall survival in NF1/TP53 double knockout mice was similar to NF1/BRAFVE double mutants. NF1 knockout did not shorten the latency to tumor formation in the setting of BRAFVE/TP53 mutation but did intensify melanocytic hyperpigmentation in all genetic backgrounds tested. To facilitate preclinical and functional studies, we derived 22 congenic cell lines from harvested mouse tumors from NF1 knockout mice, with and without BRAFVE mutation, and tested their sensitivity to targeted agents. As expected, loss of NF1 conditioned the response to BRAF inhibition, while NF1-mutant cells retained sensitivity to MEK inhibition. To abrogate the effects of adaptive RAS reactivation after MEK inhibitor therapy, combined MEK/SHP inhibition in NF1/TP53 knockout cells and BRAF/SHP inhibition in NF1/TP53/BRAFVE mutant cells strongly blunted ERK phosphorylation and cell proliferation better than single agent therapy. However, this response to the addition of SHP inhibition was transient and ERK rebound was driven by continued MEK activation and dependance. In syngeneic xenograft models of NF1/TP53/BRAFVE mutation, MEK inhibition alone, or in combination with RAF and/or SHP inhibition, induced tumor regression and delayed the onset of resistance and progression as compared to doublet RAF/SHP inhibitor therapy. Overall, we demonstrated the efficacy and feasibility of vertical MAPK pathway targeting in a novel cohort of genetically relevant mouse a