Distinct gene expression, mutational profile and clinical outcomes of V600E and V600K/R BRAF-mutant metastatic melanoma (MM).

9541 Background: BRAF V600E and V600K/R mm have distinct clinicopathologic features suggesting different etiology. V600K/R mm appears less responsive to MAPK inhibitors (MAPKi) compared to V600E MM. We investigated potential mechanisms for this by comparing the gene expression and mutation profiles of these two melanoma subgroups. Methods: BRAF V600 mutant mm patients (pts) treated with MAPKi (BRAFi +/- MEKi) between July/2009 and July/2013 were selected. Demographics, clinicopathologic features and clinical outcomes were examined. Pre-treatment FFPE tumors underwent RNA expression profiling (795-gene nanostring panel) and DNA sequencing (239 gene NGS panel). Molecular results were validated using an independent cohort from the The Cancer Genome Atlas (TCGA). Results: 95 mm pts were included (78 V600E, 17 V600K/R), with median (med) follow-up of 18.4 months (mo). 74 (78%) had BRAFi, 21 (22%) had BRAF/MEKi. At MAPKi start, there were no differences between subgroups regarding age, gender, ECOG, AJCC stage or LDH level. V600K/R pts had a trend to less tumour regression by RECIST (med 30% vs 51%, p = 0.08) and shorter PFS (med 5.1 vs 7.1mo, p = 0.08) than V600E, with no difference in OS (20.8mo vs 17.9mo, p = 0.64). V600K/R had lower expression of the MAPK-pathway feedback regulator DUSP6 and glycosyltransferase GCNT1, compared to V600E (p < 0.05). Analysis of TCGA data (122 V600E, 21 V600K/R) confirmed these findings. There was a trend toward higher mutational load in V600K/R than V600E, confirmed with TCGA data (p < 0.05). V600K/R had a higher proportion of mutations in PIK3CA and several tumour suppressor genes (FBXW7, NF2, RB1 and SMAD4), with only FBXW7 confirmed using TCGA data. Conclusions: V600K/R mm has inferior response and shorter survival with MAPKi than V600E, potentially due to less reliance on MAPK pathway activation (lower DUSP6 expression) and greater use of alternative drivers of oncogenesis (higher mutational load, particularly in tumor suppressor genes). Further analyses will be performed, including comparison of MAPK and additional pathway signalling in cell models. Response to immunotherapy will also be examined.