Abstract 5557: Mutation analysis of melanoma tumor samples from ECOG 2603 clinical trial

Abstract

Abstract The BRAF kinase, a member of the MAP kinase pathway, has emerged as an attractive molecule to target in melanoma. In an initial attempt to target BRAF in melanoma, a randomized, phase III trial (ECOG 2603) of carboplatin, paclitaxel, and sorafenib versus carboplatin, paclitaxel, and placebo was performed. Patients enrolled on the trial had either unresectable, locally advanced Stage III or Stage IV melanoma. We analyzed patients’ tumor samples for 74 mutations in 13 genes using a custom iPlex (Sequenom). We have preliminarily analyzed 157 tumor samples from patients enrolled in this clinical trial. Analysis of the initial data set demonstrates that 62 samples (42%, 62/148, 95% CI 34%,50%) carried BRAF mutations - 60 samples with V600 and two samples having a BRAF V601 mutation. Thirty eight samples (26%, 38/148, 95% CI 19%,33%) were positive for NRAS mutations, with 33 with Q61, two G12 and three G13 mutations. With the exception of one tumor sample, the BRAF and NRAS mutations were negatively correlated with each other, consistent with prior observations. We also observed a number of rare mutations in our samples including one in AKT1, one in AKT3, four in CDK4, five in beta-catenin, two in GNAQ and three in KIT. The results of this study demonstrated no difference between the two treatment arms in OS and PFS and no association with treatment outcome and BRAF and NRAS mutations. Consequently, the treatment arms were collapsed, and we examined the relationship of BRAF and NRAS mutations with OS and PFS. Interestingly, our data demonstrated that the BRAF and NRAS mutations did not correlate with OS or PFS. This sample population is unique in that it provides a large data set of melanoma tumor samples, all of whom were entered on a clinical trial, unselected for pre-existing mutations. Using the Sequenom assay platform, we have identified a number of mutations in a subset of genes known to be involved in melanoma in patient tumor samples. In the future, these tumor samples will be informative as we continue to examine and identify additional genetic alterations in melanoma, providing information regarding the natural history and distribution of certain mutations. Future studies will use array comparative genomic hybridization to further investigate additional somatic mutations occurring in melanoma tumors and correlate these with clinical outcome, as well as provide insight into the pathogenesis of melanoma. These results may lead to further opportunities for clinical trials. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr 5557. doi:1538-7445.AM2012-5557