PGx7589: PGx Investigation of Efficacy by Meta-Analysis of Dabrafenib/Trametinib Melanoma Studies BRF113710, BRF113929, BRF113683, MEK114267 and MEK115306

Trial description: Dabrafenib (GSK2118436) is a potent, ATP-competitive and selective inhibitor of mutant BRAF kinase (V600E/K) and trametinib (GSK1120212) is a selective, non-ATP competitive, allosteric inhibitor of MEK1 and MEK2 kinases. The U.S. Food and Drug Administration recently approved dabrafenib and trametinib as single-agent therapies as well as in combination for the treatment of unresectable melanoma or metastatic melanoma in adult patients with the most common type of BRAF mutations: BRAF V600E (dabrafenib) and BRAF V600E/K (trametinib). The BRAF V600E/K mutation is found in 40-60% of melanomas causing constitutive activation of BRAF and, in turn, the MAP kinase pathway. Both dabrafenib and trametinib prolong median survival of BRAF V600E/K mutant melanoma patients compared to chemotherapy, however, similar to other kinase inhibitors, secondary resistance typically develops after an initial robust response to these drugs. Multiple resistance mechanisms have been identified (mostly somatic), but there are no validated biomarkers that could predict the response. Germline genetic variations may help explain the variation in treatment response (in addition to the somatic changes in the tumor). A prior PGx investigation of efficacy of dabrafenib (200632) in melanoma subjects from 3 dabrafenib studies (BRF113710, BRF113929 and BRF113683) provided evidence of association trend (p<0.05) of three functional candidate markers (1 in KRAS and 2 in IL8) with the efficacy endpoint of progression-free survival (PFS). While the association trends with KRAS and IL8 markers were not seen in trametinib treated melanoma subjects from MEK114267 (200633), the meta-analysis of the three dabrafenib studies and the trametinib study showed an association trend with KRAS and IL8 markers. The primary objective of this study is to further follow up previously identified genetic association trends with PFS (KRAS and IL8) by meta-analysis inmelanoma subjects from BRF113710, BRF113929, BRF113683, MEK114267 and MEK115306. The exploratory objectives are to investigate any other genetic associations (candidate gene and genome wide) with PFS by meta-analysis in subjects from BRF113710, BRF113929, BRF113683, MEK114267 and MEK115306. The two treatment arms in MEK115306 will be analyzed independently and meta-analyzed with the previous pharmacogenetic data for PFS from BRF113710, BRF113929, BRF113683 (200632) and MEK114267 (200633) studies. PGx subjects from MEK115306 (n=380) will be genotyped for all markers on the Affymetrix Axiom® Biobank Plus GSK Custom array. After genotype QC, approximately 650K SNPs across the genome, are expected to be available for analysis. While all 650K markers will be used for Principal Component Analysis to characterize the genetic ancestry of the PGx population, the following markers will be selected and stratified into tiers for genetic association analysis with the study endpoints: Tier I: Functional variants in KRAS and IL8 genes (n=3). Tier II: Other functional variants in 30 candidate genes (n=64). Tier III: Genome-wide imputed variants (N=10,000,000 variants in total).Meta-analysis will be conducted on subjects from five metastatic melanoma studies (BRF113710, BRF113929, BRF113683, MEK114267 and MEK115306). The subjects from BRF113710, BRF113929, BRF113683 and MEK114267 (200632 and 200633) will be re-analyzed in genome-wide imputed variants. The two arms (dabrafenib monotherapy and a combination of dabrafenib and trametinib) in MEK115306 will be analyzed independently using Cox proportional hazards model with covariates selected by evaluation and the first top 10 principal components and meta-analyzed with results from 200632 and 200633. Genotype association tests will be performed assuming an additive genetic model.AXIOM is a registered trademark of AFFYMETRIX, INC.