Hitting the Target in BRAF-Mutant Colorectal Cancer

The development of drugs that target specific oncogenic driver mutations has been integral to personalized cancer medicine. BRAF encodes a serine/threonine kinase that is activated by RAS and phosphorylates mitogen-activated extracellular signal–regulated kinase kinase (MEK), leading to downstream activation of the mitogenactivated protein kinase (MAPK)/extracellular signal–regulated kinase (ERK) pathway, a key mediator of cellular proliferation. Somatic mutations causing constitutive activation of BRAF, most commonly a valine-to-glutamic acid substitution at codon 600 (V600E), were first described in human cancer in 2002. It is now known that BRAF mutations are present in approximately 40% to 50% of melanomas, 40% to 45% of papillary thyroid cancers, 20% to 25% of anaplastic thyroid cancers, and 5% to 15% of colorectal cancers and are at a lower prevalence in numerous other tumor types. Moreover, the V600E mutation is the defining genetic event of hairy cell leukemia. Preclinical studies have supported a role for the BRAF mutation as an oncogenic driver, and simultaneous with the discovery of BRAF alterations in cancer, the multikinase inhibitor sorafenib became available for clinical investigation. Sorafenib has in vitro activity against several targets, including BRAF (wild type and V600E mutant), vascular endothelial growth factor receptor isoforms, platelet-derived growth factor receptor beta, FLT3, KIT, and RET. Disappointingly, despite promising preclinical data, in several phase II and phase III clinical trials in patients with melanoma, sorafenib demonstrated poor clinical efficacy either alone or in combination with chemotherapy, and the presence of a BRAF mutation was not predictive of response to sorafenib. Although these negative results initially instilled doubt about the therapeutic utility of targeting mutant BRAF, later correlative studies suggested a mechanistic explanation for the failure of sorafenib in BRAF-mutant melanoma. A comparison of pretreatment and post-treatment tumor biopsies did not reveal a difference in the levels of phosphorylated MEK and phosphorylated ERK (pERK), suggesting that sorafenib was ineffective at inhibiting the MAPK pathway in patients receiving the maximum tolerated dose. The lessons from these efforts were simple, albeit invaluable. First, target inhibition by a drug in cultured cells or xenografts does not imply similar activity in humans. Second, it is premature to disregard a promising target based on the failure of a specific putative inhibitor. Therefore, pharmacodynamic studies are essential to validate that a drug inhibits its intended target within tumor cells in the human host and that this interaction causes sustained downregulation of signaling through the relevant oncogenic pathway. In the case of melanoma, subsequent development of the more potent and selective BRAF inhibitors vemurafenib and dabrafenib led to a dramatic improvement in progression-free survival (PFS) and overall survival compared with chemotherapy, effects that correlated with inhibition of ERK phosphorylation in paired tumor biopsies. Treatment with the MEK inhibitor trametinib also demonstrated efficacy, leading to the testing of the combined inhibition of BRAF and MEK, which showed that dabrafenib with trametinib outperformed dabrafenib or vemurafenib monotherapy. The success of BRAF inhibition in melanoma was a springboard for investigation of this therapeutic strategy in BRAF-mutant colorectal cancer, which is typically refractory to standard treatments and confers a poor prognosis. It was logical to assume that the same inhibitors that had shown such exceptional activity in melanoma would have similar efficacy in colorectal cancers harboring the identical BRAF V600E mutation. In one of the two articles accompanying this editorial, Kopetz et al conducted a phase II pilot study of vemurafenib in 21 patients with BRAF V600-mutant metastatic colorectal cancer. In striking contrast to melanoma, they observed no complete responses, only one (5%) partial response (durable for 21 weeks) and seven patients (33%) with stable disease over a range of 8 to 50 weeks. The median PFS was 2.1 months (range, 0.4 to 11.6 months), and the median overall survival was 7.7 months (range, 1.4 to 13.1 months). In the other article that accompanies this editorial, Corcoran et al evaluated the combination of dabrafenib and trametinib in 43 patients with BRAF V600mutant metastatic colorectal cancer, showing modest but enhanced efficacy compared with vemurafenib monotherapy. With combined BRAF/MEK inhibitor therapy, one patient (2%) had a complete response lasting more than 36 months, four (9%) had a partial response, and 24 (56%) had stable disease with a median PFS of 3.5 months (95% CI, 3.4 to 4.0 months). Notably, the sole patient with a complete response was the only patient on the trial not previously treated with cytotoxic agents. Both trials are important first steps in exploring the vulnerability of BRAF-mutant colorectal cancer to existing targeted therapies. Unfortunately, neither strategy elicited the robust clinical response observed in melanoma. To investigate the etiology of the relative failure of these agents in colorectal cancer, Corcoran et al performed paired JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 33 NUMBER 34 DECEMBER 1 2015