This review highlights that preclinical data and discusses several ongoing clinical trials that are leveraging this information to explore new therapeutic strategies in targeting BRAF-mutated colorectal cancer.
The V600E BRAF mutation occurs in 5% to 15% of colorectal cancer patients and is a negative prognostic feature. BRAF-directed therapies in metastatic melanoma have shown great promise. However, unlike BRAF-mutated melanoma, single-agent BRAF inhibition has been an ineffective strategy in BRAF-mutated colorectal cancer. In recent years, preclinical experiments have increased our knowledge of BRAF-mutated colorectal cancer. This review highlights that preclinical data and discusses several ongoing clinical trials that are leveraging this information to explore new therapeutic strategies in targeting BRAF-mutated colorectal cancer.
James M. Cleary MD, PhD
In recent years, one of the most encouraging developments in oncology has been the success of BRAF inhibitors in BRAF- mutated melanoma. In addition to BRAF-mutated melanoma, there have been reports of significant responses to BRAF inhibitors in patients with BRAF-mutated lung cancers, thyroid cancers, and hairy cell leukemia.1-4 Unfortunately, to date, attempts to target BRAF-mutated colorectal cancer have been largely unsuccessful. In this article, we will review the molecular and clinical characteristics of BRAF-mutated colorectal cancers and ongoing research efforts to target this aggressive molecular subtype.Colorectal adenocarcinoma is classically thought to develop from adenomatous polyps.5 Recent work has demonstrated that colorectal cancers can also arise from serrated polyps.6-8 Serrated polyps, which include sessile serrated adenomas and traditional serrated adenomas, differ from traditional adenomas in that they have a sawtooth (ie, serrated) appearance. Colorectal cancers harboring BRAF mutations predominantly develop from these serrated polyps.6,9 Many tumors arising from serrated polyps demonstrate the CpG island methylator (CIMP) phenotype.6,8
BRAF mutations occur in approximately 5% to 15% of colorectal cancers.10-15 BRAF-mutated colorectal cancers are located predominantly in the right side of the colon and are typically poorly differentiated.13,14 In addition, BRAF-mutated colorectal cancers occur more commonly in older patients, and predominantly in females.13,14 Multiple studies have demonstrated that BRAF-mutated tumors carry a poor prognosis. This is especially true in microsatellite stable (MSS) colorectal cancers that harbor a BRAF mutation.10-12,14 Roth et al and Popovici et al found that V600E BRAF mutations were a negative prognostic feature in MSS stage II and III colorectal cancers.14,16 The prognosis of BRAF-mutated colorectal cancers with high levels of microsatellite instability (MSI-H) is still unclear. Two recent studies suggested that the prognosis of BRAF-mutated MSI-H colorectal cancers is favorable compared with BRAF-mutated MSS colorectal cancers.10,11The RAF oncogene was discovered in 1983 when it was observed that retroviruses carrying the RAF oncogene transformed murine and avian cells in vitro and in vivo.17,18 Mutations in the BRAF oncogene have now been described in multiple cancers including melanoma, adenocarcinoma of the lung, papillary thyroid cancer, testicular cancer, and hairy cell leukemia.19,20 The V600E BRAF mutation is the most common form of BRAF mutation and results in the constitutive activation of the BRAF kinase domain.19,21
V600E BRAF mutations drive colorectal cancer by causing sustained activation of the MAPK pathway. BRAF is serine/threonine kinase that is downstream of KRAS and is immediately upstream of MEK (Figure). Demonstrative of the importance of the MAPK in colorectal cancer, BRAF and KRAS mutations are mutually exclusive in colorectal cancer and only a few exceptions to this mutual exclusivity have been described.22
In colorectal cancer, the V600E BRAF mutation clearly functions as a driver mutation. Selective blockade of the expression of V600E BRAF by RNA interference led to apoptosis of V600E BRAF-mutated colorectal cancer cell lines.23,24 In addition, V600E BRAF mutation has been shown to transform a “normal” human colon epithelial cell line.23 There is also emerging evidence that V600E BRAF mutations may be an early event in the oncogenesis of serrated colon cancers, as V600E BRAF mutations have been found in precancerous polyps.9,25 Two groups have generated murine models where there is forced expression of mutated BRAF in the gastrointestinal tract.26,27 Both of these models exhibit pathological changes resembling serrated colorectal cancer. Rad et al also demonstrated that some of these mice, which expressed BRAF V637E (the murine equivalent of human V600E BRAF mutations), developed carcinomas in the small and large intestine. Interestingly, many of these cancers exhibited microsatellite instability.26
Mutations in KRAS have been clearly demonstrated to predict resistance to EGFR-directed therapy in colorectal cancer.28-30 However, an area of controversy is the treatment of patients with V600E BRAF-mutated colorectal cancers. Conceptually, BRAF mutations would also seem to suggest resistance to EGFR-directed therapies since both KRAS and BRAF are downstream from EGFR. Consistent with this, several studies have demonstrated a lack of benefit of EGFR-directed therapies in patients with BRAF-mutated colorectal cancer.31-34 However, controversy exists, because other studies have not reproduced this finding. 15,35 Despite this controversy, oncologists agree that current therapies are inadequate in treating metastatic BRAF-mutated colorectal cancers and advances are greatly needed.
A major breakthrough in BRAF-mutated melanoma was the demonstration of the effectiveness of vemurafenib (Zelboraf).36 Vemurafenib was designed by analyzing the structure of the V600E BRAF protein and preferentially binds V600E with a high affinity (IC50=13nM).37 The BRAF Inhibitor in Melanoma 3 (BRIM-3) study, a randomized open-label phase III trial, demonstrated a 48% response rate and improved survival for vemurafenib compared with dacarbazine chemotherapy in BRAF-mutated melanoma.36,38 Vemurafenib was FDA approved for patients with BRAF-mutated melanoma in 2011. Another BRAF inhibitor, dabrafenib (Tafinlar), has shown similar results in BRAF-mutated melanoma and is also FDA approved.39 There have also been several reports of responses to BRAF inhibition in lung cancer, anaplastic thyroid cancer, papillary thyroid cancer, and hairy cell leukemia.1-4 It is important to note that single-agent MEK inhibition also has activity, albeit less than vemurafenib, in BRAF-mutated melanoma. The phase III MEK versus DTIC or Taxol in Metastatic Melanoma (METRIC) trial demonstrated that trametinib (Mekinist), a MEK inhibitor, had a 22% response rate and improved overall survival compared with chemotherapy in BRAF-mutated melanoma.40
Given the encouraging activity seen in other types of cancer, it was a great disappointment that single agent vemurafenib was found to have poor activity in BRAF-mutated colorectal cancer. Kopetz et al found that only 1 of 19 patients with BRAF-mutated colorectal cancer had a partial response to single-agent vemurafenib.41 Similarly, a phase I trial of the BRAF inhibitor dabrafenib found that only 1 of 9 patients with BRAF-mutated colorectal cancer had a partial response to single-agent dabrafenib.2
The next step in developing targeted therapy for BRAFmutated colorectal cancer again paralleled the development strategy used in BRAF-mutated melanoma. Building upon the success of single-agent BRAF inhibitors in BRAF mutated melanoma, Flaherty et al then explored whether combined BRAF and MEK inhibition would improve these results. Their hypotheses was that inhibiting BRAF and its downstream signaling partner MEK would be more effective by increasing the level of inhibition of the MAPK pathway and blocking some mechanisms of acquired resistance. A randomized phase II clinical trial clearly demonstrated that a combination of the BRAF inhibitor dabrafenib and the MEK inhibitor trametinib had a higher response rate (76% vs 54%) and longer progression-free survival (PFS; 9.4 months vs 5.8 months) than single-agent dabrafenib in patients with BRAF-mutated melanoma.42 These striking results led to the 2014 FDA approval of combination therapy with the BRAF inhibitor dabrafenib and the MEK inhibitor trametinib in BRAF-mutated melanoma.
The strategy of combined BRAF/MEK inhibition was also tested in BRAF-mutated colorectal cancer. Interestingly, combined BRAF/MEK inhibition did show more activity than single-agent BRAF inhibition. However, unlike the results seen in melanoma, the effectiveness of combined BRAF/MEK inhibition was still limited. Corcoran et al found that 1 of 36 patients with BRAF-mutated colorectal cancer achieved a complete response to dabrafenib/trametinib.43 In addition, 3 of 36 patients with BRAF-mutated colorectal cancer had an unconfirmed partial response (1 patient achieved a confirmed partial response).43 Analysis of tumor biopsies demonstrated that the BRAF/MEK inhibition decreased downstream activation of the MAPK pathway as phosphorylated-ERK levels were decreased in each of the posttreatment biopsies analyzed.43 Notably, the BRAF/MEK combination was well tolerated. Toxicities included a poorly understood cytokine release syndrome which could cause fevers and flu-like symptoms. This cytokine release syndrome was often treatable with a brief course of steroids.BRAF-mutated colorectal cancer is being intensively studied both preclinically and clinically. Several preclinical studies have sought to explain why V600E BRAF-mutated colorectal cancer is refractory to BRAF inhibitors. Interestingly, 2 groups have suggested that the differential sensitivity to BRAF inhibition that BRAF-mutated melanoma and BRAF-mutated colorectal cancer display may be partly explained by the EGFR protein.44,45 A key observation contributing to this work is that the vast majority of melanomas do not express EGFR and that forced expression of EGFR in BRAF-mutated melanoma cell lines leads to resistance to vemurafenib.44 In an elegant series of experiments, both groups demonstrated that inhibition of BRAF triggers the activation of EGFR in colorectal cancer cell lines.44,45 Importantly, combined EGFR and BRAF inhibition caused decreased cell proliferation and triggered apoptosis of BRAF-mutated colorectal cancer cell lines and xenografts.44,45 These preclinical observations highlight the importance of EGFR in BRAF-mutated colon cancer and offer a possible explanation for why BRAF/MEK inhibition is an effective therapy in BRAF-mutated melanoma but not in BRAF-mutated colorectal cancer.
Capitalizing on the preclinical findings, clinical trials testing this strategy are currently underway. One trial is testing a combination of dabrafenib, trametinib, and panitumumab, an anti- EGFR antibody (ClinicalTrials.gov identifier NCT01750918). Results from this trial are eagerly awaited. Notably, a recently published case report details a patient with refractory BRAFmutated colorectal cancer who responded to cetuximab and off-label use of vemurafenib.46
Mao et al also explored the mechanistic difference between BRAF-mutated melanoma and BRAF mutated colorectal cancer.47 The investigators demonstrated that BRAF-mutated colorectal cancer cell lines have higher levels of PI3K activation than BRAF-mutated melanoma. Importantly, they demonstrated that resistance to vemurafenib could be overcome in BRAF-mutated colorectal cancer cell lines by combining vemurafenib with PI3K inhibitors.47 In agreement with this, Rad et al also found that combined BRAF/PI3K inhibition was an effective strategy in a murine model of BRAF-mutated colorectal cancer.26 The importance of the PI3K pathway in BRAF-mutated colorectal cancer is now being tested in a clinical trial of the combination BRAF inhibitor LGX818 and cetuximab, with and without the PIK3CA inhibitor BYL719 (ClinicalTrials.gov identifier NCT01719380).BRAF-mutated colorectal cancer represents an important molecular subpopulation of colorectal cancer. V600E BRAF mutations are important driver mutations and preclinical data suggests that the 5% to 15% of colorectal cancers that harbor these mutations have a distinct oncogenesis. In addition, compared with other patients with metastatic colorectal cancer, patients with metastatic colorectal cancer with BRAF mutations have a poor prognosis and typically have rapidly progressive disease. Unlike the success of BRAF inhibitors in BRAF-mutated melanoma, BRAF inhibitors have thus far been ineffective in colorectal cancer. However, significant progress in understanding BRAF-mutated colorectal cancers has been made in preclinical studies. Current clinical trials, including trials combining inhibitors of BRAF and EGFR, are leveraging these preclinical advances and there is great optimism that effective targeted therapy may soon be available for patients with BRAF-mutated colorectal cancer. If successful, BRAF-directed therapies could dramatically change the outlook for colorectal cancer patients whose tumors harbor these BRAF mutations.
ABOUT THE AUTHOR
Affiliation: James M. Cleary, MD, PhD, is an instructor of medicine at Harvard Medical School, Dana-Farber Cancer Institute.
Disclosure: Dr. Cleary has no relevant conflicts of interest to disclose.
Address correspondence to: James M. Cleary, MD, PhD, Dana-Farber Cancer Institute, 450 Brookline Ave, Boston, MA 02215; phone: 617-632-6073; fax: 617-632-5370; e-mail: email@example.com