Originally published June 2.
CHICAGO (GenomeWeb) – Updated Phase III analysis showed that the combination of GlaxoSmithKline's BRAF inhibitor Tafinlar (dabrafenib) and MEK inhibitor Mekinist (trametinib) halted advanced melanoma patients' disease progression for three months longer than those on just Tafinlar, according to data presented at the American Society of Clinical Oncology meeting held here this week.
Because a substantial number of patients are still on the trial and responding to the drug, the overall survival analysis is not final. But at the time that researchers from GSK and elsewhere analyzed the data in the COMBI-d trial, 93 percent of patients on the Tafinlar/Mekinist arm were alive at six months, compared to 85 percent of patients in the single Tafinlar arm. COMBI-d was funded by GSK, which recently sold the rights of its oncology assets to Novartis.
Researchers have known with development of the first BRAF inhibitor, Roche/Genentech's Zelboraf (vemurafenib), that patients with BRAF mutations respond dramatically for a few months, but then develop resistance. The pathways through which the disease reemerges are complex, and researchers have long wondered how well a resistant patient would do if treated by multiple targeted agents that hit different activated pathways driving the cancer. However, conducting a trial to study combinations of targeted agents is challenging due to the risk for increased toxicities for the patient and the need to closely monitor dosing schedules. And in melanoma, until more molecularly targeted agents came to market, a comparative, randomized control trial was not possible.
After the approval of Zelboraf in 2011, the US Food and Drug Administration last year approved GSK's Tafinalr and Mekinist as single agents. Then earlier this year, the agency granted accelerated approval to the Tafinlar/Mekinist combination for advanced melanoma patients with BRAF mutated tumors.
The FDA granted GSK approval for the combination regimen based on data from a clinical trial involving 162 patients with unresectable or metastatic melanoma with BRAF V600E or V600K mutations. These patients, most of whom had received prior therapy, received either Mekinist and Tafinlar or just Tafinlar until their disease progressed or they could no longer tolerate the side effects. In the study, 76 percent of the patients treated with the combination regimen saw their tumors shrink or disappear, an effect that lasted an average of 10.5 months. Comparatively, 54 percent of those who received just Tafinlar had an objective response for an average of 5.6 months.
The COMBI-d study presented at the meeting this week represents the first Phase III randomized controlled trial in melanoma comparing a combination targeted agent against another targeted agent.
Researchers led by Georgina Long of Melanoma Institute Australia randomized 423 patients in this trial – half receiving Tafinlar at 150 mg twice daily and Mekinist at 2 mg once daily, and the other half receiving Tafinalr plus a placebo. Median progression-free survival in the combination arm was 9.3 months for those receiving the combination compared to 8.8 months for those receiving just Tafinlar. The objective response rate was 67 percent versus 51 percent in the combination versus single agent arms, respectively. At median nine months followup conducted in August 2013, 42 percent of patients were still on the monotherapy agent while 53 patients remained on the combination therapy.
"Eighteen patients on the dabrafenib arm were censored and not included in the analysis," Long said at the meeting. "The majority of [these patients] had clinical progression without a scan or a new anti-cancer therapy. This [censoring] occurred in six patients in the combination arm." If this censoring didn't happen, the median progression-free survival for the monotherapy arm would have dropped to 7.6 months and would have remained stable for the combination arm.
Antoni Ribas from the University of California, Los Angeles, reviewed the data from COMBI-d, and asserted that the results were positive despite critics who weren't that impressed by the difference in progression-free survival between the combination and monotherapy arms. "I've been around in the meeting and I hear people say to each other, 'Have you heard of this study? There was two weeks difference between dabrafenib/trametinib versus dabrafenib," he said during a talk. "Well that's the median. That's a number. If we have 100 patients, that's what happens to number 50 of those."
The overall data suggests to Ribas, however, that all the patients do "a little bit better" on the Tafinlar/Mekinist combination compared to just Tafinlar. "Actually, there is no doubt that they did and … this was prospectively tested and met the primary endpoint," said Ribas, who was on the steering committee of the COMBI-d study, but received permission from ASCO to review the study.
The hazard ratio (0.63) in the interim overall survival analysis favored the combination arm, with 40 deaths in the Tafinlar/Mekinist cohort compared to 55 deaths in the Tafinlar arm. "The median overall survival has not been met," Long said. Researchers will conduct the final overall survival analysis when 77 percent of the study's participants have died.
Patients with elevated levels of the enzyme LDH – which may increase during illness – tended to do exceptionally well on the Tafinlar/Mekinist regimen. But Long cautioned that this marker only identifies a good prognosis subset, and that it should not be used to gauge whether patients will respond to the combination treatment, noting that those with normal LDH also did well with Tafinlar/Mekinist.
Adverse event rates were similar for both cohorts, but more patients needed toxicity-related dose modifications after receiving the combination agent than the single agent. Fevers occurred in 51 percent of those receiving the combination compared to 28 percent in the Tafinlar arm, and more patients had Grade 3 fevers while receiving Tafinlar/Mekinist compared to the single agent (6 percent versus 2 percent of patients). The fevers were manageable, according to Long, if investigators educated patients to stop taking the combination therapy for a few days and resume after the fever broke.
"If this is done early, you tend not to get complicated pyrexia [fever] with hypertension or dehydration," she said. Meanwhile, there were fewer inflammatory skin disorders with Tafinlar/Mekinist compared to just Tafinlar.
The Phase III study results were in line with a Phase II analysis using the same doses of Tafinlar and Mekinist. During the same session at the meeting, Keith Flaherty of Dana-Farber Cancer Center reported that in that study of a much smaller cohort, median overall survival for those receiving the combination therapy was 25 months.
"I conclude that this is the first randomized trial demonstrating the superiority of this combination therapy, dabrafenib/tremetinib versus the single agent dabrafenib," Ribas said at the conference of COMBI-d. "I think we're making a lot of progress … in a short period of time."
Activating BRAF mutations are present in between 40 percent to 45 percent of patients with cutaneous melanoma and at lower levels in those with non-cutaneous forms of the disease. During the session, Michael Davies of the University of Texas MD Anderson Cancer Center highlighted the need for targeted agents for patients with wild-type BRAF tumors, whose tumors harbor other oncogenic mutations, such as NRAS.
Advances in NRAS mutant melanoma
Mutations in NRAS are prevalent in 20 percent of cutaneous melanoma, and are mutually exclusive with BRAF mutations in treatment-naïve patients. Moreover, patients with NRAS mutant melanoma tend to have poor prognosis, studies show.
RAS mutations activate the MAP kinase pathway, which is interrogated by MEK inhibitors. Still, patients with NRAS mutant tumors have had variable responses to such agents. For example, Davies highlighted that in a Phase I trial of Mekinist, seven melanoma patients who had NRAS mutations did not respond to the therapy, while in other studies patients have had modest response to such agents.
Davies suggested that MEK inhibitors might serve as a good "backbone" therapy but that doctors would need to "add something to it" for patients with NRAS mutations. In this regard, researchers led by Jeffrey Sosman of Vanderbilt University Medical Center presented data at the meeting showing that the simultaneous inhibition of MEK1/2 and CDK4/6 could more potently inhibit the MAP kinase pathway in melanoma patients with NRAS mutations.
These melanoma patients don't have any targeted therapies directed at their specific molecularly characterized disease, and the trial performed by Sosman and colleagues offers early signs that the combination of Array Biopharma's MEK inhibitor binimetinib and Novartis' CDK4/6 inhibitor LEE011 might be an option in this setting. In the Phase Ib/II open label trial, Sosman and colleagues hope to enroll around 15 patients with NRAS mutations in a dose-finding portion of the study and then around 40 NRAS-mutant melanoma patients in a Phase II cohort to look at clinical activity of the drug. Patients received LEE011 once daily for 21 days per 28-day cycle at 200 mg or 300 mg, and binimetinib twice daily continuously at 45 mg.
The study is still ongoing, and researchers are hoping to establish a recommended Phase II dose. Reporting on the responses of 21 melanoma patients with NRAS mutations, Sosman said that 33 percent of patients to date have had partial responses, and seven had 20 percent to 30 percent tumor regression. "Over 86 percent of patients had some sense of clinical benefit, including stable disease," he said.
At the time of analysis, 12 patients were still on treatment, from between two to eight months. Sosman and colleagues analyzed the tumor profiles of 15 patients using Foundation Medicine's next-generation sequencing-based FoundationOne test, and validated that they were all NRAS mutation positive.
Patients experienced dose-limiting toxicities such as Grade 3 acute renal injury, Grade 4 asymptomatic creatine phosphokinase elevation, Grade 3 edema, and Grade 4 atrial fibrillation. Common drug-related events, included CPK elevation and rashes, were associated with the MEK inhibitor. There were patients who had mild leukopenia (decrease in white blood cells) from the CDK4/6 inhibitor. Researchers reported that there didn't appear to be any drug-drug interactions in the study.
Reviewing the data, UCLA's Ribas called the results impressive despite the caveats with this type of combination targeted treatment strategy, such as added toxicity and tricky dosing. "It's the first time that we have a group of patients with RAS-driven tumors … and we have the majority of patients responding," he said. "That's an exciting advance and we have to build on this."
Noting the encouraging signs that combinations of molecularly targeted agents can potentially be an option for melanoma patients who have become resistant to targeted monotherapies, and that the toxicities may be manageable, conference attendees wondered how long it would be before three targeted agents can be studied. A number of presentations at ASCO explored various combinations of targeted agents, pairing immunotherapies such as the PD-L1 antibody MEDI4736 with Mekinist and Tafinlar.
"Obviously for the MEK/CDK4 combination [in NRAS patients], with another targeted agent, we need more time to look at scheduling especially," Sosman said.
With regard to the BRAF/MEK combination, Flaherty said that researchers must "consider that there is clearly a lower toxicity rate with the BRAF/MEK combination validated across three BRAF/MEK combinations and that leaves room for the addition of a third agent from a tolerability perspective."