Keith T. Flaherty, MD
Long-term data from the COLUMBUS trial showed that encorafenib (Braftovi) plus binimetinib (Mektovi) had superior overall survival (OS) and progression-free survival (PFS) compared with encorafenib or vemurafenib (Zelboraf) alone in patients with BRAF
At a median follow-up of 48.8 months, the updated results, which were presented at the 2019 ASCO Annual Meeting, demonstrated that the median OS was 33.6 months for encorafenib/binimetinib, 23.5 months for encorafenib alone, and 16.9 months for vemurafenib alone. Overall, the combination decreased the risk of death by 39% versus vemurafenib alone (HR, 0.61; 95% CI, 0.48-0.79). Additionally, the median progression-free survival (PFS) was 14.9 months for encorafenib/binimetinib, 9.6 months for encorafenib alone, and 7.3 months for vemurafenib alone.
In June 2018, the FDA approved the combination of encorafenib and binimetinib for the treatment of patients with BRAF
-mutant unresectable or metastatic melanoma, as detected by an FDA-approved test, based on earlier COLUMBUS data.
“With this updated analysis, the goal is to understand if these results hold up over time for this new BRAF/MEK combination. The answer is, ‘yes’—it had superior PFS and OS compared with BRAF inhibitor monotherapy,” said Keith T. Flaherty, MD, a professor of medicine at Harvard Medical School and director of the Henri and Belinda Termeer Center for Targeted Therapy and director of clinical research at Massachusetts General Hospital.
In an interview with OncLive
, Flaherty discussed the updated COLUMBUS data and its implications on the melanoma field.
OncLive: Could you provide some background on the COLUMBUS trial?
: The COLUMBUS trial was the one and only randomized trial evaluating the new BRAF inhibitor, encorafenib, combined with the new MEK inhibitor, binimetinib, versus encorafenib alone versus vemurafenib. Vemurafenib was the first approved BRAF inhibitor, and of which we had the most data at the time that we were designing the study.
It was a 3-arm randomized trial. Part of the purpose of the study was to test whether encorafenib might be a better BRAF inhibitor than vemurafenib. When [encorafenib] was being developed preclinically in head-to-head comparison studies and animal models, laboratory-based evidence suggested that it might be a better BRAF inhibitor. Within this trial was a direct comparison comparing encorafenib with vemurafenib. It's the only time there has been an attempt to beat out one of the "first-generation" BRAF inhibitors vemurafenib and dabrafenib (Tafinlar), which were FDA approved years before as monotherapy and later in combination with MEK inhibitors.
The preclinical concept was that encorafenib was engineered to be a better BRAF inhibitor in this phase III trial that was correctly tested, and that was shown. There was a statistically significant improved outcome in progression-free survival, the primary endpoint of the study, and overall survival.
That was gratifying, but that wasn't the primary purpose of the trial. The primary purpose of the trial was to vet the new BRAF/MEK combination. Dabrafenib/trametinib (Mekinist) was the first BRAF/MEK inhibitor combination to be approved in 2014. About 1.5 years later, [the combination of] vemurafenib and cobimetinib (Cotellic) were FDA approved, as well. Those were emerging at the time that this trial was designed and executed. The goal was to try to understand if this was another example of a BRAF/MEK combination that would prove superior to a BRAF inhibitor. We thought the odds were pretty good that this would be the case since we had already seen it twice.
What questions still remain with this study?
The biggest issue is that all trial results break down between efficacy in safety insights. In the BRAF/MEK domain, about 20% of patients can have complete responses (CRs) with this type of therapy with the other regimens. Additionally, the vast majority of patients respond to some degree; that has been seen across the board.
Encorafenib was brought forward because it looked like a superior BRAF inhibitor. We were hopeful we might be able to improve the benchmarks of CR and PFS. With this data set being mature about 4 years—which is meaningful in a disease such as melanoma, in which metastatic patients used to succumb to this disease within 1.5 years— a 4-year follow-up is a significant. We've been tracking these PFS and OS milestones and that's where this new information is coming. There's not much shift with response rates numbers in the updated analysis, because those are judged within the first few months of therapy. Much of our focus has been trying to judge where the tail of the curve is emerging,
Cross-trial comparisons are always challenging. These regimens have never been compared head-to-head and likely never will be within the context of one randomized study. We make attempts to adjust to patient characteristics and they're roughly comparable. Running the highest numerical outcomes across endpoints is the best we can do. Encorafenib does appear to be a BRAF inhibitor that's pulling more weight and, when combined with MEK inhibitor, produces the best outcomes we have seen.
This regimen has a unique and better safety profile, which we learned in early studies and was confirmed in the phase III trial. It does not have unique adverse events (AEs) like other BRAF inhibitors; this regimen does not have those issues.
However, it shares some of the class AEs. Added up, the number of serious AEs and quality of life [issues] with this regimen from the early analyses through now appears to be [fewer] compared with these other studies. That's where I feel more confident in performing cross-trial comparisons when looking at the tolerability and safety data. In this analysis, it remains clear that as the efficacy data continue to hold up, the safety characteristics of this regimen are unique.
Is there rationale to add a third drug to the regimen?
We have been searching for combination strategies beyond the 2-drug BRAF/MEK approach because we have a fraction of patients who do not respond, and those who do respond can lose response over time. At a 4-year follow-up, we have a subpopulation of almost 30% of patients who remain progression free up to 4 years. About 40% of patients are alive, meaning some of them progress and were able to benefit from subsequent therapy. Some even remained on therapy through disease progression.
While that is encouraging, it speaks to the room for improvement. There are still patients who are succumbing even though there are other available therapies, notably PD-1 inhibitors as the primary other consideration for these patients. We need more therapies.
An opportunity that has been pursued is BRAF/MEK [inhibition] plus a PD-1 antibody. Multiple studies have investigated that, and we have 2 large randomized phase III trials for which we are awaiting results. I would anticipate that if we see positive results from those studies, we would want to understand if this BRAF/MEK regimen could do even better.
The other opportunity is to undercut resistance with other targeted approaches. There's some very intriguing evidence that's emerged within the past year or 2 within the laboratory suggesting that there may be metabolism or antigenic targets that help to alter the resistant pool of cells in a way that would make them more susceptible to these therapies. However, these are preclinical results and we have a long way to go in terms of clinical application, in part because we understand relatively little about how to target or drug metabolic or antigenic changes that occur in cancer that allow plasticity and emergence of resistance. I'm optimistic that we're at least fingerprinting the problem and know what types of therapies we want.
With the emergence of these updated COLUMBUS data, how is sequencing affected?
For patients who don't have a BRAF
mutation, we only deliberate on immunotherapy options, whether as monotherapy or combination. In the BRAF
-mutant population, it's more complicated because we have all of these options and the clinical trials, including this trial, that were conducted in treatment-naïve patients; therefore, we don't know about the benefit of these therapies as second-line treatments. We know about them as frontline treatments. We stare at the data and talk to our patients about what decision we'll make in the first-line setting, knowing we'll turn to these therapies with a bit less confidence regarding how much benefit they'll offer in the second-line setting.
This is making it hard to talk to patients about whether immunotherapy or targeted therapy is the right upfront approach. The COLUMBUS data became helpful when combined with other data sets. This long-term follow-up data is helping us understand the likelihood of a patient [responding to treatment], based on their disease characteristics—particularly being a long-term non-progressor or having long-term benefit from therapy when they receive upfront targeted therapy or immunotherapy. It looks highly comparable, which makes it very hard to tell patients there is a superior approach. The only way we're going to make progress in this domain is understanding the molecular features—the more patient-specific features that identify those patients who are more likely to get long-term benefit from these available therapies.
The unfortunate update from translational studies is that the patient population that seems to not respond to targeted therapy seems very similar, in terms of their molecular features, to those who don't respond to immunotherapy. There's a consensus emerging in the field that we have a common problem of a population that doesn't get benefit from either. That leaves us with a lot of overlap. There is a lot of work that needs to be done to understand who should get what, in what order, and in what combination.
Liszkay G, Gogas H, Mandalà M, et al. Update on overall survival in COLUMBUS: a randomized phase III trial of encorafenib (ENCO) plus binimetinib (BINI) versus vemurafenib (VEM) or ENCO in patients with BRAF V600–mutant melanoma. J Clin Oncol. 2019;37(suppl; abstr 9512). doi: 10.1200/JCO.2019.37.15_suppl.9512.