'Very Impressive' Overall Survival for Melanoma Combination

Hello. This is Dr Jeffrey Weber. I am a medical oncologist and deputy director of the Laura and Isaac Perlmutter Cancer Center at New York University Langone Health in New York City.

Today I will be continuing to report on some very interesting melanoma abstracts presented at the 2018 American Society of Clinical Oncology (ASCO) meeting.

This time we are going to talk about targeted therapy. I am reporting on an abstract delivered by Dr Reinhard Dummer on the COLUMBUS study^[1] which was a three-arm randomized phase 3 study of two new drugs, encorafenib (BRAF inhibitor) and binimetinib (MEK inhibitor), compared with encorafenib alone or vemurafenib, the then-standard single BRAF inhibitor.

This study was [instigated] by the fact that encorafenib seemed like a very interesting drug with a relatively low off-rate from BRAF. This might give it some advantages because you could either administer it less often or it might have a better physiologic advantage over drugs that equally inhibit BRAF but have a quicker off-rate and might not be quite as effective. This drug was combined with binimetinib, which, interestingly, has a shorter half-life than the standard BRAF inhibitors that we use.

This was a well-designed, fairly large, 1:1:1 randomized study with 577 total patients and about 190 patients per arm. Encorafenib was dosed at 450 mg once daily and binimetinib at 45 mg twice daily. Again, this dosing is different from that of the usual BRAF-MEK inhibitors. The MEK inhibitor was given twice daily and the BRAF inhibitor was given once daily because of their differing pharmacokinetics and off-rates. The control arms were either vemurafenib at the standard dose of 960 mg twice daily or encorafenib at a lower dose of 300 mg/day.

Interestingly, as with most BRAF/MEK combinations, you reduce the toxicity of the BRAF drug by adding the MEK drug, and by adding the BRAF drug you reduce the toxicity of the MEK drug. This makes sense when you look at the biochemistry of these compounds.

The median duration of follow-up is now 32 months. We heard reports on updated progression-free survival (PFS) and overall survival data for the first time The initial results of this COLUMBUS study were published in Lancet Oncology^[2] earlier this year, but they only had response and PFS data; now we have updates on those endpoints as well as the critical overall survival data.

The initial data published already showed that the median PFS for the binimetinib/encorafenib arm was 14.9 months—highly favorable compared with any phase 3 study of a BRAF/MEK combination. Again, those data have now been updated. The patients, as in any well-performed large, randomized study, were well balanced between the arms for stage, substage, BRAF status, Eastern Cooperative Oncology Group (ECOG) performance status, gender, age, and number of organs involved, which is critically an important parameter for benefit with BRAF [and] BRAF/MEK drugs.

A relatively low number of patients had had prior immunotherapy. About a quarter of all patients had adjuvant interferon, which is surprising, but [patients were] well balanced across the arms. One percent had adjuvant ipilimumab, and a total of 6% of patients with metastatic disease had either ipilimumab, PD-1, PDL-1 antibodies, interleukins, or interferon. Most of the patients were previously untreated in the metastatic mode and had not received prior therapy.

Interestingly, the investigators were very keen on understanding what posttreatment or postrelapse treatment or postprogression treatment was used. Even those were very well balanced between the arms in terms of CTLA-4 alone, PD-1/PDL-1 alone, CTLA-4/PD-1 combos, or other BRAF [drugs], BRAF/MEK drugs, or chemotherapy. Any potential difference between the arms could not really be accounted for by differences in subsequent immunotherapy, which we know, of course, is effective.

The bottom line of this presentation is that there was a very impressive overall survival advantage at a median follow-up of over 32 months: 33.6 months median survival for the combo versus the control of vemurafenib at 16.9 months, which is pretty much in line with what we have expected from prior randomized studies. The hazard ratio (HR) was a healthy 0.61 (P < .001). Looking at whatever parameters you wish, whether you look at the 1-, 2-, or 3-year landmarks, shows clear benefit for the combo versus either vemurafenib or the lower dose of encorafenib. And if you look at the forest plot, virtually everything is to the left of 1, favoring the combination. As you would expect, overlap on the 95% confidence intervals occurred when you had an elevated lactate dehydrogenase or more than three organ sites involved, which we know from prior data from Georgina Long^[3] is associated with a poor outcome with BRAF/MEK treatment.

If you look at the combination of encorafenib and binimetinib versus a lower dose of encorafenib alone, there is no surprise that there was a difference. The P value was modest because those differences were actually pretty close and the HR was 0.81. That was not a primary endpoint of the study. The primary endpoint, of course, was looking at vemurafenib as the then control. Encorafenib, interestingly, beats vemurafenib, with a median overall survival of 23.5 months versus 16.9 months. If you look at the updated median PFS, it is 14.9 months for combination encorafenib/binimetinib versus 7.3 months for vemurafenib, which is kind of what you would expect.

Very impressive data. If you look at the responses, as you would expect, the combination was 64% by central review and 76% by investigator review, which would reflect some level of optimism; but again, if we look at the central review, [it's] in line with every other BRAF/MEK combination [and] looking very good. The results with the BRAF drugs alone were between 40% and 51%.

The toxicities show 15% in either arm—encorafenib alone or binimetinib/encorafenib—leading to discontinuation. This is little more than you would expect—say, for example, with a PD-1 antibody. The side-effect profile is somewhat different from what we have seen with other BRAF/MEK drugs. There was some serous retinopathy, a little bit of pyrexia, and some [increased] transaminases, and relatively little photosensitivity or skin issues with this particular combination.^[4]

All in all, [this demonstrated] a very intriguing and impressive overall survival with a median of 33.6 months [for the BRAF/MEK combination], which is better than any phase 3 BRAF/MEK combination [data] that we have seen, all of which have been in and around 24-25 months level. It really makes you wonder whether this will be better than any other [regimen]. We are going to look at the landmarks at 2, 3, 4, and 5 years to decide whether this is really a better regimen.

Please feel free to [write] in with comments. This is Dr Jeffrey Weber.