Expert Weighs in on Targeted and Immunotherapy Combinations in Melanoma

Ryan J. Sullivan, MD, associate director of the Melanoma Program at Massachusetts General Hospital Cancer Center, assistant professor of medicine at Harvard Medical School

Ryan J. Sullivan, MD

Although multiple combinations of BRAF and MEK inhibitors have been approved by the FDA for the treatment of patients with melanoma, more trials are underway to identify regimens of targeted agents plus immunotherapies to better benefit this patient population, according to Ryan J. Sullivan, MD.

Dabrafenib (Tafinlar) plus trametinib (Mekinist) was the first BRAF/MEK combination to receive regulatory approval for use in patients with melanoma. About 70% of patients have major responses to these agents, said Sullivan, associate director of the Melanoma Program at Massachusetts General Hospital Cancer Center.

Five-year survival outcomes with the combination were confirmed in a pooled analysis of the phase III COMBI-d and COMBI-v trials. Results showed that the OS rates were 37% and 34% at 4 and 5 years, respectively. Specifically, the 5-year OS rate with dabrafenib plus trametinib in patients who had a complete response (CR) was 71% compared with 32% in those with a partial response (PR) and 16% in those with stable disease. However, investigators are still looking for other potential combinations to improve upon this efficacy, said Sullivan.

For example, investigators are now looking to pair immunotherapeutic agents with targeted therapies. The PD-1 inhibitor spartalizumab is being examined in combination with dabrafenib and trametinib, and the PD-L1 inhibitor atezolizumab (Tecentriq) is being evaluated in combination with vemurafenib (Zelboraf) and cobimetinib (Cotellic).

In an interview with OncLive^® during the he 16th International Congress of the Society for Melanoma Research, Sullivan, who is also an assistant professor of medicine at Harvard Medical School, discussed the rationale for BRAF/MEK combinations and immunotherapeutic combinations in melanoma and ongoing research examining triplet regimens.

OncLive^®: What is the rationale for using BRAF/MEK inhibitor combinations in melanoma?

Sullivan: About 40% to 50% of patients with melanoma have a mutated gene called BRAF, which is a critical signaling pathway initiator. When that is active, uncontrolled signaling occurs that is associated with cancer. BRAF mutations in that pathway include KRAS, NRAS, and MEK mutations and are found in a significant minority of patients with cancer.

The discovery of BRAF was made not that long ago, in the early 2000s. Inhibitors of BRAF V600 mutations have been tested now for well over a decade, and when given as an individual drug, we have seen responses in about half of patients, though most have reductions in tumor volume. However, the pathway inhibition by BRAF mutation is usually not enough to be associated with durable benefit in the majority of, or even a significant minority of patients.

Blocking the same pathway with a protein called MEK, is also associated with benefit. In the early 2010s, there were approvals of vemurafenib, which is a BRAF inhibitor; dabrafenib, another BRAF inhibitor; and trametinib, a MEK inhibitor. All were approved for use as single agents because they improved responses, progression-free survival, and overall survival (OS).

As with everything, when you have 2 drugs that work, you always wonder if combining them is going to be an important next step. In fact, preclinical models support the combination [of BRAF and MEK inhibitors]. Interestingly, there was some suspicion that the combination may not be more toxic, rather that it might mediate some of the toxicities [seen with both agents alone]. If you give a BRAF inhibitor, the agent actually enhances signaling in those cells. It’s also true when patients develop hand-foot syndrome and thickened callouses that are associated with the use of BRAF inhibitors by themselves. MEK inhibitors cause AEs by blocking the pathway. When you combine the 2 inhibitors in a cell that is not mutated, that signals to the pathway, and actually balances out the toxicity. It became quite clear even before vemurafenib was approved in 2011 that combination therapy was likely going to be better. Not only does there appear to be mediation of toxicity, but there is also superior efficacy with this approach.

To date, 4 randomized phase III trials have been done in patients with metastatic melanoma and BRAF mutations that have demonstrated the superiority of combination BRAF/MEK versus a BRAF inhibitor alone. Three combinations have received FDA approval, including vemurafenib plus cobimetinib, dabrafenib plus trametinib, and encorafenib (Braftovi) plus binimetinib (Mektovi). We don’t use single-agent BRAF or MEK inhibitors, unless some weird scenario dictates it in a patient with BRAF-mutated disease. We use combination therapy.

Dabrafenib plus trametinib was the first of these combinations to be approved for use in this patient population. Could you discuss the impact of this regimen on the treatment paradigm?

This combination was the first to be tested in phase I studies. The treatment network of providers who have been caring for patients with melanoma have the most experience with dabrafenib and trametinib. The effectiveness of the combination is undoubtable; about 70% of patients will have major responses.

Recent data characterized the 5-year outcomes of patients who were enrolled on 2 of those phase III studies: COMBI-d and COMBI-v. These [trials examined] trametinib and dabrafenib versus dabrafenib, or dabrafenib/trametinib versus vemurafenib. [Investigators] pooled the data from the 2 combination arms of those randomized trials, [which included] around 500 to 600 patients.

The 5-year survival rate is about 30%, while the 5-year PFS ranges from 18% to 20%. That doesn’t sound too amazing, because we want more than 30% of patients alive at 5 years, but those therapies certainly have durable benefit. What is interesting about that data set is that we might not have predicted that 20% of patients would be progression-free with frontline BRAF-targeted therapy, […] but it is higher than we thought.

Now, it’s really just a matter of trying to figure out which patients are right for that therapy who may or may not be right for a PD-1—based therapy or other immunotherapy approaches.

You mentioned the potential of combining BRAF/MEK inhibitors with immunotherapy. Could you discuss the rationale for these triplet regimens?

We are waiting for data from 2 randomized phase III trials of a PD-1 or PD-L1 inhibitor plus a BRAF/MEK regimen versus a BRAF/MEK regimens plus placebo. There is the combination of the PD-1 inhibitor spartalizumab plus dabrafenib and trametinib versus dabrafenib/trametinib, and the other combination is atezolizumab, a PD-L1 inhibitor, plus vemurafenib and cobimetinib versus a BRAF/MEK inhibitor. The rationale for these combinations is: We have these drugs, so let’s put them together. Additionally, there actually are pretty significant data to support the use of those combinations. First, it looks like BRAF inhibitors changed melanoma cells in culture to be more expressive of tumor antigens. One way that immunotherapy may be ineffective is when there aren’t antigens to be recognized by those T cells that are, theoretically, all ready to go.

In the setting of PD-1 inhibition, 1 of the major checkpoints of their functioning as cytotoxic T cells is PD-1/PD-L1 interaction, so you’ve taken that away. If they can’t recognize the antigen, it doesn’t really matter. Expression of the antigen happens when you give BRAF inhibitors or MEK inhibitors to cell lines. If they have biopsy before they start and 10 days in, you’ll see those changes that the tumor is becoming more recognizable to the immune system. That alone was enough to launch several clinical trials looking at various combinations.

The first of [those combinations] was atezolizumab plus vemurafenib. That study had a little bit of a rocky start. A lot of toxicity was observed when we put the drugs together, so we had to come up with a way to lead in with the BRAF inhibitor, modify the dose, and then start the PD-L1 inhibitor. As the field changed, and BRAF/MEK combinations became the thing to look at, we evaluated vemurafenib and cobimetinib plus atezolizumab. That phase I study had a phase II component to it, and those data supported the [launch of the] randomized phase III trial.

The data suggest that the response rate is high, but not significantly high, at least when you’re comparing across the data seen with vemurafenib and cobimetinib. [Responses do not appear to be] dramatically higher. However, you do see more profound responses, and really nice PFS and OS data. The caveat is that it is a selected population of patients. You can’t compare that survival data across [the board] other than to say that patients seemed to do well for a while when they received that combination.

The other combinations that have been studied tremendously are dabrafenib, trametinib, and pembrolizumab (Keytruda). KEYNOTE-022 is a study that had several different parts to it, including a phase I part, which treated about 15 patients, and a randomized phase II part, which enrolled 120 patients. Sixty patients were randomized to receive dabrafenib/trametinib and pembrolizumab, while 60 were randomized to receive dabrafenib/trametinib. These were the first randomized data that we had [with the combination], and it appeared that the response rates were no better. In fact, numerically, [response rates] were worse for the triplet versus the doublet. Specifically, the response rate for the doublet was what you would expect: 72% with dabrafenib/trametinib, but rates were in the mid-to low 60s with the triplet. Why would that happen when you have all these other data suggesting that this was a good idea?

It also seemed like [the triplet] was more toxic. The toxicity, however, seems to do more with the BRAF/MEK combination, rather than the [addition of the] PD-1/PD-L1 inhibitors. We are not seeing immune-related toxicities at a higher degree than what we would expect with giving those drugs as single agents. We are seeing more of the toxicities from the kinase inhibitors. [With dabrafenib and trametinib,] we are seeing alanine aminotransferase (ALT) and aspartate aminotransferase (AST) elevation that is not associated with lymphocytic infiltrate, which is a dominant toxicity. We are seeing other toxicities related to the kinase inhibitors, predominantly. [These AEs] requires dose holds and discontinuation, so presumably that toxicity is the reason why patients aren’t able to stay on the therapy and achieve the same responses that they have with the BRAF/MEK combination alone.

What’s also interesting is that there were more profound responses observed with triplet therapy, which was consistent with other data sets. There were more CRs and the percentage reduction of target lesions was substantially greater with the triplet than the doublet.

Ultimately, what we care about is survival. The OS data that have been published [with the triplet] are too premature to make any calls on it. If the new data are, indeed, suggestive of an OS advantage to triplets, that is going to be important. The first data that we see from the 2 randomized phase III trials are PFS data. Ultimately, if we have some randomized data sets showing OS advantage and we have larger data sets in definitive studies showing PFS advantage, we will feel more comfortable using those types of approaches upfront for a patient with BRAF-mutated melanoma, [especially] if those regimens were to become FDA approved.

What is the risk-benefit ratio with immunotherapy versus targeted therapy?

As a general rule, BRAF-targeted therapy is associated with significant toxicity that is virtually immediately reversible with stopping or holding [treatment]. Immunotherapy is associated with immune-related toxicities that tend to require active treatment, particularly severe treatment, to reverse. When these events are reversed, [patients] usually get better, but not all do. Depending on what setting you are treating, and depending on whether you are giving patients single-agent or combination therapy, there may be a difference in the rate of permanent toxicity. However, ultimately, thyroid, pituitary, and pancreatic endocrine toxicity, such as type 1 diabetes, tends to be permanent. These are all toxicities that we see and believe are permanent when they occur; they require replacement of whatever that gland was making. If it’s the islet cells of the pancreas, you need to give insulin. If it’s the pituitary glands, you need to give cortisol and the thyroxine. If it’s the thyroid gland, you need to [give] thyroxin. Permanent toxicity is much more common with immunotherapy than it is with BRAF-targeted therapy.

With that said, the data suggest phenomenal long-term outcomes [with immunotherapy], based on where we were before; probably 30% to 40% of patients who receive checkpoint inhibitor therapy in the frontline metastatic setting are going to be progression-free at 5 years. OS rates are in the 40% to 50% range at 5 years, which is unprecedented and better than anything else we have seen. In the metastatic setting, it does seem that the long-term data favor frontline immunotherapy.

In the adjuvant setting, that may be different. It seems like the risk-benefit of immunotherapy and targeted therapy in the frontline metastatic setting probably favors immunotherapy. In the adjuvant setting, I don’t believe that we are as confident saying that. This is the first time that an adjuvant therapy may be potentially curative in the metastatic setting. Substantial data support the use of adjuvant therapy because it does, in fact, cure patients; it saves lives.

Robert C; Grob J-J; Stroyakovskiy D, et al. Five-year outcomes with dabrafenib plus trametinib in metastatic melanoma. N Engl J Med. 2019;381(7):626-636. doi: 10.1056/NEJMoa1904059.

<<< View more from the 2019 International Congress of the Society for Melanoma Research