Medical oncologist Jeffrey Weber, MD, PhD, a frequent presenter at ASCO meetings, is a professor of medicine at NYU Langone Medical Center in New York City and deputy director of the Laura and Isaac Perlmutter Cancer Center at NYU. Dr. Weber specializes in lab work based on clinical studies, especially innovative phase 1 studies adding new drugs to established drugs. He is also investigating biomarkers and taking them to clinical design trials, bench to bedside and back, to better target the right therapies to the right patients.
After attending a melanoma tumor board course led by Dr. Weber during ASCO’s annual meeting this year, Mark Teich interviewed him about the state of the art in treatment of advanced melanomas today.
Mark Teich: The two dominant forms of treatment for advanced melanoma now are checkpoint blockade immunotherapy and BRAF-MEK targeted therapy. What are their relative strengths and drawbacks, and which would most physicians consider the frontline therapy today?
Jeffrey Weber, MD, PhD: If you look at the longest-term data with combination checkpoint blockade therapy, ipilimumab-nivolumab (Yervoy®-Opdivo®), there’s about 53 percent four-year overall survival (OS) now, and it looks like this plateaus. So we’re essentially going to cure about half of all patients on ipilimumab-nivolumab. Using the anti-PD-1 checkpoint blockade monotherapies, either pembrolizumab (Keytruda®) or nivolumab, you probably plateau at about 40 percent. So you gain about 10 percent with the combo, but with a tripling of toxicity.
With targeted therapy, if you look at the latest data at this year’s annual meeting, the results of the Combi-D and Combi-V trials of combination dabrafenib-trametinib (Tafinlar®-Mekinist®), it’s about 34 percent long-term survival. The second combination targeted therapy developed, vemurafenib-cobimetinib (Zelboraf®-Cotellic®), is probably going to have about the same results once the long-term studies are in, and the newest combo, encorafenib-binimetinib (Braftovi®-Mektovi®), may have slightly better results, but it will take some years for the long-term results to come in. To date, average OS with this combination has been 33.6 months.
So, do the math. Would you rather have a 34 percent plateau with BRAF-MEK targeted therapy at four years or a 40 to 50 percent plateau at four years with checkpoint blockade therapy? That’s why most people in the academic world now go with frontline checkpoint inhibition, usually combo ipi-nivo. We’ll save the targeted therapy for later.
Now, some patients are too ill to go on the combo checkpoint blockade therapy or have had an allograft transplant requiring immunosuppressants or have autoimmune disease, and those folks won’t go on checkpoint blockade; those folks will go on targeted therapy instead if they have mutant BRAF. But most, if they can, will choose to go on immunotherapy first, given the lower survival plateau of the targeted therapy.
MT: In what other situations would you go with targeted BRAF-MEK therapy first?
JW: The advantage with targeted therapy in the metastatic mode is that you get rapid regression in someone symptomatic with rapidly growing disease. This advantage disappears in someone with brain metastases, where combination ipilimumab-nivolumab is probably going to be more beneficial. Although no one has done a head-to-head study, the data for those with central nervous system (CNS) disease look much better giving combination ipilimumab-nivolumab, even for those with mutant BRAF, then following with targeted therapy after treatment failure or after side effects of the initial therapy become too serious.
With respect to adjuvant therapy, in the first year after treatment there’s 88 percent relapse-free survival (RFS) with combination dabrafenib-trametinib (the one BRAF-MEK targeted therapy approved for adjuvant therapy), which is better than with pembrolizumab in the Keynote 045 study, with a similar patient population. So, it looks like there’s a bump in the first year with the targeted therapy, but then things come down, so the big question is whether the pembrolizumab and nivolumab adjuvant studies will have a higher plateau at four and five years in RFS and OS. [Combination ipilimumab-nivolumab therapy has not been approved yet as an adjuvant therapy.] We don’t know. There’s not enough follow-up yet, but we’ll find out. A study is underway, and so far, two years and three years after treatment starts, targeted and checkpoint inhibition adjuvant therapy look similar. The question is, will that hold up over time?
Another advantage of targeted dabrafenib-trametinib therapy in the adjuvant setting is that, unlike the approved frontline checkpoint blockade therapy adjuvants (nivolumab or pembrolizumab), it doesn’t risk irreversible side effects such as endocrinopathies and diabetes. Although total toxicity is higher for the targeted therapies, the irreversible toxicity is lower. So if, for example, I see a patient, especially a young person, with a low-risk stage IIIA tumor, with good data for BRAF-MEK but not so much data for nivolumab or pembrolizumab, I might be tempted to go with frontline targeted adjuvant therapy rather than immunoadjuvant therapy.
However, because of the distant metastatic data showing a better plateau with immunotherapy, most academicians prefer adjuvant immunotherapy for their first choice rather than targeted therapy. To be honest, that’s not directly backed up, because adjuvant trials haven’t covered comparable patient populations, and they don’t have the same follow-up. That’s more of a gut feeling rather than a data-driven decision. But most of my colleagues who give adjuvant therapy will go with nivolumab or pembrolizumab first rather than combination dabrafenib-trametinib targeted therapy in the adjuvant setting.
MT: Since there’s not long enough follow-up with the adjuvant therapies approved in the past couple of years [nivolumab, dabrafenib-trametinib and now pembrolizumab] to prove a long-term overall survival advantage, why risk the side effects from all of them?
JW: That’s not an unreasonable question, except that recurrence-free survival tends to go hand in hand with overall survival. If you look at the combi-AD study (a randomized, double-blind phase 3 adjuvant trial of combination dabrafenib-trametinib versus two placebos), the RFS and OS benefits pretty much went in lockstep. For most adjuvant therapies, when you see an RFS advantage, it usually converts ultimately to overall survival advantage. In our treatment, that may be a little different. Let’s say you get adjuvant nivolumab alone, and if that fails, you can go on targeted combo therapy if you’re BRAF-mutated, so there’s an inherent crossover. Survival falls out as a useful marker. If you’re on ipilimumab and you fail, you can also get nivolumab, and if you’re on nivolumab and you fail, you can get ipilimumab. That obviates the survival endpoint, and you can only use RFS as your endpoint. But a truly virtuous treatment will probably show an OS advantage, not as big in OS as you did in RFS, but you may well see that advantage. That’s going to take another year of follow-up at least.
MT: Since combination ipilimumab-nivolumab has the highest survival rates in the metastatic setting, what are its chances of being approved soon in the adjuvant setting?
JW: A trial testing combo ipilimumab-nivolumab in the adjuvant setting finished its accrual, and we should know in 2020 or 2021 whether that’s a positive study. The difference between nivolumab monotherapy and ipilimumab-nivolumab is probably not going to be as big as the difference that was found between pembrolizumab versus placebo, say, in the study that got adjuvant pembrolizumab approved. I think we’re going to need longer follow-up with enough events to see a difference, but my gut says there will be an RFS and OS advantage with ipilimumab-nivolumab over nivolumab alone, just as there is in the metastatic setting. We’ll see how that plays out.
MT: In the past, most people with distant metastases died within months. With these vastly improved survival rates from targeted therapy and checkpoint blockade therapy, why do the 2019 data from the American Cancer Society’s respected Cancer Facts and Figures still show only 23 percent average five-year survival for these patients?
JW: The survival figures always lag a year or two behind in the tumor registry. The death total is more important, and deaths are going down significantly, down to 7,230 this year from 9,320 last year (a 22 percent reduction). It’s probably going to keep dropping even more. That’s the key.
MT: Considering that combo ipilimumab-nivolumab has triple the risk of serious side effects, wouldn’t a lot of patients choose the far safer monotherapies, either nivolumab or pembrolizumab, even if 10 percent more patients have long-term survival with the combo? How do patients make this choice, and how do you advise them?
JW: Yes, the monotherapies are significantly less toxic, but it’s all in how you present it, and whether or not you show you’re enthusiastic about the combo and say you would choose it yourself. If you show hesitation for the combo checkpoint blockade, then patients mostly won’t stick with it.
Frankly, if I were a patient, and you told me I had a 6 percent higher chance of increased survival, hell, I want to live longer, so give me the toxic stuff, as long as you can reverse most of the toxicity. As we learn, we are able to reverse more and more of the toxicity, though still not all.
Now, if someone is not willing to risk the toxicity, or is older or more fragile, we’ll put them on a single-agent monotherapy, but usually paired with something else, an investigational drug, usually with low toxicity.
MT: One of the key things investigators have been looking into recently is optimal sequencing of these therapies. Does starting with combo ipilimumab-nivolumab, then when it fails or becomes intolerable moving to targeted therapy, work best, or does starting with targeted therapy and then moving to immunotherapy?
JW: There are two trials testing this, but they are difficult to accrue. Michael B. Atkins, MD, leader of one of those studies, has been very frustrated. His Intergroup Trial is an active 300-to 400-patient randomized study, but recruitment is taking a long time, since most patients want immunotherapy frontline. They see the data from the COMBI-D and COMBI-V studies, and they want immunotherapy. They don’t want the targeted therapy, because dabrafenib-trametinib in the metastatic setting has a lower plateau of survival. I think they’re finally getting decent accrual, and though it’s taking a long time, that trial will eventually tell the story.
MT: One bit of exciting news this year appears to be some of the early findings with neoadjuvant therapy, which is adjuvant therapy given to high-risk melanoma patients before tumor surgery rather than after. Is it true that early success with neoadjuvant therapy appears to be a marker for durable treatment success?
JW: Yes, if I had to venture a guess, I’d say that’s exactly true — there’s a lot of excitement around neoadjuvant therapy, though it’s not yet standard of care. And it appears that it can distinguish early on those who would benefit from adjuvant therapy versus those who would not. In his trial presented at ASCO this year, Alexander M. Menzies, MD, showed that patients who have a pathologic complete response (PCR) to neoadjuvant therapy do better over time than those who don’t. But keep in mind, when Alex showed that compilation of data, he’d had probably about 100 patients with no relapse. However, the median follow-up was only 10 months, so who knows what’s going to happen after 20 or 30 months? That’s why you need to do a clinical randomized trial where you take patients who had complete responses to neoadjuvant therapy, and either treat or don’t treat them with adjuvant therapy after surgery. That would be a trial worth doing.
MT: All the drugs now being used as frontline adjuvant therapies for stage III melanoma — nivolumab, dabrafenib-trametinib and now pembrolizumab — had to gain specific approvals in the adjuvant setting after already being approved for stage IV. Will they also have to gain separate approvals in the neoadjuvant setting?
JW: I don’t think anyone is going to go for registration. They’re already approved drugs. I think the U.S. Intergroup will do a study of pembrolizumab neoadjuvant therapy versus no neoadjuvant therapy followed by pembrolizumab adjuvant therapy after surgery and a complete lymph node dissection. If that result is positive, it will be practice-changing, and it will get into the compendium and probably the National Comprehensive Cancer Network Guidelines, but I doubt they’ll go for an FDA approval. The therapy will just enter common practice. You don’t need a registration for every possible treatment.