Jeffrey S. Weber, MD, PhD, discusses the combination of nivolumab (Opdivo) and ipilimumab (Yervoy) for the treatment of brain metastases, the tolerability of the combination of ipilimumab and talimogene laherparepvec (Imlygic; T-VEC), and what is next on the horizon for pembrolizumab (Keytruda).
Jeffrey Weber, MD, PhD
If mature data from the ongoing phase III trial of pembrolizumab (Keytruda) combined with epacadostat are positive, the combination could become the preferred frontline treatment regimen for patients with melanoma, says Jeffrey Weber, MD, PhD.
The ongoing KEYNOTE-252/ECHO-301 study is exploring the efficacy, safety, and tolerability of the PD-1 inhibitor pembrolizumab with the first-in-class IDO1 inhibitor epacadostat in patients with stage III/IV unresectable or metastatic melanoma (NCT02752074).
Findings of the multi-arm, open-label phase I/II ECHO-204 trial of pembrolizumab/epacadostat in patients with advanced solid tumors presented at the 2017 ASCO Annual Meeting showed the combination demonstrated an objective response rate (ORR) of 63% and a complete response (CR) rate of 5% for patients with treatment-naïve melanoma. Based on the projected 2-year survival data, the combination could represent a new, less toxic frontline standard of care for patients with melanoma, said Weber, a medical oncologist and deputy director of the Laura and Isaac Perlmutter Cancer Center at NYU Langone Medical Center.
“The 1- and 2-year survivals are projected to be as good as ipilimumab (Yervoy)/nivolumab (Opdivo), which means a 64% 2-year survival rate,” Weber said. “Let's say it's around a 62% 2-year survival rate. Given the very favorable toxicity profile, a lot of doctors will want to use pembrolizumab/epacadostat in the frontline setting.”
Weber gave a lecture on the past, present, and future landscape of immunotherapy in the field of melanoma at the 2017 OncLive® State of the Science SummitTM on Melanoma and Immuno-Oncology. In an interview during the meeting, Weber discussed the combination of nivolumab and ipilimumab for the treatment of brain metastases, the tolerability of the combination of ipilimumab and talimogene laherparepvec (Imlygic; T-VEC), and what is next on the horizon for pembrolizumab. Weber: I went over the history of melanoma immunotherapy, not from the very beginnings, but from the beginning of the modern era, where we understood what a checkpoint was and how you inhibit a checkpoint with an antibody, and then I went over some of the initial data. I won’t spend a lot of time on the old data, because there is so much interesting stuff, especially at the recent ASCO meeting. I decided I would put in some of the classic stuff showing the results from trials that led to the approval of pembrolizumab and nivolumab, and not harp much on the really old stuff with ipilimumab from back in 2010 and 2011.
I spent time talking about the updated data from ASCO, which are very interesting—looking at new combinations, activity in brain metastases, long-term outcomes, and new adjuvant data—a whole host of different presentations that I thought would be of interest to the practicing doctor, because that is our audience. My talk had a compendium of data, virtually all of which will have important implications for how they practice, in my view. That was a very interesting study. That was the initial results from the Bristol-Myers Squibb CA209-204 study. It was essentially a phase II study that was in 2 parts. The data that were presented at ASCO described 75 patients from what they called part A, which were the patients who previously had not received treatment for their brain metastases and were not on steroids.
The results were pretty impressive in that, in the brain, you could get as high of a response rate as you could extracranially, meaning outside the brain. The complete response rate in the brain, and in some cases with some pretty big tumors, was 20%, and that is not bad. It's at least as good as you would see with systemic disease and there was no real disparity between the systemic response, that is the extracranial response, and the response on the brain.
The very impressive data suggest that, in some selected patients with small volume or individual brain metastases, you could start treatment with just ipilimumab plus nivolumab, and you may not need to radiate the metastases. However, relatively few patients will present with isolated single metastases. My personal feeling is that given that stereotactic radiosurgery for subcentimeter lesions is very successful at causing total eradication of those lesions.
I would generally treat first with stereotactic radiosurgery for oligometastatic disease in the brain, and then I would treat with ipilimumab plus nivolumab. It also supports the idea in the relatively unusual patient with central nervous system-only disease, where a patient might have 3 brain metastases, you radiate them, and then what do you do?
It says that giving them ipilimumab/nivolumab is a good idea because they will get as much benefit from the ipilimumab/nivolumab with disease in the brain as they would as if the disease were extracranial. It will have some impact on how people practice. It suggests that having brain metastases is not a death sentence.
If you go way back to the original MDX010-020 study, which was the study that led to the registration of ipilimumab in 2011. A good proportion of those patients had prior brain metastases that were generally radiated, and their survival was no worse than those who had no brain metastases. The suggestion being that having brain metastases in the year of checkpoint inhibition is not a death sentence, and some of those patients will go on to long-term survival.
We all understand that often brain metastases are associated with other poor outcomes indicators like high lactate dehydrogenase (LDH), multiple sites of disease, and poor performance status, but if you are a low LDH, low tumor burden patient who happens to have brain metastases, you may do very well with ipilimumab plus nivolumab. That is the message.Pembrolizumab has been in many combinations, [but] relatively few of them have matured. The most interesting ones to me and to most of my colleagues are the phase III studies of pembrolizumab with T-VEC or without and pembrolizumab with the IDO inhibitor epacadostat or without.
Since progression-free survival (PFS) is the endpoint, at least of the pembrolizumab/epacadostat trial, we will probably hear something within the year. It could be as early as the 2017 ESMO Congress, but perhaps it will be at the 2018 ASCO Annual Meeting. If that's a positive study, that would be very impressive.
This then brings up the issue of what if you fail—and half or more of the patients will fail in the first couple of year—what do you do then? Do you just give single-agent ipilimumab? Do you give ipilimumab/nivolumab? It suggests that we should be paying attention to ipilimumab combinations. Ipilimumab was always the bad guy. It was useful with nivolumab or pembrolizumab but, other than that, single-agent ipilimumab was not that popular.
If pembrolizumab/epacadostat becomes a frontline treatment of choice, then we have to start thinking about what you would combine with ipilimumab. Ipiliumumab and T-VEC? Ipilimumab and T-VEC—in the randomized phase II study of 198 patients presented at the 2017 ASCO Annual Meeting—had greater than 30% response rate. That's pretty good.
We don't know the long-term outcome; it's too early. However, that would be a potential second-line choice if pembrolizumab/epacadostat become the first-line choice. Then, what do you do if the flipped dose of ipilimumab/nivolumab is a lot less toxic but is just as effective at 1 and 2 years?
Then, the pendulum will swing back and now people will use ipilimumab/nivolumab in the frontline setting and then can use pembrolizumab/epacadostat in the second-line setting. You'll have a lot of choices and that's good for patients.
It will create some controversy with physicians as to what they use first. It will be a wealth of resources to be able to treat patients, which will be great. You will get more shots on goal—meaning you will get more chances to push someone into a long-term remission and maybe cure them.T-VEC is a very well tolerated, directly-injected agent. Obviously, the limitations are that you have to have an injectable subcutaneous disease and you have to clean the room before and after. It's a bit of an inconvenience to be injecting it, because it's an engineered herpes virus and you can't handle it like it's just a chemical or an antibody. It has other precautions, but the data looked pretty impressive when you combine it with other things.
Alone, it is a treatment of modest benefit, although it is FDA approved and it met its registration endpoint. The initial ipilimumab plus T-VEC data looked very interesting and it's been born out in a large randomized phase II study. That may have some legs to it and it would represent a terrific second-line treatment in a patient with an injectable disease. We are still stuck back in the empiric era, where we decide on who to treat and how to allocate them by clinical parameters, such as performance status, pace of disease, high LDH, low LDH, and tumor burden.
These are 1970s concepts. We still lack reliable predictive biomarkers to differentiate patients who would benefit from just PD-1 blockade alone with modest side effects versus those who will not do well with just PD-1 blockade and need ipilimumab plus nivolumab, for example, or would need pembrolizumab/epacadostat or ipilimumab/T-VEC.
Understanding what those biomarkers are is one of the key frontiers. We have lots of interesting drugs out there. There are more trials than we have possibly available melanoma patients right now. One of the biggest unmet needs is defining the predictive biomarkers that will help you apportion patients to different treatments.One of the most important things for us as academics at a melanoma research center is we think that virtually all patients with high-risk disease, stage IIIb, IIIc, and stage IV resected patients, and anyone with metastatic disease, should be seen at a referral center and at least given the option of going on a clinical trial.
Let's say you're living in Suffolk County and it's a 2-hour schlep to get into Manhattan. Those patients may elect not to come in for a regimen that has them be here every 2 to 3 weeks. That is not an unreasonable decision. Also, they may be very motivated.
When I practiced at Moffitt Cancer Center, in Tampa, Florida, we had patients who regularly would drive across the state from Jacksonville—a 2.5- to 3-hour drive—to come to Moffitt to be on a trial every 2 weeks. Everybody should be able to hear about the new possibilities with these immunotherapeutic agents and be considered for a trial. They may decide not to go on it, but it should be offered and they should come to pay us a visit.