Seven years of famine and 7 years of feast in melanoma breakthroughs
For those with early-stage melanoma, surgery may be the only treatment needed. However, when melanoma has spread beyond the skin, additional treatments before or after surgery may be necessary. Multiple lines of therapy are also considered.
With the development of immunotherapy and targeted agents, the treatment landscape for melanoma has changed immensely over the years. ONCOLOGY recently sat down Jeffrey Weber, MD, to discuss the successes and failures in developing new treatments for patients, whether they be alone or in combination, to ultimately prevent relapse. Although there have been many highs and lows, Weber is enthusiastic as to what is to come in the future.
Q: What are the major trends in melanoma treatment right now?
DR. WEBER: Well, the field is obviously moving to detecting the optimal combination. The late 2000s to 2013 saw a whole variety of [trials evaluating] nivolumab (Opdivo), pembrolizumab (Keytruda), ipilimumab (Yervoy) plus nivolumab, and ipilimumab plus pembrolizumab, and in 2013 it was an absolute big deal because all the initial data that were presented [sparked] a lot of excitement. Then by 2014 to 2015, those drugs were all approved and they were used in the adjuvant [setting]. What is happening now is people are interested in neoadjuvant approaches which is not a new drug development issue, it is just how we reuse old drugs.
Looking at new combinations, the first of the new combination trials was Incyte’s indoleamine 2,3-dioxygenase (IDO) inhibitor plus pembrolizumab, that was a complete bust. That was a little worrisome, but then there were 6 or 7 large phase III trials that came in the next year, like the MASTERKEY-265 trial and the trial evaluating an anti-lymphocyte-activation gene 3 (LAG-3) monoclonal antibody and nivolumab. There are also trials evaluating SD-101, CMP-001, and ipilimumab plus IMO-2125, so we have all these trials of mostly programmed cell death protein 1 (PD-1) blockades but at least 1 ipilimumab phase III trial that will be complete and the data will mature in the next year and a half. And we hope that at least some of them will be positive.
My concern is [that] a lot of them were embarked upon without adequate randomization, without an adequate phase II randomized experience, so there is a risk and the risk is that these may not be positive trials. And that is going to hurt the field. For example, the IMO-2125 trial, I believe, was a phase III trial embarked upon with having treated 26 patients. That is a very aggressive drug development pathway.
But everything that is happening is about combinations and what will the best combination be—a direct injection of the toll-like receptor (TLR) agonist? Is it going to be anti–LAG-3? Is it going to be anti–mucin domain-containing protein 3 (TIM-3)? Is it going to be a new checkpoint inhibitor? Nobody knows, and we are not going to know until we see the results of these phase III studies. The best-case scenario is all of them are positive and then we will have a wealth of options to choose from. But where the field is going is toward optimal combinations of new drugs with PD-1 blockade that would not significantly improve or worsen the toxicity, so you would no longer have to use ipilimumab/nivolumab, which frankly is a pretty toxic combination. Right now, that is the regimen to beat, at least in the United States. The best long-term survival data come with ipilimumab and nivolumab, no question.
The original question was where is the field going. The field is going toward defining optimal combinations of checkpoint inhibition to try to keep the toxicity at a reasonable level or certainly not more than what you see with anti–PD-1 agents alone.
And then there is Iovance’s tumor-infiltrating lymphocytes (TIL) trial. TIL therapy has been around for 30 years, and it is finally going to get its turn with the FDA to see whether they will be able to register it. I would say the initial data that have been made public suggest you are looking at a 30-plus percent response rate in all-comers that grow cells, which is almost all the patients. It looks interesting and promising. We will see whether that pans out. It depends on the duration of response, and the progression-free survival, and ultimately the overall survival. We will know that soon [from results] in a large phase II study. There is a lot of excitement about these bispecific T-cell enhancers. I think they will come along in melanoma and other tumors. So, it is just a variety of things, almost all of which depend on [if]…novel new drugs [can be combined] with PD-1 blockade.
Q: What are the most interesting trials?
DR. WEBER: The studies that are probably furthest along would be the direct injectables, things like SD-101, CMP-001, and T-VEC. Those are all [being investigated] in randomized trials that are large, definitive registration phase III trials. I think MASTERKEY-265 probably will read out among the earliest. So those are all TLR agonists that are involved in large phase III trials. I think they have probably accrued most, if not all, of their patients already. They are probably close to complete accrual. There is also an anti–LAG-3/nivolumab trial. There is a phase II anti–TIM-3 trial. There are several vaccines… that are in phase II [trials], but [they] are not yet in phase III trials. Those are the [most exciting trials with data] that will be mature the [earliest].
Then there is the IMO-2125 ipilimumab trial. That is a separate trial with ipilimumab. So, we are looking at nivolumab/ anti–LAG-3 and direct injectables. The idea of nivolumab/anti–LAG-3 is that LAG-3 is a checkpoint that inhibits T-cell reactivity in addition to PD-1 blockade, especially in the PD-1 refractory population. And by blocking it you will overcome resistance which is a reasonable hypothesis.
And the idea of using a direct injectable like T-VEC, or TLR-9, or other agonists is that too many of the tumors have an immunologically privileged tumor microenvironment. You do not have enough...T cells. It is not a so-called inflamed tumor microenvironment, so the direct injection will inflame the tumor microenvironment, attract T cells, and try to induce a stronger immune response in the tumor microenvironment. And those T cells can then exit the tumor and be disseminated elsewhere into distant tumors. And, if that works out, that is a good idea.
There is 1 more phase III trial that started that is going to take a while to accrue [patients] other than [those with] metastatic disease, and that is the neoadjuvant ECOG SWOG trial of neoadjuvant pembrolizumab. There is the CheckMate 915 trial, an adjuvant trial of ipilimumab/nivolumab versus nivolumab alone. All…patients [were accrued] as of 6 months ago, and we will probably need to wait another year to see any information. [Also,] there is a stage II…pembrolizumab versus placebo trial, and…a nivolumab versus placebo trial for resected stage IIB and IIC melanoma. So these will be adjuvant/neoadjuvant trials.
Q: In lung cancer, if you have a druggable, targetable mutation, the patient does not seems to respond well to immunotherapy. Is there worry that it will be the same situation in melanoma?
DR. WEBER: Well, it has been kind of a bust so far. I mean, there is this Novartis phase III trial, called COMBI-i which will read out very soon, in the next 3 to 6 months. That [trial assessed] ...the combination of dabrafenib (Tafinlar) plus trametinib (Mekinist) plus PDR001 versus dabrafenib/trametinib alone. And, if that is a negative study that is going to really kill the targeted plus immunotherapy field.
There was a randomized phase II trial of dabrafenib/trametinib/pembrolizumab versus dabrafenib/trametinib. For survival, [the results were]…negative. It was unimpressive. Even though the response rate was higher, the results were not so great. I am cautiously hopeful, but I am a little concerned that if that was a negative trial, it is going to kill the targeted immunotherapies field [in melanoma]. And you wonder about the potential for MEK inhibition downmodulating immune activity. Antoni Ribas, MD, PhD, has not seen that in animal models or in vitro, but you wonder whether that is happening in vivo. And that could be a reason why we are going to kill that field, if that happens, we will see.
Q: What is the rule of thumb right now? If I have a patient who has both a targetable mutation, where do I start and why?
DR. WEBER: Well, if you look at the median survival, patients who start out with targeted therapy, it is between 25 and 33 months. If you look at the median survival of frontline ipilimumab/nivolumab, remember those data just came out from James Larkin, FRCP, PhD, at The European Society for Medical Oncology Congress a month ago, the median survival had not been reached at 5 years. You tell me what you would rather do, you want to have a median survival of 25 to 33 months or you want to have a median survival of more than 5 years? You decide. I mean, I would accept the increased toxicity and go for the long survival. I think that there is no question and that most people in our academic field would probably opt for immunotherapy frontline even in the patient with a BRAF mutation. The only exception would be someone with really bulk, aggressive high lactate dehydrogenase-high disease.
Q: What research are you working on right now?
DR. WEBER: We have submitted a manuscript showing that a resulting (C-reactive protein) and other acute phase reactants are immunosuppressive. So, the best way to suppress them is [by] blocking interleukin-6 (IL-6), and there is such an antibody called tocilizumab (Actemra), from Genentech. Also, I have an [investigator-sponsored trial] with [Bristol-Myers Squibb; BMS] to look at ipilimumab/nivolumab plus tocilizumab to potentially reduce toxicity and increase effectiveness or both.
Obviously, if neither works it is a bust of a trial, but that trial is ready to go. It is going to be done here at Dana Farber. It is a phase II designed trial. I like the idea that there are correlative markers built into it. And I think it will be a very interesting trial. So, I like that idea. I also like the idea of IDO or histone deacetylase (HDAC) inhibition, and there are a couple of trials out there looking at PD-1 plus HDAC, and we have a trial that we are trying to get together of ipilimumab plus nivolumab plus HDAC inhibition which I think has some promise.
Q: If this is successful, it could increase the response rate pretty significantly?
DR. WEBER: If we see an increased response rate, I will be tickled pink. You evaluate 18 patients in the first stage and you then move on to 49 in the second stage. So, if we see a promising response rate, we will go up to 67 patients. If it still looks that promising, I am sure that it is going to become a corporate-sponsored trial, and it will just be an expanded, maybe a randomized phase II trial. If I were BMS and I saw promising data in a 67-patient phase II study, I would want to do a randomized phase II trial and really decide whether the idea of tocilizumab plus ipilimumab/nivolumab is promising. I would say it is probably more likely that we are going to see reduction in toxicity than increased response rate. But I would be tickled pink if we saw both. We will see. Obviously, the worst-case scenario is you see neither, and after 18 patients you declare it a busted trial and you move on and do something else.
Q: What is the theory on why it would reduce the toxicity or the adverse events?
DR. WEBER: Oh, there is already ample evidence in the literature that patients who have steroid-refractory colitis and other immune-related adverse events respond to tocilizumab. Tocilizumab is an IL-6 receptor blocking antibody that is approved for arthritis of different types, and it is used to block the cytokine release syndrome in patients getting CAR T-cell therapy, so there is ample evidence that it can reduce immune toxicity in certain scenarios. This will be the first time we use it preventively to try to block the possibility of the side effects occurring. And it would make sense because it seems unlikely that it would have a negative impact on efficacy. In fact, it would promote efficacy because we have data submitted showing that high levels of IL-6 are associated with bad outcomes with immunotherapy and chemotherapy. So, it is a negative prognostic marker.
Q: Among patients who are not responding to ipilimumab plus nivolumab, how many have the mutation that you were just talking about that you are going to try and suppress?
DR. WEBER: BRAF is probably present in about half the patients who do not respond to ipilimumab/nivolumab, and then they would go to dabrafenib/trametinib or encorafenib (Braftovi)/binimetinib (Mektovi), or vemurafenib (Zelboraf)/cobimetinib (Cotellic) as their second choice if they [have the mutation]. If they [do not have the mutation], that is a difficult situation. Those patients almost all go on trials, or if they are at an institution that does not have trials, they may go on chemotherapy, or if they are lucky and near an institution that does high-dose interleukin-2 or TIL [therapy], they’ll go on high-dose IL-3 or a TIL trial. That would be the perfect thing for such a patient.
DR. WEBER: Oh, there [are] many, many studies going on currently to assess whether biomarkers can be defined, both for immune-related adverse events and efficacy…People are looking at the genome, RNA expression peripherally in the tumor. So, people are looking in the tumor, in the periphery. They are looking at proteomics, epigenomics, genomics, expression, tumor microenvironment by microscopy, phenotyping, high-resolution phenotyping in the periphery and in the tumor.
There is just huge amount of efforts going on to define biomarkers because everybody wants to be able to tell you who not to treat or who to treat with these expensive therapies. Keep in mind, in the adjuvant mode, you are treating 4 patients to benefit 1, right? In the adjuvant setting to treat stage III patients, you have a 50% average risk of relapse, and you have a hazard ratio of 0.5. So that means that for every 4 patients, 1 of them is not going to relapse. I would rather have that be for every 2 patients, 1 of them will not relapse. But treating 4 patients to benefit 1 gets expensive because you know how much these drugs cost? A lot of money, hundreds of thousands of dollars a year to the insurance company. So, it is a big deal. There is a huge effort in biomarkers. We are also doing microbiome work to define biomarkers.
Q: How large of a population should currently be getting adjuvant treatment?
DR. WEBER: Well it is a pretty big chunk. I have a lot of patients who are getting adjuvant therapy, and it has really changed the landscape because it [is associated with] a 50% risk reduction. You have 20,000 stage III patients in the United States per year, and 10,000 of them would have relapsed back in the bad old days, now only 5000 of them are going to relapse. So, you just cut it down by 5000 patients. That is a big diminution. And we are already seeing lowering death rates and lowering rates of development of stage IV melanoma. Probably a bit of it is the useful adjuvant therapy that we have.
Q: Has any of that work come through to the point where we can say patients probably are not going to respond to immunotherapy and you should go on trial with something else?
DR. WEBER: None of it is at the point of clinically impacting care. As you probably remember, when you are defining biomarkers, you have to prove that you have an analytically valid assay. So, analytical validity. First it is a proof of concept, the idea that it might work. Then you have to show that you have analytic validity, that you have a reliable, robust, useful marker. Then you have clinical validity, meaning it seems to be associated with the clinical outcome in question. And then you have clinical utility showing that you can use it to actually impact on care. We are just at the analytical validity and clinical validity level, and we are not at the point of proving clinical utility. Clinical utility means you do a trial where you use the biomarker to impact care [decisions]; [for example,] if…[the] biomarker is negative, [the patient] will not get treated. If it is positive, [the patient] will. I do not think there are any valid biomarkers that are quite at that level yet.
So, for example, in our trial, we will look at IL-6 levels. So, let us say we find that in our studies that patients whose IL-6 levels go down the most with tocilizumab [will] benefit. Then we can say, okay, well, IL-6 is an established Clinical Laboratory Improvement Amendments-certified enzyme-linked immunosorbent assay test, we are going to stratify the patients. If somebody has [high] IL-6 [levels], they are going to go on ipilimumab/nivolumab/tocilizumab. If they have low IL-6 [levels], we are just going to give them regular ipilimumab/nivolumab. That is the kind of thing that you could do. But that is clinical utility. That is a couple years away.
Q: When is melanoma usually diagnosed?
DR. WEBER: : Most melanomas are diagnosed early in stage I and stage II and cured. If there are 100,000 melanomas a year, I would say 70,000 were caught early, and diagnosed at the earliest possible time, no question about it.
But, that still leaves a significant bunch of patients who will have stage III disease or a high-risk stage II disease or stage IV disease. And the stage III patients are going to have surgery and most of them will require adjuvant therapy because if I had a 50% risk of relapse at 5 years of my melanoma, I would want to do something rapidly.
Q: Is there anything new, interesting, or different along the lines of early prevention or early detection?
DR. WEBER: : There is researcher at our institution who just got a Melanoma Research Alliance grant with 1 of the dermatology residents, [and] they are going to do electronic assessment of patients who go to the dermatologist and they are going to have this electronic remote assessment of the way their skin looks to try to make a diagnosis of melanoma, which is kind of a cool thing. I mean, as you know, getting in to see a dermatologist in this country is like pulling teeth; it is not easy.
They are going to try to assist the dermatologist by having this done remotely, digitally, and electronically where you can do this digitized photography and have it sent remotely to a dermatologist sitting at a computer terminal. That is a very good idea and I think one of the directions early detection is moving in. But there [are] no magic prevention early detection trials going on in melanoma right now.
Q: Are any of the blood biopsies that are hopefully going to be able to detect a number of important cancers early showing promise in melanoma?
DR. WEBER: : Ironically, the same researcher is working on circulating tumor DNA, that is not as much for early diagnosis as it is for diagnosis of early diagnosis of relapse. So, again, if [a patient] has a melanoma surgically removed, instead of scanning them or examining them every 3 months, they would just get a blood sample. And you look and see whether there is any evidence of circulating tumor DNA that has arisen or gone up, and if that is the case, then you get worried and you assume that there is an occult tumor lurking somewhere. And that is the patient you then treat, you could potentially treat them in the absence of disease. So, I think that is a very clever idea.
Let us say you will treat them with sort of adjuvant therapy and there you would have to do a randomized trial to decide whether it was useful. But still, I think that is a very good idea. I applaud that idea, I like it a lot, and I think that is something that will get pursued, no question.
Q: What advice would you give to oncologists and hematologists about how to best position themselves to be ready for development in the field?
DR. WEBER: : Be open-minded and nimble and agile on your feet. And as soon as the data look really promising, adopt new ideas. Be open to new ideas, no question about it.
What are the most recent phase III trial results that have come out that could possibly change practice?
DR. WEBER: The most recent ones that came out were both negative. That was the Incyte trial of pembrolizumab/IDO inhibitor versus pembrolizumab/placebo. And that was not just negative, that was dead negative. I mean, that was the most negative trial I have seen in a long time which is really depressing. And then there was the atezolizumab (Tecentriq)/mitogen-activated protein kinase kinase (MEK) inhibitor versus atezolizumab alone trial. And [those results have not] been published but it was reported [that the trial did] not…me[e]t its end point, which is really a bummer. [Results from] the 2 largest trials that have come out most recently were negative. So, it makes you wonder. You remember the book of Genesis? You remember [the] Old Testament? Remember what it said in the book of Genesis, that “There will be 7 years of feasts followed by 7 years of famine,” I just hope we are not witnessing the end of the 7 years of feast, which actually would have been 2011 or 2010 to 2017, and then 2018 you see [results from] the Incyte trial. I just hope that did not initiate the beginning of the 7 years of famine. That would be depressing, but we will see. We will see how it plays out.
Financial Disclosure: The authors have no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.