Dermatologic Oncology Update, Issue 1, 2019 (Video Program)
Dermatologic Oncology Update, Issue 1, 2019
Featuring interviews with Dr Michael A Postow, Dr Karl Lewis, Dr Mario Sznol and Prof Caroline Robert.
5.25 AMA PRA Category 1 Credits™
To complete the Post-test and Educational Assessment and Credit Form, click here. A statement of CME credit will be issued for scores of 80% or better. Your statement of credit will be mailed to you within 3 weeks or may be printed online.
Mechanism of action of immune checkpoint inhibitors and rationale for combining anti-PD-1/PD-L1 and anti-CTLA-4 antibodies DR LOVE: To begin, I asked Dr Michael Postow about the use of the combination of anti-PD-1 and anti-CTLA-4 antibodies for patients with melanoma. DR POSTOW: It’s a really important question. It’s an important question for patients with melanoma but also, as we are seeing now, combination immune therapy for kidney cancer, for lung cancer and likely in other cancers as time goes on. And the real issue is that we need to know who is going to respond to single-agent immune therapy so that we could, ideally, give people the best chance of response with as little toxicity as absolutely possible. I think the challenge always is that we know when you combine more drugs together, you get higher response rates in the short run than if you give 1 drug by itself. So I think the controversy arises when you combine treatments together. You have a lot more toxicity. Who is it really worth it for to impose that additional toxicity burden on, and what are the long-term benefits of this type of combination strategy? DR LOVE: Before you kind of get into the clinical research data, maybe you can take a step back and just review with us now — I mean, immunotherapy’s been out there a while. We’re learning a lot more about the pathophysiology mechanisms. Kind of how you see how anti-CTLA-4 agents work, how PD-1/PD-L1 agents work and what the potential synergy is there. DR POSTOW: Absolutely. So CTLA-4 was the first immune therapy checkpoint that was developed. And CTLA-4 normally is a molecule on T cells that prevents T cells from becoming too activated. So when T cells get activated to fight off something that shouldn’t be in the body, including cancer, CTLA-4 comes up on the cell surface and turns off those T cells or turns them down. And so when you block CTLA-4, it prevents this turnoff signal from being active, and therefore the T cells are more active. And it’s a very early signal in the immune response. And so when you block CTLA-4, it causes a lot of toxicities, because it really globally disinhibits the immune system in a very generalized way. PD-1/PD-L1 blockade, the other type of immune checkpoint blockade that’s very common in cancer patients, acts a little bit more downstream, if you will, than the CTLA-4 molecule. And, therefore, one of the theories why PD-1 or PD-L1 blockade is better tolerated than CTLA-4 blockade is because the action of the drugs that block PD-1 and PD-L1 occur more in tumor microenvironments, more at a later step in the immune response. And, therefore, the T-cell activity is more focused on where you want it to be, which is in the tumor microenvironment rather than early in an immune response where you get more global disinhibition of the immune response with CTLA-4 blockade. And so that’s maybe why toxicities are a little different with PD-1/PD-L1 and CTLA-4 blockade, but that’s also the reason why there’s been interest in trying to combine these strategies, because they both work at boosting the immune response in very different steps along the immune cascade. And, therefore, there’s the idea that you could maybe get at least additive benefit with combining blockade of both. “Hot” versus “cold” tumors and effect of the tumor microenvironment on response to immunotherapy DR LOVE: So within that model, one of the concepts we’ve heard discussed over the years is the so-called hot tumor. What does that mean, and how does that fit into this model you started to describe? DR POSTOW: A hot tumor is a tumor that we believe is ripe for an immune response to come and destroy it. So a hot tumor may be one that has a preexisting immune response already in it, so a lot of T cells that are trying to destroy the cancer, but they just can’t quite get over the hump. So an immune response has already been generated against this tumor, maybe because it has a lot of mutations, a lot of neoantigens. It looks really different to the immune system for some way. And the T cells are all in the tumor. They are trying to fight the tumor. That’s a hot tumor. There’s a lot of immune reaction going on, but for some reason the immune cells can’t completely destroy the tumor, and they just need a little push with immune therapy to get over the hump and cause the tumor to be destroyed. A cold tumor, by converse, is basically a tumor where there is not much of an immune response at all. So when you try to mount an immune response to a cold tumor, you may need to use combination treatment to get more T cells into the tumor microenvironment or boost it with multiple different strategies. Because if you think about it, it’s much more difficult to conquer a cold tumor for the immune system than a hot tumor because the immune system is already against a hot tumor in many ways, and it just needs a little extra push. But the cold tumors, for whatever reason, you need probably multiple different strategies and combination approaches to help destroy the cold tumors. DR LOVE: The other thing I’ve heard talked about related to this idea of combination therapy is somehow it’s related to the node. That the anti-CTLA-4 works what, in the node? Or before it — does it have anything to do with anatomy here? DR POSTOW: It does a bit. So when you block CTLA-4, it boosts T-cell responses early in an immune response in the priming phase. And the priming phase is when these T cells are first coming in contact with antigen that’s getting these T cells excited, and many times that occurs in lymph nodes. So if you think about CTLA-4 and where the molecule’s really acting, it’s very early in an immune response at the priming phase. And that’s when different cells like dendritic cells are showing off the antigen from tumors to T cells, and many times that’s occurring in the lymph nodes, so that’s one of the sites in the body where there’s T-cell priming and CTLA-4 works. So CTLA-4, one could say, has a lot of activity in the lymph nodes, and PD-1/PD-L1 is more in tumor microenvironments. But as we’re seeing as well that that’s a little bit of an oversimplification, and I don’t want to get into all the different minutiae of all the different multiple places that CTLA-4 blockade works. But another theory is that CTLA-4 blockade does not only work in lymph nodes in increasing the T-cell activity there, but also, CTLA-4 blockade may have a role in the tumor microenvironment as well. So there’s a lot of interest in new ways to get CTLA-4 in the tumor microenvironments to help immune response within that location in addition to the lymph nodes. Correlation between tumor mutational burden and activity of immune checkpoint inhibitors DR LOVE: So I want to bring in another concept that you hear about, and it kind of seems like we’re hearing more about it in lung cancer related to combination therapy, which is tumor mutation burden. Do you want to take a crack at explaining what that is? It seems like I don’t hear much about that in melanoma. You hear a lot about it like in other things, like lung cancer. Go for it. DR POSTOW: Absolutely. So a lot to talk about for tumor mutation burden. It’s a really important concept. And the idea here is that one of the reasons people think that the immune system doesn’t normally destroy cancers is that because cancers come from our normal body tissue, and our immune system is programmed to avoid reacting against normal body tissue, otherwise we would have too much autoimmune disease. So because cancer cells arise from our normal body tissue, the immune system normally is restrained against dealing with cancer cells emerging from our own body tissue, because our immune system is programmed not to attack normal body tissue. But tumors are different from normal body tissue. There are different mutations that arise within them that make the immune system recognize them as something foreign to normal body tissue. And tumor mutation burden is a term that specifically reflects how many different kinds of mutations are present within a cancer that might be different from a normal, healthy tissue in the body. So if you have a lung cancer and a lung tumor that’s growing and there’s a lot of mutations within that lung cancer tumor, that looks different to the immune system than a patient’s normal lung tissue. And the idea there is, hopefully the immune system would recognize it as more different than normal lung tissue. So if you have a lot of mutations in the tumor, the idea is that hopefully the immune system would say, “That looks really weird and really, really different, and I’m going to attack that as something that’s different from normal body tissue.” And what we’ve seen from many, many studies of patients getting immune therapies, namely checkpoint inhibitors, is that patients that have either cancers that have lots of mutations like melanoma — because ultraviolet light causes lots of mutations — or lung cancer, especially people who smoke, that causes lots of mutations in the lung tumors. And there’s some theories that patients who smoke have better responses to immune therapy. Not a reason to go out and start smoking, but an interesting scientific principle that if you have a lot of mutations that have been caused by cigarette smoke, maybe those are the lung tumors that respond better to immune checkpoint blockade. And so the idea there is that one of the measures of this hot tumor that we were talking about, one that has this immune response already activated against it, is, how different is this tumor from normal, healthy body tissue, specifically, how many different mutations are present that might differ from normal, healthy tissue? So tumor mutation burden, one of the factors that’s associated with response to checkpoint blockade, not everything. There are patients with low tumor mutation burden that have great responses to treatment, patients with high tumor mutation burden that don’t respond at all. So this is all along a spectrum, and we don’t really have a cut-point quite yet where we can really say you will or you will not respond to checkpoint inhibitors. But it’s an interesting scientific principle. PD-L1 expression as a predictive marker of benefit with combination immune checkpoint blockade for melanoma DR LOVE: So your colleague in lung cancer, Matt Hellmann, presented data in non-small cell, and now we’re seeing data in small cell. But in non-small cell, it kind of looked like the patients with low PD-L1 and high tumor mutation burdens did better with the combination. Has that concept been looked at in melanoma, and do you buy into it for lung cancer? DR POSTOW: So PD-L1, in addition to tumor mutation burden, is associated with this hot tumor microenvironment. And I think the theory there really is, when you have low PD-L1 and you don’t have a lot of immune reaction against the tumor, you may need combination treatment. And in melanoma in the CheckMate 067 study, which is the Phase III study comparing ipilimumab plus nivolumab, nivolumab alone and ipilimumab alone, the first data reports, at least the 3-year survival data, suggested that there may be a bit of a better benefit for combination over PD-1 alone in patients who were PD-L1 negative, suggested that if you don’t have a lot of immune response against your tumor, you may need combination treatment. That’s not completely perfectly aligned, but I think the same story is emerging in lung cancer and other tumors that if you look at the people who are defined as cold tumors, low tumor mutation burden, PD-L1-negative, maybe those are people that really should be treated with combination checkpoint blockade. The lung data are very interesting. Ipilimumab and nivolumab has been explored most extensively in patients with high tumor mutation burden. And that, perhaps, might be a group that ultimately would get the best benefit from combination immune therapy. We don’t really completely quite know that just yet. It’s still ongoing research in large lung cancer trials. DR LOVE: When you look, though, at, for example, that data set, the clinical research data set you just talked about, where the PD-L1 levels seem to maybe predict benefit in melanoma with the combination, were those the data or were those patients looked at in terms of tumor mutation burden? DR POSTOW: In the melanoma data set, not yet. We only have PD-L1. DR LOVE: Is there enough variation that would be worth looking at? Is it going to be looked at? DR POSTOW: I think it would be worthwhile to look at. DR LOVE: I’m mean, it’s kind of weird that you have lung cancer doing — DR POSTOW: Lung cancer’s ahead of us now. We in melanoma were first, because we didn’t, unfortunately, have treatments that worked all that well for melanoma in the past, namely chemotherapy, interleukin-2, interferon. They worked to a degree in some patients really profoundly, but the vast majority of patients didn’t benefit. So melanoma was right for trying all kinds of crazy ideas, and that’s where all these treatments originated. However, lung has really, I think, in many ways, taken the lead, and I think they’re doing so because they have so many more patients they can answer questions much more quickly, because they can run big randomized trials much more quickly just due to the number of people with lung cancer. So they’re answering really important questions about is PD-L1 a predictive biomarker? Is tumor mutation burden a predictive biomarker? And I think we now have a lot to learn from lung in the melanoma experience. Perspective on the role of PD-L1 testing for patients with melanoma DR LOVE: The thing that’s been kind of interesting, and we’ve been doing a lot more lung CME than dermatologic or melanoma, unfortunately, so I’ve been trying to figure out what’s been going on. But when those data came out, it looked like it was telling a logical story from a translational point of view, clinical point of view. And then I started to ask investigators, and they just kind of weren’t doing it. They weren’t even doing PD-L1. I don’t know, do you actually even look at assays? I’ve heard a lot of your colleagues say they don’t even get PD-L1 levels. DR POSTOW: In melanoma, PD-L1 is really controversial. Melanoma, thankfully, is a tumor type that responds very, very well to checkpoint blockade. And so we have learned that even if someone is PD-L1-negative, they can still have an amazing response to PD-1-based immunotherapy. And, therefore, PD-L1 should never exclude a patient from trying one of these immune therapy drugs. DR LOVE: But, I mean, why wouldn’t you want to look at it, though? Do you look at it? DR POSTOW: I don’t look at — DR LOVE: You don’t even order — incidentally, is there any issue in your mind? Because I have heard people question the quality of the assay data itself. Is that part of the issue here? But you don’t really hear that, like, in lung and other tumors. DR POSTOW: Yes, but we’re getting better across the whole immune therapy field at understanding what does PD-L1 positivity mean and what does negativity mean. There’s a lot of discussion about since this is a continuum, you can be strongly positive, you can be weakly positive. There are percent positivities. You can be 95% positive, 2% positive. So essentially if you call someone positive or negative, you’re making a decision that’s kind of a black and white decision, but really there are a lot of shades of gray here. So that’s one issue with PD-L1. The second issue with PD-L1: What type of antibody are you using to test for this positivity or negativity? And the third issue with PD-L1, which I think is actually the most important, is what we’ve seen is even within 1 tumor, there are pockets of the tumor that can be PD-L1-positive and pockets of the tumor that can be PD-L1-negative. And within the same patient, you can have some tumors that are PD-L1-positive and some tumors that are PD-L1-negative, so this just leads to questions of — it probably just matters where you stick the needle in terms of what response you may get. That being said, with all the caveats of PD-L1 and how much we can really truly believe 1 specific result, there has been still this consistent signal that it is a better outcome from immune checkpoint blockade if you’re PD-L1-positive than if you’re PD-L1-negative. So there is some inclination that this matters, but it’s really hard to make treatment decisions, because in the clinic you have to go one way, or you go the other way. And patients want you to come up with “We need to do this and here’s why, or we need to do that and here’s why.” My personal belief around PD-L1, at least in melanoma where there are so many other factors that you have to consider in the decision do you treat a patient with BRAF/MEK targeted therapy, do you treat them with PD-1 alone or do you treat them with combinations of CTLA-4 and PD-1 alone — and together? The decision is so much more on all these other intangible factors that I can’t really find in my practice that PD-L1 makes a major difference in that decision-making. Efficacy and safety of combination versus single-agent immune checkpoint blockade in patients with melanoma and brain metastases DR LOVE: So some of the factors I’ve heard about, and I’m curious how you think them through, one is the brain. How much better is the combination than single agent in terms of the brain? DR POSTOW: The brain is one of the areas where combination checkpoint blockade, I think, has really definitively proven superiority to single-agent PD-1 blockade. We’re talking about response rates in the high 40s to mid-50% range for combination checkpoint blockade in the brain. And we’re talking about response rates of around 20-ish percent for single-agent PD-1. So this is more than twice as good of a response rate in the brain with combination checkpoint blockade. And the brain is also a place where you don’t want to lose a response. It’s such an important area that it’s really critical if you have a patient with brain metastases that you start to turn that disease in the right direction. So we’ve now seen from 2 prospective studies in melanoma of the combination of nivolumab plus ipilimumab how well it can work in the brain. And the response rates in the brain really look like the same as the response rates in the body. So the brain is an organ, just like every other organ in the body, and although it’s a separate compartment, we have to think about unique issues related to brain metastases in melanoma. I really think that we should be thinking more strongly about combination checkpoint blockade for patients with brain metastases than single-agent because of the more than doubling response rates that we’ll see. DR LOVE: And, I mean, brain’s an issue in people who don’t have clinical brain mets, too, because it’s always a potential concern. Which leads to what I kind of am hearing as a clinical algorithm that has evolved in melanoma concerning this, which is kind of a preference to use the combination if you think the patient can get through it, ie, they’re younger, they’re in good condition, they’re motivated, whatever it might be. But to be fine, if not, the patient’s older, whatever. But I don’t know, is that your philosophy? DR POSTOW: If you take toxicity out of the question — and I know that’s not fair to do because that’s an important consideration — but if you just put that to the side for a moment and you look at pure efficacy, I think the combination is the most effective when you look at all the immune therapy strategies in melanoma. It has the highest response rate. You can argue about the survival results long term. It was not a powered study to look at nivo plus ipi versus nivo alone, and pembrolizumab and nivolumab are equivalent drugs in terms of efficacy as well. So lots of debate in terms of whether any perceived differences, which are not statistically significant in terms of long-term survival, favor combination. But in the end of the day, if you put toxicity to the side, it’s really a question about the combination still driving the most pure efficacy in the brain, highest response rate. So the question then becomes a toxicity question — how much are we going to be hurting these patients with the combination? How much experience do we have nursing patients through these toxicities that they experience? Which patients is it worth taking the much bigger risk to give them combination checkpoint blockade? Because that’s not an easy question. And there are people that if you select them poorly for combination blockade, you can really, really hurt them with the toxicities. And so it’s very, very hard to try to pick the patients who can tolerate the combination and who you may think really need the combination. And you hope that maybe some people could just get away with PD-1 by itself. And I think the bottom line for medicine — in melanoma, at least — is both treatment strategies: single-agent PD-1 and combination checkpoint blockade. Both are totally reasonable because we don’t have a real clear-cut winner in all situations yet. Dosing considerations and adverse events associated with anti-PD-1/PD-L1 and anti-CTLA-4 antibody combinations DR LOVE: So in terms of toxicity when you look at the combination, are you pretty much basically talking about the ipi, the anti-CTLA-4 as adding the toxicity, and how much of that is dose dependent? How are you approaching dosing? DR POSTOW: When I’ve been speaking about combination checkpoint blockade in melanoma, I’m specifically referring to the ipilimumab plus PD-1 combination strategy. As time goes on, we are clearly going to be talking about many more combinations with PD-1 than the ipilimumab combination. That’s what we’ve started with, because we’ve had these drugs, and we’ve decided to combine them in melanoma first. So over time, we hope other PD-1 combinations are just as effective with lower toxicities. In terms of the ipilimumab/PD-1 combination specifically, there definitely is a question about how much ipilimumab do you need to add to PD-1 to get this enhancement in response rate? And the Phase III studies, the Phase II studies in melanoma, they use the high doses of ipilimumab, which is 3 mg/kg, combined with the lower dose of nivolumab, 1 mg/kg. So that’s more ipilimumab relative to nivolumab in the melanoma ipi/nivo combination. In lung cancer, in kidney cancer, it’s a lower dose of ipilimumab combined with nivolumab. And in melanoma, there’ve been studies that have tested lower doses of ipilimumab with PD-1 agents and specifically low dose of ipilimumab plus pembrolizumab. And we’ve seen that even when you add a low dose of ipilimumab to PD-1 — either pembrolizumab, for example, or nivolumab — you could still see that it looks like there’s something additive to PD-1 that you’re getting. But ipilimumab has a dose-dependent effect in terms of toxicities. So our hope is that perhaps with more investigation of lower doses of ipilimumab plus PD-1 that you’ll still get that benefit in terms of efficacy with, hopefully, better toxicities. We’ve known in the past that higher doses of ipilimumab monotherapy have much more toxicity than lower doses of ipilimumab monotherapy, and that’s proving true in the combination with PD-1 as well. So hopefully we will try to sort this out. Melanoma used the higher dose of ipilimumab historically, because that was the dose that was shown to have an overall survival benefit as a single agent, the 3 mg/kg. But over time, it could be quite possible that it changes to 1 mg/kg of ipilimumab with nivolumab. Emerging data with novel anti-CTLA-4 antibodies under investigation for melanoma DR LOVE: So what do we know, and what will we know in the future — I’m not even sure exactly what’s being studied in melanoma right now — in terms of, particularly, other anti-CTLA-4 agents? I mean, there are some out there, and even anti-PD-1/PD-L1 agents. Are we going to see clinical research being reported over the next few years with other agents? DR POSTOW: In melanoma, the other CTLA-4 antibody, tremelimumab, had been tested, and, unfortunately, that was a negative Phase III study. But there is an emerging area of interest with CTLA-4 blockade in general in many, many different kinds of cancers. So tremelimumab is being tested in multiple different types of cancers, namely in combination with PD-1/PD-L1-type drugs. So there’s definitely interest in that treatment in those different kinds of cancers. We’re also seeing different strategies of CTLA-4 blockade in early-phase clinical trials, so ways that different CTLA-4 antibodies can be engineered to help facilitate certain mechanisms of their action and help reduce toxicities. So different strategies like different antibodies that only get activated in the tumor microenvironments, that’s one strategy of CTLA-4 blockade that’s emerging. There are strategies that enhance CTLA-4 blockade depletion of T-regulatory immunosuppressive cells. That’s another strategy that’s being tested. Those are mostly Phase I clinical trials, so not a lot to speak of in terms of efficacy at this point. But all kinds of different ways we’re testing new CTLA-4 blocking antibodies in many different types of cancers. And PD-1/PD-L1 blockade similarly, not just giving an antibody that blocks PD-1 or PD-L1 but trying to engineer those antibodies so they can localize some of their effects to tumor microenvironments more specifically. Those are the kinds of ways we’re trying to engineer these antibodies to really try to improve upon their efficacy and reduce toxicity. Efficacy of combined immune checkpoint blockade versus BRAF/MEK inhibitor combinations for patients with melanoma and BRAF tumor mutations DR LOVE: So I want to ask you in a minute in terms of other checkpoints that are being looked at in terms of blockade. But I kind of want to get a little bit of a feel for you. You were mentioning how immune responsive melanoma is in general. And I’m curious, as you kind of look at it globally, everything that you would consider — the rate of response, the depth of response, cure, the brain, et cetera — how would you globally compare the antitumor efficacy of, let’s just say, the best immunotherapy, so combination treatment with BRAF/MEK combinations in BRAF-positive disease? DR POSTOW: That’s one of the most important questions in the melanoma field. If you have a patient with a BRAF mutation, how do you think about BRAF/MEK inhibitors versus PD-1 plus CTLA-4 blockade? And, in fact, there are many randomized trials that are going on to test this specifically. It’s amazing in the melanoma field, there are trials going on testing targeted therapy versus immune therapy, but we have no prospective randomized data yet to comment on what’s more effective between BRAF/MEK-targeted therapy versus PD-1 plus CTLA-4 efficacy. So most of what we’re left to do is compare across different trials to try to say, how do these combinations of BRAF and MEK inhibitors look in one trial versus how do combinations of PD-1 and CTLA-4 blockade look in other trials? I think people generally feel that there is still no more certainty that you’re going to have for a patient’s immediate benefit than you get with the BRAF and MEK combinations of treatment. If you have a patient that you really need to make sure you turn their disease in the right direction, the BRAF and MEK inhibitors still offer the highest confidence of having that immediate response, immediate palliation within hours or days of starting the BRAF and MEK inhibitors. So if I have a patient with a BRAF mutation and I need to really get their disease turned around right away, I still reach for BRAF and MEK inhibitors. And those can be really, really powerful treatments. And then there’s been this kind of whole discussion — everyone responds to BRAF and MEK inhibitors, but then the problem is, everyone gets resistant to them. And that’s just absolutely not true. We have seen with long-term data from BRAF and MEK inhibitor clinical trials that there are a subgroup of patients — namely those that you would expect would do well, good prognostic patients with few sites of disease, normal LDH and generally good performance statuses — those patients can have a really prolonged benefit with BRAF and MEK inhibitors. And I have people that have been on BRAF and MEK inhibitors for years in many contexts. And so we just still need to figure out, who are the people that are going to benefit long term with BRAF and MEK inhibitors? Who are the people that we need to treat up front with BRAF and MEK inhibitors, and who are the people that we should think about some kind of sandwich of immune therapy, BRAF/MEK inhibitors or vice versa? And I think we don’t really know yet how we’re going to best do that. DR LOVE: Do you have patients in your practice who you’ve treated for metastatic disease with either immunotherapy or targeted therapy that you think are cured? DR POSTOW: I really hope so. I think the notion of a cure is a really important one for patients. There’ve been people that I’ve treated, usually they have a very small amount of melanoma and then it disappears on their scans, and we’ve stopped treatment for different various reasons — side effects or maybe they finished a year or two of treatment. We’ve stopped treatment — there’s no evidence of melanoma on their scans. I see them for what I call well-baby checks and social visits to catch up on their travel all around the world and what’s happening with their family and looking at photos of their kids’ weddings or new grandkids and things like this. These people might be cured, I don’t know. Now, when patients ask me that all the time. They say, “Am I cured? Am I in remission?” What kind of words do I use for those kinds of patients? I don’t really know what the right word exactly is for these kinds of situations because we still, unfortunately, don’t have really long-term follow-up, so I don’t have 5-/10-year follow-up on some of these patients and drugs. But what I’ll say is that I use the word dormant, and I use the word sleeping. And I say, “Your melanoma’s dormant. Your melanoma’s sleeping. We have no evidence of disease on your scans, and we’re babysitting you.” I’m a little bit of a conservative person, so I hesitate to come in and really celebrate and say “You’re cured,” because maybe I’m just superstitious, and I don’t want to invoke any bad karma that could cause a problem with the melanoma and moving forward. But there’s a quite distinct possibility that some of these patients are going to go for years and years and years, if not decades, without any melanoma problems at all. And eventually, unfortunately, someone — no one lives forever, but if the melanoma kind of goes away, you don’t see it anymore, and someone grows to be 100 years old and passes away in their sleep for some other reason, then I would say their melanoma was cured. Clinical presentation and frequency of hypophysitis associated with immune checkpoint blockade DR LOVE: So let’s talk a little bit about toxicity. Obviously, it’s a huge question that goes way beyond melanoma. But I have to tell you, the case you presented at a conference became my all-time toxicity case that I present in breast cancer, everywhere. That was the patient who had the hypophysitis. And I think the patient had brain mets, and I don’t know, I think it was ipi/nivo, I don’t remember. Was it?DR POSTOW: Ipi/nivo, yes. DR LOVE: Right. And then the patient comes in with headache, and the family’s all panicked that it’s the brain mets getting worse. And the brain mets are gone, and the patient has hypophysitis. So anyhow, that’s my favorite case. How many times have you seen hypophysitis, incidentally? DR POSTOW: Oh, it happens all the time. DR LOVE: Really? DR POSTOW: It really happens all the time. DR LOVE: Really? Really? DR POSTOW: In many ways, it’s a very interesting kind of — it happens a lot with CTLA-4 blockade plus PD-1. It doesn’t happen that much with PD-1 alone. DR LOVE: Huh! DR POSTOW: But in many ways, I’m very pleased to explain a headache in a patient with a cancer like melanoma that goes to the brain all the time with inflammation of the pituitary. DR LOVE: I know. Just thinking about that whole thing is incredible. Even you, I’m sure, go through a lot of anxiety. DR POSTOW: Oh, because headache and cancer equals brain mets until proven otherwise, and I’m very glad to have another explanation. DR LOVE: But I have heard — and I think maybe you — I don’t remember with this patient that there’s a characteristic headache. If you hear that, you kind of think that’s what it is, hypophysitis? DR POSTOW: Headache and fatigue and the lack of other neurologic symptoms. DR LOVE: Right. DR POSTOW: I think that’s a big issue. If you have a lot of brain metastases that are causing headache, usually someone may have another neurologic symptom. And you kind of just know it when you see it when it’s the brain metastases, but you really have to get an MRI to be certain. Hypophysitis can masquerade as so many different types of things. I’ve seen people blown off as just being depressed because they have cancer and being tired because you have cancer. And patients that come in and just say, “I’ve just not been myself. I’m exhausted, I can’t work, I just feel down.” And everyone says, “Of course, you have metastatic cancer now, and you’re getting treatment, and go see the psychiatrist,” or, “It’s okay to feel this way, because it’s a tough diagnosis.” DR LOVE: Wow. DR POSTOW: But at the end of the day, then you find out that this is because of secondary adrenal insufficiency from hypophysitis. You put someone on steroids, and they immediately start to feel better. They go back to work. It’s an amazing kind of diagnosis to make in many ways, and I think it’s good to make sure you’re thinking about it. If you don’t think about it, you won’t diagnose it. DR LOVE: I’m curious that you say you think that it is common in your experience. What kind of endocrine abnormalities do you see in these patients? Are they panhypopit or not necessarily? DR POSTOW: When hypophysitis presents itself, often it is the adrenocortical axis that can be affected, so low adrenal function due to poor ACTH secretion by the pituitary. And then the other common hormone that can be affected is secondary hypothyroidism. So if the pituitary is not producing ACTH, you’ll have low cortisol, and you can just supplement that with hydrocortisone. If the pituitary is producing low amounts of TSH, you will have low thyroid function, and you need to supplement that with drugs like levothyroxine. Those are the most common two hormones to be affected. But we really haven’t studied a lot about gonadotropin deficiency. What does it mean if you have hypophysitis? Is your testosterone low? What are the consequences of that? Is your estradiol low for women? Does it affect the menstrual cycle? What are the consequences of low FSH and LH in patients that have hypophysitis? What does that mean for long-term effects? This is really relevant for adjuvant checkpoint blockade when we’re treating a lot of young, healthy men and women, many of whom are planning future fertility issues and other things. What happens when the hormones get messed up in the pituitary gland, and they may want to have kids in the future? I’ve been instructed by a reproductive endocrinologist that as long as it’s only isolated to pituitary dysfunction, those hormones can be supplemented in those patients to have future children. So it’s invoking a whole new field of research, which is really bringing in this whole other hormones affected, long-term consequences of hormone deficient. Patients who are otherwise cured from melanoma or perhaps other cancers when you’re using these checkpoint inhibitors in the adjuvant setting, and a whole new issue of long-term issues with this, mostly because the endocrinopathies can be permanent. Monitoring and management of hypophysitis DR LOVE: So I have to confess I didn’t read the article in the JCO on toxicity of checkpoint inhibitors to see what they said about hypophysitis. But I assume the way you’d screen for it would be with blood work rather than imaging. DR POSTOW: One can check blood work. It’s easier than getting brain MRIs, which is the other way to look for hypophysitis. The blood work that can often be tested would just be a simple TSH or ACTH test in cortisol. So if you see TSH and ACTH low in the blood work, that might get one thinking about hypophysitis. It’s much easier to check those labs when you’re seeing patients than it is ordering brain MRIs all the time. In melanoma patients, I start with the brain MRI at baseline, because it’s important to have the brain image at baseline in metastatic patients with melanoma to ensure that there’s no brain metastases. But when I have a patient who’s symptomatic with headache or other symptoms of hypophysitis, you may need to get a brain MRI to look for hypophysitis. And it’s really important you communicate with your radiologist, because the radiologist will see in the clinical history metastatic melanoma headache, and they’ll tell you no brain metastases, but they won’t necessarily look carefully at the pituitary gland. So you need to say metastatic melanoma or metastatic lung cancer on immune therapy, evaluate pituitary gland and rule out brain metastases. And that’ll prompt these radiologists, who are learning about this, to look for that. DR LOVE: And when you do see imaging evidence of hypophysitis, does it reverse? DR POSTOW: It does. After the event of hypophysitis has occurred, usually the pituitary gland, which often will swell in size, can shrink. And sometimes it’ll actually shrink to smaller than it even was in the beginning. So it’s kind of like a little shrunken-down scar of what it was previously. DR LOVE: Interesting. How long’s it take the headache to go away? DR POSTOW: Once you start steroids, the headache usually gets better in a day or two. Immune-related adverse events in patients with melanoma DR LOVE: I was just flashing on the fact that the other day I was interviewing Dave Spigel from Sarah Cannon, a lung cancer investigator. I think he said that he felt — because we were talking about, again, patients with complications, and he was commenting that he thought that when people get autoimmune complications, they tend to be just 1 organ, although I see your patient here had several problems. So first of all, is that your experience with melanoma? I’m not even sure it’s true in lung cancer. He had that clinical feeling that you don’t see multiple organs attacked. You see one that maybe shows up. And the other thing that he said very anecdotally is that he thought maybe there’s a correlation between toxicity and benefit. So pure speculation, but it’s always fun. What about these 2 questions? DR POSTOW: The first question about how many numbers of organs are involved with treatment when you’re giving checkpoint inhibitors, we know when you combine 2 checkpoint inhibitors together, like CTLA-4 blockade and PD-1 blockade, you have an increase in the number of organs involved. So in melanoma, we’re giving more CTLA-4 blockade and PD-1 blockade, perhaps, than they are in lung, because it’s not yet — DR LOVE: Yes, absolutely. DR POSTOW: — standard clinical practice in lung. So I think in melanoma, we’re seeing many more patients with multiple organs involved probably because we’re giving more combination checkpoint blockade. Single-agent PD-1 blockade, of course, it’s much, much better tolerated than the combinations of CTLA-4 and PD-1 blockade. So a single-agent PD-1 blockade you may only really have 1 organ involved — liver, skin, something like this. But with the combinations in melanoma, it is quite common to have multiple organs involved, and I think that’s also compounded by the fact that in melanoma, we’re using a dose of ipilimumab that’s 3 times higher than the dose that they’re using in lung and renal cell carcinoma. So we’re giving more combination ipi and nivo in melanoma than other diseases, and we’re also using a dose that’s 3 times higher. Again, that may change over time as we get more data about lower doses of ipilimumab, but I think the issue is the clinical feeling of how many organs will be involved is somewhat dependent on where investigators may be coming from with their own clinical experiences. Taking care of melanoma patients, I see multiple organs involved, unfortunately, all the time. And sometimes all different organs get involved at the same time. Sometimes you get 1 organ involved. You get through it. Everyone’s happy. And then there’s something else that happens in a different organ. So it’s really interesting how the timing of all these types of things takes effect and why certain organs characteristically are involved early, middle and late along the course. DR LOVE: Would you like to kind of state in your own mind what you tell your fellows about those 3 time periods, what you see? DR POSTOW: I talk to fellows, I talk to patients — the first toxicity that characteristically occurs is skin toxicity and rash. And so many people will have that present first. It might just be because we know that patients can examine their own skin, so they’ll see the rash very quickly, and that just may be the first toxicity that people report. That happens earliest. Side effects like diarrhea usually happen after a few weeks of getting into treatment. And then the liver inflammation can also happen after a few weeks into treatment. And characteristically, endocrinopathies like hypothyroidism or this hypophysitis that we’ve been speaking about usually happens a little bit later. So that’s kind of the pattern that we’ve seen emerge: skin, GI toxicities and then the endocrinopathies subsequently. But the range of when these things can happen is really, really wide, so clinicians need to be alert to anything at any time. And I think that’s important when you’re talking to patients too, because you need to tell them to call you the day of the treatment, the day after the treatment, anything can really happen all the time and everyone needs to be alert. Correlation between toxicity and benefit with checkpoint inhibitors DR LOVE: I’m curious. When you see a patient who has a lot of toxicity, in your mind, do you think it’s going to be more likely that they’re going to have some benefit? DR POSTOW: The data from all the retrospective studies that have addressed this are a little bit all over the place in terms of some studies have shown clear correlations with toxicity and benefit. Others have been a little bit less certain. It’s not so clear in all different treatment modalities. I think we think with PD-1 monotherapy, any toxicity that the patient has is kind of a good thing. But I guess I kind of use it in that I try to reassure the patients either way. If a patient’s suffering from lots of toxicity, I encourage them, “Your immune system is raging. I think this is a good thing that you’re going to have a good tumor response.” But I think the key is that you don’t have to have terrible toxicity to have a benefit from these treatments. And so if I have a patient without any toxicity, I’ll reassure them that you don’t have to have toxicity to have a benefit. There is this notion that toxicity may be a good thing with immune therapy responses, but it’s not so clear cut in my mind across all the different studies that have been done. But it is a good way to reassure people if they’re having bad toxicity that that does mean their immune system is turned on. DR LOVE: So yes, I guess you could say on the right target, we think. DR POSTOW: Mm-hmm. Case: A 53-year-old man with metastatic mucosal melanoma discontinues the combination of ipilimumab and nivolumab because of immune-related adverse events DR LOVE: So let’s finish out this part of our discussion by hearing about your 53-year-old man, if you can briefly tell me what happened there. DR POSTOW: Sure. So this was a 53-year-old patient that I had. He had a mucosal melanoma in the mouth area, so head and neck mucosal primary and had a lymph node involved in his left neck that was biopsy-proven melanoma. We obtained an MRI of the brain at the beginning, as a routine standard staging tool. And he had 2 brain metastases that were present. One was in the thalamus, and the other was in the frontal lobe. And so with this patient, he had a very low-volume of extracranial disease, just this mucosal melanoma in the mouth and also this lymph node in the neck, but then he had 2 brain metastases. And my decision really was a couple of things. He did not have a BRAF mutation, so we put the BRAF and MEK decision to the side. We have a patient with mucosal melanoma that’s metastatic to lymph node and brain. What’s the best initial systemic treatment option for that patient? And the decision really is between single-agent PD-1 and combination checkpoint blockade. There were 2 reasons I chose combination checkpoint blockade for this particular patient. Number 1, the brain metastases. Number 2, mucosal melanoma is one of those subtypes of melanoma where the response rates to combination checkpoint blockade are also higher than single-agent PD-1 checkpoint blockade. So metastatic mucosal melanoma single-agent PD-1 responses are a little lower than they are with cutaneous melanoma in general. DR LOVE: Any translational correlations of why that’s the case? Is their profile different, or immune system, or — DR POSTOW: Mucosal melanoma has less mutations than cutaneous melanoma, so one theory is that because we know the story that mucosal sites of disease don’t get that UV bombardment that the skin gets all the time — DR LOVE: Interesting. DR POSTOW: — maybe that’s why a little less responsiveness to PD-1. But still very responsive to PD-1 compared to other cancers. So I think if you have a patient with mucosal melanoma and you don’t think they’re right for the combination, you should feel very comfortable to be able to give them PD-1 alone. But in this patient with the brain metastases and given my comfort with the combination, I felt like I would give them the combination of 2 checkpoint inhibitor drugs. But it’s very reasonable to go either way. But both the mucosal reason in this patient and with the brain metastases, that’s why I went with combination checkpoint blockade. DR LOVE: So what happened? DR POSTOW: So this patient got the combination of 2 checkpoint inhibitors and, unfortunately, had a lot of different toxicities — 3 toxicities. The first one was a thyroiditis, which is something that’s very, very common. Management of immune checkpoint inhibitor-associated thyroid dysfunction, hepatitis and pancreatitis DR POSTOW: We always talk about hypothyroidism happening — low thyroid, people are tired and they need levothyroxine. But many times when patients have hypothyroidism, that hypothyroidism is preceded by a period of hyperthyroidism due to thyroiditis. So you get started with the checkpoint blockade. The thyroid gland gets really, really inflamed, spewing out free T4 all through the body, and people can get really sick with that. They lose tons of weight. They can develop tachyarrhythmias — like, this patient had atrial fibrillation — DR LOVE: Wow. DR POSTOW: — that had been happening. DR LOVE: So the patient was clinically hyperthyroid. DR POSTOW: Clinically hyperthyroid and — DR LOVE: I thought that usually that’s thought to be, like, silent. DR POSTOW: Sometimes it can be silent, and you just see the TSH go down and the free T4 go up. But sometimes people can be really symptomatic with — DR LOVE: Wow. DR POSTOW: — hyperthyroidism from thyroiditis. And then the inclination is always, I’m going to give an antithyroid drug like methimazole, for example, or propylthiouracil to try to block the thyroid hormone secretion. But the reality is that this is not necessarily always like a Graves’ disease hyperthyroidism, this is because the thyroid gland is — it’s like a factory on fire that’s just spewing out thyroid hormone everywhere. DR LOVE: Right, I mean, they don’t have exophthalmos, obviously. DR POSTOW: Correct, they don’t have exophthalmos, exactly. DR LOVE: So it’s really chemical hyperthyroidism, as opposed to the autoimmune hyperthyroidism. DR POSTOW: Right, exactly, yes. The thyroid gland is not necessarily — DR LOVE: It’s immune hyperthyroidism, I don’t know. DR POSTOW: Yes, it’s all immune related, but it’s like if you think about the thyroid gland as a factory, it’s not necessarily that the factory production of the thyroid hormone is enhanced, which would be the case in Graves’ disease. I like to think of it — the factory is just on fire, and everything is just kind of spewing out of this factory — DR LOVE: Wow, interesting. DR POSTOW: — that’s on fire with thyroid hormones. So the point, I think, for the education about what to do in this situation is really to tell patients this will burn out over time, because you’ll eventually become hypothyroid when this inflamed thyroid gland dies down a little bit. To treat that hyperthyroid phase with symptomatic treatment as needed, like beta-blockers or even sometimes low doses of steroids to dampen down the inflammation that’s going on, that can help. And then be ready to start levothyroxine, because what will happen is that the thyroid hormone will eventually start to go down, the TSH will start to rise and then you’re hypothyroid. So be ready to start levothyroxine even in someone who’s hyperthyroid. Don’t start it at the time that they’re hyperthyroid, but just wait and be mindful that you’re going to need to start it soon. So I’ve had people lose tons of weight and then gain tons of weight once they get to that hypothyroid phase. DR LOVE: So how long was it after the first dose of ipi/nivo that this started to appear clinically? DR POSTOW: Only a few weeks later. DR LOVE: Wow. DR POSTOW: Pretty soon afterwards. And that was the first toxicity that this person experienced. And usually the thyroid problems can happen later with PD-1 single agent, but this was a situation where the hyperthyroidism happened pretty early, and the thyroiditis. DR LOVE: So then what? DR POSTOW: Once that got under control, then, unfortunately, we had other toxicities that kind of manifest themselves. So namely, this patient developed 2 other toxicities — hepatitis and pancreatitis. And one of the reasons that I suspected thyroiditis was initially the cause is because his liver was inflamed right around the time that the thyroid gland was being overly active. So usually, these kinds of things happen as a constellation of different symptoms altogether, especially with the combination of 3 mg/kg of ipilimumab and the lower dose of nivo that this patient received. So the hepatitis was treated with steroids, and I was using the steroids to also dampen the thyroid inflammation. But once that situation got better, then the pancreas got inflamed. And pancreatitis is a very interesting side effect that we’re still really just trying to learn, how do you manage immune-related pancreatitis? Do you need to put people in the hospital and give them fluids and hydration and bowel rest like they may manage — DR LOVE: They have pain and the whole scenario? Is that what this man had? He had pain, he was sick? DR POSTOW: Exactly. Epigastric pain — DR LOVE: Wow. DR POSTOW: — some kind of low-level nausea, poor appetite. And those are nonspecific symptoms. So you just have to think — DR LOVE: How high were his enzymes? DR POSTOW: Amylase and lipase were in the several hundreds, but sometimes in some patients it can go in the thousands. So I don’t really believe in checking amylase and lipase in people who are asymptomatic, because many times it’ll go up and down, and people with abdominal disease like cancer in their abdomen will have high amylase and lipase values, and I don’t think that that’s reflective of pancreatitis. But when you have someone who’s symptomatic, I do suggest checking amylase and lipase, especially with that characteristic epigastric pain and poor appetite. And many times when you get a CAT scan, or if you get a PET scan in these situations, the pancreas will be enlarged. It will be inflamed. And this is something you also have to talk to your radiologist about, because they won’t necessarily look for it. We’re all good at looking for colitis. I think there’s a lot of familiarity with colitis or enteritis, small bowel inflammation that can happen with these immune checkpoint-inhibiting drugs. But it’s important to pay attention to the pancreas as well, because with more attention on this, we’re seeing pancreatic inflammation. I haven’t seen anyone with terrible pancreatitis like pseudocysts needing to be in the hospital for weeks on TPN and that kind of really severe case, fortunately. Most cases of immune-related pancreatitis have been successfully treated with prednisone. I’ve used infliximab in some cases as well, and people do get better from this. DR LOVE: So all of this is going on in the first 2 cycles of treatment, right? DR POSTOW: Yes, very early on, within the first month or two of treatment. DR LOVE: I’m kind of curious. I don’t know if you could feel the node, or I guess you could see the primary. What was going on with the disease that you could see and feel? DR POSTOW: So we had a hard time really seeing the disease in the back of the mouth. It was something that showed up on an MRI. But one of the issues that I usually like to do is, after someone’s been dealing with such toxicity, you can’t treat them with more immune therapy drugs, so I like to image to see, where are we with the melanoma? Where are we with the cancer right now? Is this continuing to progress, and am I going to really need to be worried about this, or is this patient already having a response? And I think in many cases, when someone’s going through terrible toxicity like this, you will see a disease response. And there’s nothing that’ll make someone feel better psychologically than reassuring them that their cancer is behaving in the way that we want to do. So yes, you feel sick because you’re having pancreatitis and thyroiditis and hepatitis or colitis or hypophysitis or whatever it is, but your cancer’s really going in the right direction. Then the good news is, most of these side effects will resolve with time with treatment with immunosuppressants. As long as the cancer’s heading in the right direction, then we should do well. This patient’s cancer was going in the right direction. So the thing in the mouth was shrinking. The lymph node was fine. The brain metastases were also responding gently, so that was very, very reassuring. And the message at that point was, we need to hold immune therapy, nurse through these side effects and hope that we continue that response even if we discontinue immune therapy permanently. Risks and benefits of radiation therapy for patients with melanoma and brain metastases DR LOVE: And I’ve heard stories like this with bad toxicity, had to discontinue treatment. And then the end of the story is, so we decided to just watch the patient. The tumor went away, and now it’s 4 years later. But I see here that you gave radiation therapy. DR POSTOW: Yes, this patient’s brain metastases were large, and they were in bad spots, at least the one in the thalamus was in a bad spot. So I decided to double down on the brain in this particular patient and offer focal radiation to the brain metastases. It’s a big question in our field now. Now that we have proof of benefit of ipi/nivo in the brain, who do we also give focal radiation to in the beginning, and who do we see if ipi/nivo will work in the brain and follow it carefully? That’s a really important, difficult question. So how do I make that decision in the clinic? When I see a patient with brain metastases, who do I say, “Let’s get you set up with the radiation oncologist to give you radiation and start systemic treatment,” or who do I just say, “Let’s just start systemic treatment and see how things go in the brain”? That’s a very difficult question. The way I make the decision is, if patients have large brain metastases, brain metastases in bad places, if they’re symptomatic or if they’re on steroids, those are patients I use radiation up front in addition to systemic treatment, and that’s because we don’t really know how well ipi/nivo works in the brain when people are symptomatic or on steroids. And although there is a very promising response rate for ipi/nivo alone in the brain, it’s not 100%. And if I have a patient with a metastasis in a bad spot that I don’t want to take a basically 50/50 chance on progressing, that’s a patient I really want to make sure sees the radiation oncologist and goes for radiation treatment. If I have a patient who’s really, really sick with extracranial metastases — so big liver metastases, ascites, shortness of breath for lung disease and they may have one or two 1-cm brain metastases that we pick up as just an incidental finding on a brain MRI, those are the patients I think could start with combination immune therapy alone and follow the brain really carefully, because I think their prognosis is going to be driven by the extracranial metastases. And I’m not sure that they need to go through this really aggressive up-front radiation for a very small brain metastasis at the same time they’re dealing with so much extracranial disease. DR LOVE: How would you globally compare the impact, the benefit that you see with radiation therapy for brain mets in melanoma to, for example, some of the common solid tumors — lung cancer, breast cancer, colon cancer? DR POSTOW: I think the one take-away is that radiation can be very effective for patients with melanoma, including in the brain. There’s this old wives’ tale, I guess I would say, that melanoma is radioresistant and radiation doesn’t work for melanoma. And the reality is, that’s just not true. It absolutely can work very well for patients. And there’s emerging literature — it’s not the highest level of data, it’s retrospective data — but that suggests, perhaps, when patients are receiving immune checkpoint blockade, radiation in the brain would even be maybe more effective at the spots that get radiation than, even historically, radiation alone was for melanoma brain metastases. So I think in the setting of small brain metastases that are approached with focal radiation, melanoma, lung cancer, breast cancer, kidney cancer, whatever type of disease there is in the brain, I support treating these patients aggressively with radiation and/or checkpoint blockade if you can, because it can be very, very effective, and you don’t want to lose control of that area. Case: A 31-year-old woman presents with back pain and is diagnosed with metastatic melanoma with a BRAF tumor mutation DR LOVE: Let’s talk about targeted therapy, a good segue into that. And we’ll start with the metastatic setting. How about this 31-year-old lady, what happened there? DR POSTOW: Yes, so I had a young patient who was very symptomatic with metastatic BRAF-mutant melanoma. She had back pain from bony metastases, she had a lung mass and also — DR LOVE: Where was her primary? DR POSTOW: This was unknown primary melanoma, actually. So she presented with back pain — DR LOVE: Wow. DR POSTOW: — and it’s a really tough story. You come in with back pain. You’re otherwise young and healthy. Had some imaging of the spine and found bony metastases there. And then that led to a whole bunch of other imaging. Fortunately, there were no brain metastases, but it led to a whole kind of workup and found metastatic melanoma of unknown primary, which can happen in many patients. The primary melanoma’s just never detected. Either it was in an area that was not able to be, obviously, seen, or maybe the body’s immune system fought the primary melanoma off and it disappeared but that it still had metastasized inside the body. So this is metastatic melanoma of unknown primary, BRAF mutant and very symptomatic disease with back pain as the initial presentation. UV radiation exposure from the sun as an etiologic factor for melanoma DR LOVE: Just out of curiosity, when you see very young patients like this, is there a correlation — I’m curious what her sun exposure history was, and is that even relevant? DR POSTOW: I do think it is relevant in many patients. There are a lot of young people, unfortunately, with melanoma, and there’s some suggestions that that intermittent sun exposure — so not necessarily people that have been out working in the fields for their whole life like older people that have squamous cell, basal cell carcinomas and then get a melanoma on their scalp or forehead. But younger people that have had a lot of tanning bed exposure or weekend warriors that go out and fry on the weekend and then work in an office all week or go on vacation every now and then and sit out for a week and get totally burned. There is some discussion that these BRAF-mutant cases are associated with this younger group of demographic patients that have this intermittent sun exposure. And so it’s definitely an issue that sun exposure can be a problem, especially given the fact that young people will have melanoma. Unfortunately, melanoma can still happen in places where the sun doesn’t shine, for lack of a better expression, and it’s clearly not only driven by the sun. But it’s really important from a population health perspective to tell patients to be careful in the sun, use sunscreen, limit exposure to sun at the bright times of the day from 10 am to 2 in the afternoon and even longer than that. But I think it’s hard to really know in any one individual case how much the sun contributed to things, but many times when you do mutation analyses of these patient’s tumors, you’ll see a lot of what are called C to T genetic changes, and that can reflect UV-induced damage that likely resulted in the melanoma in many of these cases. DR LOVE: So I’ve asked a lot of lung cancer investigators and oncologists about the issue of regret in lung cancer, for example, with smokers. But I never thought about it in melanoma. Do you find patients, particularly younger ones, who bring this up to you or that you become aware of who have regrets who feel self-blame? DR POSTOW: That does happen sometimes, and I think it’s unfortunate that people feel this way. And I try to get people to — everyone’s obviously very anxious, very depressed even to have this diagnosis in the first place. So I try to really gear my conversations to focus on how we’re going to move forward with the diagnosis and that it’s not so much helpful to go back, and say, “You can’t beat yourself up for the whole lifetime that you’ve lived enjoying yourself, whether you live in a sunny climate or you’ve been an outdoor kind of person.” That’s been part of people’s lives. It’s just who they are, and that’s something that you can never take away from somebody. So at this point, I focus on moving forward. One of the questions I get all the time with people with metastatic melanoma is, “Should I stay inside now, or how much care do I need to take in the sun?” And for people with metastatic melanoma, I really tell them that the disease that you have at the moment is not going to be further exacerbated by additional sun exposure. So people are worried that being out in the sun, does that make the metastatic melanoma grow faster? And to the best of our knowledge, from the research that’s available, at least that I’m aware of, it doesn’t accelerate the growth of the metastatic melanoma. So I tell people, “Just be careful. Do what you were doing before, but just be cautious.” Of course, sun exposure does put people at risk of an additional primary melanoma in the skin, but that’s not usually the problem when someone’s already facing metastatic melanoma. So I tell people they can’t live in a cave. If Stage III or Stage IV melanoma, just be smart, follow up with the dermatologist and let’s just focus on how we can try to make you better. DR LOVE: Interesting. Choosing among the BRAF/MEK inhibitor combinations dabrafenib/trametinib, vemurafenib/cobimetinib and encorafenib/binimetinib for melanoma with a BRAF tumor mutation DR LOVE: So this patient sounds like a classic — when you were talking before about the decision for metastatic BRAF-positive disease in people who need responses. She’s got epidural disease. I see also you gave her radiation therapy — getting back to your point about radiation therapy, but then the issue of systemic treatment. And can you talk a little bit (A) about the issue of immunotherapy versus targeted therapy? I think you kind of alluded to how you think it through before, but particularly the issue of choice of targeted therapy in this kind of situation. It seems like we’ve completely evolved towards a doublet of BRAF/MEK, but we now have 3 different combinations. So can you talk a little bit about how you think through those 3 options? And how you choose one versus the other? DR POSTOW: Absolutely. So first, the reason I chose BRAF and MEK in this patient was because of the symptomatic disease and also of the highest degree of confidence with an immediate response with the BRAF/MEK inhibitors. Especially someone with epidural disease in the spine, if you don’t get that immediate response, you worry about spinal cord issues or other nerve problems. And even though the response rates to combination immune therapy are high, still, this BRAF and MEK response confidence is still higher — that almost everyone shrinks initially with the BRAF/MEK inhibitors. Now, how to choose? We have 3 different BRAF/MEK inhibitor combinations that are out on the market now. Encorafenib/binimetinib is the latest one to come out. We have vemurafenib and cobimetinib and we have dabrafenib and trametinib. So all three of these are really, really effective combinations, and how do we decide among these three different options? Which one is the best? Because really all of them have really promising efficacy, and I think a lot of the decisions for this go on side effects and perceived anticipation of tolerance for side effects. DR LOVE: Just to clarify, though, in your mind — and as you mentioned, encorafenib/binimetinib is the most recent data, so I’m guessing the follow-up is not as long — but is there any way — reminds me when people talk about CDK inhibitors in breast cancer and PARP inhibitors in ovarian cancer. Is there any way to differentiate benefit from these combinations? DR POSTOW: The only way that you could really distinguish which one might be better than the other is with the encorafenib and binimetinib combination. Because in that randomized clinical trial, which was the COLUMBUS trial, vemurafenib monotherapy was one of the comparator arms. And so in that study, it did look like encorafenib was a little bit better of a BRAF inhibitor than vemurafenib. So although none of the other BRAF/MEK combinations were tested against each other, in that one study, encorafenib was kind of tested against vemurafenib, or at least encorafenib and binimetinib were tested against vemurafenib alone. And it looked like encorafenib is a little bit better of a BRAF inhibitor, actually. And so I think that it may be that the encorafenib and binimetinib combination has the best efficacy. And when you look at it, kind of cross-trial comparisons with all those caveats and all the kind of hesitation with how do you really decide on what to do, I look very favorably on the encorafenib and binimetinib combination. And I think that as we get more experience with it, because it is the newest, we’ll really try to understand, is that the go-to first BRAF/MEK combination, or do we try one of the ones we have a little more familiarity with, like dabrafenib and trametinib or vem/cobi first and then switch to encorafenib and binimetinib subsequently? And I think we’re still trying to figure this out. But I would say that because there’s that randomized data with encorafenib against vemurafenib and it looks like the encorafenib might be a little bit better, at least in the context of 1 study, I would kind of favor encorafenib and binimetinib or, perhaps, dabrafenib and trametinib. DR LOVE: I want to ask you in terms of toxicity profiles, but that’s interesting that you brought that up. I’m curious. We have this thing, we’ll go out and ask 25 investigators stuff — breast cancer, do you use pertuzumab, whatever it might be — just, like, what do you think? What do you think would happen if we asked 25 melanoma investigators? Do you think most would agree with you that they kind of have this feeling maybe there’s greater efficacy with the encorafenib/binimetinib? DR POSTOW: I think so, yes, at least over vemurafenib and cobimetinib, or at least the vemurafenib kind of story with the BRAF inhibition specifically. Dabrafenib and trametinib has never at all been compared to the encorafenib and binimetinib. And some could say, “I go with dabrafenib and trametinib because it’s had the most extensive follow-up, the longest-term data, the most number of randomized studies with survival benefits.” So dabrafenib and trametinib incredibly established in this area proven benefit in the CNS with dabrafenib and trametinib where we don’t formally know that for encorafenib and binimetinib. But I think a lot of investigators in the melanoma field are very excited about encorafenib and binimetinib and do generally feel that it’s perhaps the most effective BRAF/MEK inhibitor combination. And whether that means it will take over for dabrafenib and trametinib or take over for vemurafenib and cobimetinib still remains a little bit of a question. One of the issues is, it is a lot of pills to take with encorafenib and binimetinib. It’s 6 pills of the encorafenib every morning and 3 pills of the binimetinib in the morning and 3 pills in the afternoon. So it’s a lot of pill taking, and dabrafenib and trametinib has the advantage of fewer pills, so it’s 1 trametinib pill in the morning every day and 2 dabrafenib in the morning and 2 in the afternoon. So there’s a lot of these kind of nuances. There’s issues about whether some of the pills need to be refrigerated, like trametinib needs to be in the refrigerator. So there’s some kind of different feelings about kind of which patient might be compliant with what kind of treatment. Cobimetinib, with the vem/cobi experience, cobimetinib is taken 3 weeks on and then 1 week off. So if you’re worried about patients that might be confused taking a pill for 3 weeks and then taking a week off and then going back on it, there’s all these kinds of decisions. And I think it’s great that we actually have 3 different BRAF/MEK inhibitor combinations, so that we can try one out for a patient. If it doesn’t work from a side-effect perspective, we just switch to another one. And I hope that over time, insurance companies will understand that we might need to change from one to another due to tolerability. Side-effect profiles of BRAF/MEK inhibitor combinations DR LOVE: It’s interesting, as I said, that you brought up this issue of benefit, because I think what I’ve been hearing about encorafenib and binimetinib is the question of whether it’s better tolerated. How do you compare the toxicity profiles and quality of life with these three regimens? DR POSTOW: I think that if you look at the data, encorafenib and binimetinib does have a lower risk of fever and chills than dabrafenib and trametinib, and I think that’s one of the major problems with dabrafenib and trametinib. You have patients with rigors and chills and high, high fevers, and sometimes they have to go to an emergency room because of that. Vemurafenib and cobimetinib has a lot of skin toxicity. I think that’s kind of the main thing, sun sensitivity and rashes and some GI issues. Encorafenib and binimetinib, if you look at the clinical trial data, I think it does seem to be the best tolerated BRAF and MEK combination. That might be the case because it’s just newer, and maybe it’s our feeling that we have a lot of experience with toxicities of the other BRAF and MEK inhibitor combinations and, therefore, it’s kind of our old friend. We know them, good and bad and otherwise with dab/tram and vem/cobi, and we’ve seen all the toxicities with those. And so we’re attracted to the new kid on the block, the encorafenib and binimetinib, just because we haven’t had as much experience with it yet. And we look at the clinical trial data, but we haven’t really treated patients to a great extent with it, so we don’t really know a lot about the toxicities as much, other than looking at the clinical trial data. But I do think there is this notion that’s out there that, perhaps, it has a better tolerability profile than either dab/tram and vem/cobi. And so I would say, I think, that it remains a very attractive first-line option for patients with BRAF-mutant melanoma when they’re considering either initial treatment or treatment after immune therapy to go with encorafenib and binimetinib. I’m personally excited about it. And I think we have to try to see which of the patients for which BRAF inhibitor combination. I’ve come up with little algorithms in my mind about who might be best for one versus the other. I don’t know if this is correct, but in the end of the day, we have to choose. And so if we kind of try to think about this. And historically, older patients, I’ve found, don’t tolerate the dabrafenib and trametinib fevers very well. So that’s a patient I might think for vem/cobi or for encorafenib and binimetinib. Younger patients could tolerate the fevers with dab/tram better, and so I think that’s, obviously, a good choice for many of them. And then patients who are out in the sun a lot or being very active, the vem/cobi rashes might be more problematic. So these kinds of things I’m still trying to figure out, even in my own practice, who should get what. And a lot of times, honestly, it’s dependent on insurance approvals, what drug can you get in the patient’s mouth the fastest. So we try one. If it doesn’t work with the insurance companies, we just go for another one very quickly. Use of dabrafenib/trametinib and consolidation radiation therapy for patients with metastatic melanoma and BRAF tumor mutations DR LOVE: What happened with this lady? First of all, what did she get started on first, the — I see she got dabrafenib/trametinib. Did she get that, or the radiation? What was the timing? DR POSTOW: So started with dabrafenib and trametinib, and then radiation was incorporated as we went along. It wasn’t that the dabrafenib and trametinib — the dabrafenib and trametinib was effective. It shrunk the tumors right away. And it was kind of a question of should we just rely on the dabrafenib and trametinib alone or kind of consolidate that response with additional radiotherapy and — DR LOVE: Wow, interesting. Consolidation. DR POSTOW: Yes, consolidation with some radiation to try to just further optimize on the dabrafenib and trametinib initial response. And that’s kind of a questionable clinical decision. I was just very attentive to some of this epidural disease and how much the back pain was causing — DR LOVE: Sure. DR POSTOW: — a problem. And you had confidence that dabrafenib and trametinib, obviously, was going to work. But in these younger patients, I really think we have to offer everything that might help. And knowing that radiation can be effective in melanoma, I wanted to just offer everything that could be helpful. And so we did ultimately, additionally, give radiation for this patient. But she’s still continuing to do very well. It’s early along in the course. I think all the treatments started about 4 or 5 months ago. But doing really well in general, so I’m very pleased. DR LOVE: Again, how long was it from the time you started the dabrafenib/trametinib to the time she got the radiation? DR POSTOW: So some radiation to some other bony areas that were problematic happened right away. And some of the spinal radiation happened after a couple of months of dabrafenib and trametinib. DR LOVE: Interesting. And has she had any toxicity issues, any fevers, anything? DR POSTOW: I’ve been lucky with her. No fevers and chills so far and no other toxicities with the — DR LOVE: Nothing. Huh! DR POSTOW: — dabrafenib and trametinib. So that’s really remarkable, and many people just won’t have anything at all. And I think we need to try to understand, who are these patients that can tolerate this without any problem? And then who are the people — I’ve had other patients that take 1 dabrafenib pill, and then they’ll start getting a low-grade fever. So people are just intrinsically sensitive to these treatments. And this fever and chills problem is a real issue for some people, but other people don’t have any problems at all. So I’ve been very, very lucky with her. And the question really is, how long do you stay on BRAF and MEK when it’s working so well, or do you stop this after a certain period of time and switch to immune therapy? Because you worry about resistance. I think those are open questions in the medical literature thus far. We don’t really know you should treat people for 4 months and switch, or you should treat people for 6 months and switch. So I’m going to try to milk this out as long as I can, given how well she’s tolerating the treatment and how well it’s working for her, following carefully, of course, every couple of months with scans. And at the first sign if there’s any resistance, then make the switch to immune therapy. DR LOVE: So just out of curiosity, roughly what fraction of your patients who get combination targeted therapy like cruise, no problem, let’s say, dabrafenib/trametinib? How often do you see that like this? DR POSTOW: Most people will respond very well for the first 6 months to a year. So that’s, I would say, the vast majority of patients will have a good 6 months to a year run of dabrafenib and trametinib. DR LOVE: I’m talking about tolerance, side effects. DR POSTOW: Oh, from a tolerance perspective, yes. So I would say very few people have absolutely nothing. DR LOVE: But she’s unusual. DR POSTOW: She’s unusual, yes. Most people have something. Not necessarily side effects that are so bad that it limits ongoing dosing. So you could play with different dosing schedules, give it intermittently 4 days on/3 days off or 1 week on/1 week off. Sometimes if people have had bad fevers and chills, you can give them low doses of prednisone, 5 or 10 mg of prednisone a day to see if that can reduce the fever risk. Or NSAIDs, like ibuprofen, sometimes that’ll help with fevers. But this is a little bit unusual. It’s really, like, zero toxicities with the dab/tram, and it’s amazing. And just, I guess, to speak to the fact that you never really know who’s going to cruise through this without a problem at all. Adjuvant therapy options for melanoma DR LOVE: So that kind of leads into the issue of treatment in the adjuvant setting where it’s kind of a different mindset and also in terms of how tolerability/toxicity play out. It’s, I guess, been about a year since the world changed, ESMO 2017, in the fall of 2017. All of a sudden, PD-1 is out there and targeted therapy as adjuvant therapy. So to kind of get into that, I’d like to hear about your 72-year-old lady. But before you present her, can you just kind of summarize where we are 1 year later in terms of data? I don’t know if there’ve been updates to those original trials that were presented. But globally, when you look at this adjuvant question today, what are some of the key data points that you think about? DR POSTOW: The main updates in adjuvant treatment for melanoma are for Stage III patients that have resected lymph nodes, and what we have are recurrence-free survival data for both dabrafenib/trametinib and PD-1 drugs. So for dabrafenib and trametinib, we know that it improves recurrence-free survival compared to placebo. And for the PD-1 drugs, we know that pembrolizumab improves recurrence-free survival over placebo and nivolumab improves recurrence-free survival over ipilimumab adjuvant. So this is a win for both PD-1 drugs, nivolumab and pembrolizumab, and a win for dabrafenib and trametinib. The difference among some of these adjuvant studies is that the control treatment varied in some of these studies, so pembrolizumab beat observation, and dabrafenib and trametinib also beat observation in terms of recurrence-free survival. And then nivolumab beat an active control, ipilimumab. Now, that doesn’t mean nivolumab is better than pembrolizumab in the adjuvant setting. We have really no idea if that’s the case, because some of these trial populations differed a little bit in terms of what risk profiles some of these patients were in when they started these adjuvant trials. And we really have no idea what’s better between targeted therapy in the adjuvant setting or PD-1-based immunotherapy in the adjuvant setting. And, frankly, we don’t really know, how much better are any of these drugs compared to standard adjuvant old treatments like interferon? So randomized data with interferon are still coming. And we really still have very little information about some of these drugs in terms of overall survival. DR LOVE: So what about relative risk reduction in recurrence? With these 3 types — understanding it’s indirect, one of them compares to active treatment, but even bringing in interferon, if you think about the old data. Because I kind of always felt like it was really borderline. It was hard to see the benefit. But in any event, what are the hazard rates that you see in these settings? DR POSTOW: So the hazard ratios in some of these trials are reflective of approximately 30- to 40-ish percent reductions in risk of recurrence with PD-1 and/or BRAF/MEK inhibitors. So hazard ratios 0.57, 0.6-ish, those kinds of hazards ratios. So I think these are very, very strong hazard ratios. And in some context, it’s, like, 50/50 in terms of risk reduction with some of these hazard ratios. So these are really powerful recurrence-free survival-improving drugs. DR LOVE: What about with the nivo/ipi comparison, the hazard rate there? DR POSTOW: The hazard ratio there I think is 0.67, so it’s pretty favorable for nivo over ipilimumab. So I think I’m very convinced that adjuvant ipilimumab is done, frankly, I think, when you have a better tolerated drug with much more favorable hazard ratio in terms of reducing recurrences. DR LOVE: Just out of curiosity, though, in the old interferon — maybe not that old, but what kind risk reduction was seen? Did they do randomized trials? DR POSTOW: They did. Interferon was compared to — DR LOVE: Right, right. DR POSTOW: — observation, mostly, in some of these trials. And almost always, high-dose interferon and pegylated interferon, they had improvements in recurrence-free survival nearly kind of across many, many different trials. So really clear recurrence-free survival benefits. I think the controversy with interferon is all the toxicities that it caused — DR LOVE: But again, in that kind of range of 30%, 40% relative reduction? DR POSTOW: It was less effective. DR LOVE: That’s what I thought. Right. Yes, of course, the toxicity was a gigantic issue. DR POSTOW: I think one of the challenges in melanoma is really trying to understand how much benefit are we offering to patients and what is their baseline risk? And we have a new AJCC staging system that has just come out. So I think we’re trying to wrestle with all of this. And one of the big challenges in adjuvant for melanoma is, the new AJCC system came out after or kind of in parallel to when these adjuvant trials were reported. So the adjuvant trials were formally reported with the AJCC 7th edition staging — DR LOVE: Wow. DR POSTOW: — so we know kind of relative risk improvements for people with AJCC 7. However, for AJCC 8, we’re now retrospectively going back and trying to relook at these adjuvant trials. But with the AJCC 8 — and there are some abstracts that have been published on that topic — but I think we’re wrestling with new adjuvant data with an old staging system. And now we have a new staging system, but so — if there’s a way to kind of synthesize this so we can sit down with the patient, use the most refined staging system that we have that’s up to date and then know how that might apply to the trial data, I think that would really help inform our conversations with patients. One of the also — challenges from what we had from the AJCC staging system in general is, it still only reflects overall survival, and it doesn’t reflect recurrence risk. So in an era now where thankfully we’ve come up with better new treatments and a recurrence event in and of itself does not necessarily mean a patient’s going to die from their disease anytime soon. I think we have to recognize that recurrence in and of itself may be divorced from overall survival. So as much as we can try to get data sets where we have recurrence-free survival broken down by risk profile, I think it would be helpful and at least in melanoma where this field is so rapidly changing. I think that would be a welcome feature if we could have that. Case: A 72-year-old woman with a history of primary biliary cirrhosis presents with an ulcerated lesion on her left arm and a mass in her axilla and is diagnosed with Stage III melanoma DR LOVE: So, of course, the big question is, what do you do with the BRAF-positive patient? And I remember hearing about that — I think it was actually even before ESMO, because the data were out there and hearing people talking about this. I’m curious how it played out in this 72-year-old lady. What was her situation? DR POSTOW: Yes, so this woman that I saw, she was 72. She was a little bit different, because she had this history of BRAF-mutant Stage III melanoma that was resected. And she had a history of this primary biliary cirrhosis, so this kind of autoimmune biliary process. So in this patient with a BRAF mutation, clearly I think the right choice would be a BRAF and MEK inhibitor combination in the adjuvant setting. And I think one of the real clear messages is, BRAF and MEK inhibition in the adjuvant setting is really, really effective. And there’s actually some ways you can splice through the data to suggest that it may even be a little more effective, at least in the first year, than adjuvant PD-1. DR LOVE: So the reason you picked targeted therapy, I was assuming that you didn’t want to give her the immune therapy because of the primary biliary cirrhosis. Even if she hadn’t had that, do you usually use BRAF as adjuvant therapy? You prefer it? DR POSTOW: In anyone with a history of autoimmune disease in the adjuvant setting particularly, and a BRAF mutation, I think BRAF and MEK is the right choice in those patients for sure. Let’s put the autoimmune history to the side. Choosing between dabrafenib/trametinib and an anti-PD-1 antibody as adjuvant therapy for melanoma with a BRAF tumor mutation DR POSTOW: How do we think about BRAF/MEK versus PD-1 adjuvant, which is, I think, the majority of the decision-making for a BRAF-mutant patient with Stage III resected melanoma. I think one could think about this in terms of toxicity. So I think the efficacy looks very, very similar. If not, maybe the first-year BRAF/MEK even has more favorable recurrence-free survival advantages and even a more favorable hazard ratio in some contexts. However, the toxicities are very, very different. And so how do I think about that for an adjuvant patient because adjuvant patients might already have been cured from their surgery? I think the issue is that BRAF and MEK toxicities are transient. So when you’re taking the pills, you may have toxicity, and when you stop the pills, yes, the toxicity may linger a little bit, but it gets better over time once you stop. And it’s extraordinarily unlikely to have any permanent toxicity with BRAF and MEK inhibitors. And with immune therapy, PD-1-based immune therapy, which we have in the adjuvant setting, although the majority of people will feel totally fine and the majority of patients don’t have any significant toxicity at all, there are those people that will have really significant and/or permanent toxicities from adjuvant PD-1. And so it’s a conversation with the patients. Do you want to take a risk of a long-term, permanent side effect that might be severe with PD-1-based immune therapy, although the majority of people are fine? Or do you want to take pills like BRAF/MEK inhibitors that are probably more likely than PD-1 to cause side effects? But the good news is, if you stop the pills, you’ll get better and everything will be okay. And so it’s a tough patient conversation and a tough decision to decide who wants the pills that’ll probably make them feel less well, but if you stop them, they’ll feel better. And who wants to get the IV treatments with the PD-1 drugs that, most likely they’ll be fine, but there are some permanent toxicities like the endocrinopathies, like vitiligo, which can be permanent, whitening on the skin. And that can be a big cosmetic problem for some people with PD-1, especially in the adjuvant setting where we’re still trying to figure out who really needs these treatments. DR LOVE: That’s a really interesting perspective I actually haven’t heard before, makes complete sense. I guess now, though, really, I mean, obviously, if a patient like this were wild-type BRAF, PD-1 would be on the table almost as a no-brainer. But I am really curious what you would have done in her with the primary biliary cirrhosis. I’m curious — I guess one of the things that would have to be considered is what you think her baseline risk of recurrence is. She presented with a left axillary mass, 1 node. Where was the melanoma? DR POSTOW: She had an ulcerated melanoma on the left arm that was resected. DR LOVE: Oh, interesting. Ulcerated — and so she had 1 node, but 18 nodes were negative. So globally when you looked at her — and I don’t even know, maybe it comes up, maybe you actually talked to her about it — what were you estimating her risk of baseline recurrence would have been? DR POSTOW: We have a new melanoma staging system now, which is really great, and so, unfortunately, that it only shows melanoma long-term survival, so it doesn’t really get to the recurrence risk. But you can imply what a recurrence risk might be. So with an ulcerated primary melanoma of moderate depth and a very large involved lymph node in the axilla that was surgically resected, hard to really be definitive, but recurrence risks in these situations can really be almost up to 50-ish percent. And so that’s a pretty high risk of recurrence. And that’s a patient that I would consider whether or not I should still treat her with PD-1-based immune therapy even if she had a BRAF mutation. Even with an autoimmune history. Her primary biliary cirrhosis was generally well controlled. She’s on ursodiol but not a patient that had active hepatitis or on immunosuppressants already. So she had this history, but it wasn’t such a severe autoimmune disease and had a pretty decent risk of recurrence of around 50/50. That would be someone, even without a BRAF mutation, I would consider for adjuvant PD-1. But people that have microscopic lymph node involved, a relatively shallow primary, no ulceration in the primary, those kinds of earlier Stage III patients — I’m really still trying to wrestle with who really needs adjuvant PD-1 at all in those cases, and if a patient had any history of autoimmune disease in those cases, clearly I wouldn’t take the risk for those kinds of patients. So she was a higher-risk Stage III patient with the size of the lymph node involved and the ulceration in the primary. So that would be someone I would maybe take more of a chance. But the whole point here, I think, is a good one: You have to really risk-profile these people in terms of what is their real Stage III melanoma risk and what is their risk of the treatment in and of itself? Understanding the mechanisms of autoimmune toxicities in patients receiving immunotherapy DR LOVE: So wow, this has been so interesting. We could go on for hours. But there are a couple more things I just wanted to finish out with. I was thinking about the issue of mechanisms of autoimmune disease. Like, what’s the difference between primary biliary cirrhosis and rheumatoid arthritis and Crohn’s disease in the immune system? I assume it’s not all the same. And which pathophysiology do you think would be more likely to be affected by immunotherapy? I mean, I don’t know if those kinds of data are out there. Any thoughts about that? DR POSTOW: I think it’s a really important issue, because we need to learn from our colleagues who’ve been managing autoimmune diseases for a very long period of time. Whether it’s a GI doctor who’s been handling ulcerative colitis and Crohn’s disease or even autoimmune pancreatitis. And from our rheumatologists who are handling rheumatoid arthritis and myositis or polymyalgia rheumatica. They have been using all kinds of selective immunosuppressants for a very long time in those diseases, whether they’ve been IL-17 blockers or TNF alpha blockers as in the case of Crohn’s and ulcerative colitis. And so in the immune therapy world for treating patients with cancer, we need to really try to figure out, are the mechanisms of the autoimmune toxicities that we cause from our drugs similar or different from underlying autoimmune diseases in general? And then we can learn from each other. We can learn from the rheumatology world, and they can really learn from us. And one of the very interesting mechanisms that’s out there at the moment is that there is a drug in rheumatology that’s used for rheumatoid arthritis that’s a CTLA-4 agonist. So instead of blocking CTLA-4, like tremelimumab or ipilimumab would do, it enhances the function of CTLA-4. So that’s used to treat people with rheumatoid arthritis. And that’s a very interesting concept, because it tells us that you can agonize some of these pathways to treat patients with autoimmune disease. And it makes us wonder should you agonize pathways related to PD-1 or PD-L1 blockade to help people with autoimmune conditions. We know that when you block PD-1, you can cause an autoimmune diabetes. So could we be reevaluating diabetes in general for patients without cancer? Is routine insulin-dependent diabetes that’s caused from beta islet cell in the pancreas destruction from autoimmune phenomenon? What can we learn by the fact that PD-1 blockade causes diabetes in patients with cancer? What can we then use in patients with diabetes to, perhaps, agonize PD-1 or learn more about autoimmune pathways in people with diabetes to try to help the vast far more patients with these autoimmune conditions than patients with cancer getting immune checkpoint blockade, and how can we work with each other? I think that’s a really important area, and we need this cross-pollination of rheumatology and other people to enter into the cancer treatment field so we can talk to each other, try to learn from each other and try to help the vast number of people with autoimmune conditions. I think it’s a great topic and I think one that we still don’t know a lot about. Perspective on the use of adjuvant therapy versus observation for patients with melanoma and BRAF tumor mutations DR LOVE: I think the other thing is actually, maybe — I know it’s obvious — I mean, everybody wants to talk about what you do in the BRAF-positive patient, which one do you use, which strategy do you use? But based on what you were talking about, I think also the issue of whether you do anything. The lower-risk patient — I mean, PD-1’s not a walk in the park necessarily, or BRAF. DR POSTOW: Absolutely correct. We don’t know really that all these patients should be getting adjuvant treatment at all. And the reason we don’t know about that is because we don’t have overall survival information. The adjuvant COMBI-AD study with dabrafenib and trametinib did report some overall survival information, but it’s not at the final analysis, so we can’t really say statistically that overall survival was improved with adjuvant dabrafenib and trametinib. But for the PD-1s, we have no overall survival information in the adjuvant setting. So I think it’s very justifiable to not use adjuvant treatment in some patients, especially those that are really low-risk Stage III, because we don’t know if the survival is better by giving adjuvant treatment at the time of surgical resection or only treating the people that have recurrent melanoma in the metastatic setting. And some of the adjuvant PD-1 studies, namely the one testing pembrolizumab in the adjuvant setting, are being designed to test that question specifically: Do you treat people now after surgical resection of Stage III melanoma, or do you only treat the people that have a recurrence of melanoma in the metastatic setting? And is there that survival advantage overall for treating people in the adjuvant setting? That’s still an open question. And until there’s an overall survival benefit that’s proven for any of these drugs, I don’t think clinicians need to feel compelled to offer their patient — I think that they should offer their patient adjuvant and talk about the risks and benefits, but I don’t think observation is absolutely wrong, especially for certain patients, and particularly people with low-risk profiles. I think many patients I tell, “We should just follow you. I don’t think you need these drugs in the adjuvant setting, because your Stage III risk profile is so favorable.” DR LOVE: Again, this is comes up all the time in other parts of oncology. I think people pretty much agree: HER2-positive breast cancer, you add pertuzumab. You get about a 20% relative risk reduction in recurrence. No survival data yet. And you take the absolute risk that you’re looking at, apply the 20%, you get a number. And yes, maybe you don’t know about survival, but if it turns out to be less than 5% absolute difference in recurrence, at least you have that information. That’s way different than 50/50 chance you’ll avoid a recurrence. DR POSTOW: Absolutely. And I think it’s an interesting general fundamental philosophical question of how does a patient anticipate and feel about a difference in benefit versus how does a physician anticipate and feel about a difference in benefit? What number should we be quoting? We should be quoting relative risk reduction. Should we be quoting absolute risk reduction? Should we use something like number needed to treat? How many people do we need to subject to the toxicity of these drugs to have one less recurrence? These are important questions, and we haven’t completely fleshed them out yet, especially in melanoma. These adjuvant data are really, really still early. We have follow-up less than 2 years in many of these cases. And so I think that’s still such a short period of time to make some of these decisions. So we’re really still grasping in the dark about who should we feel should get these different treatments. So I think it’s really hard to be really dogmatic at this stage because the data, ultimately, could end up all over the place. Case: A 75-year-old man who presents with a large mass on his right cheek is diagnosed with locally advanced squamous cell carcinoma (SCC) of the skin and receives pembrolizumab DR LEWIS: So this is a guy who came to us — he’s 75 years old and presented with this large, raised bleeding mass on his right cheek and was pretty much taking up the majority of his right cheek. And he also had about a 3-cm area above the right eyebrow that was ulcerated and oozing. And biopsies of both of those sites came back as cutaneous squamous cell carcinoma. He was seen by an outside surgeon who was really planning a pretty radical resection — was going to lose a lot of the right side of his face — and he obviously got very concerned about that and came to us. And we saw him and decided to start a PD-1 antibody, pembrolizumab. DR LOVE: Before you go on, could you just kind of backtrack a little bit and talk a little bit more about him, what his social situation was and why it was that he had such advanced disease? DR LEWIS: Yes. So a lot of these patients that present with these advanced nonmelanoma skin cancers that get like this, he falls into that kind of typical person, which is somebody who tends to be isolated. He’s a farmer, lived out on a farm. He was largely retired from farming, and he was divorced and by himself. And he just was isolated and let this thing go and didn’t seek medical care. And finally came in when it got to the point where it was unbearable. DR LOVE: And had he received medical care for other reasons, or he was separated out from the medical doctors? DR LEWIS: By and large he was separated out. He hadn’t seen a medical provider in a number of years. DR LOVE: And I’m curious, how often do you see metastatic disease in this situation? DR LEWIS: We see it, but what I tend to see in my clinic, for whatever reasons, tends to be more these locally advanced tumors. And they can be either large like this or just in a bad situation. So some people who present with these will be in, for example, with a lesion that’s very close to the orbit. And they’ve tried resections, and the resections have not really succeeded in removing tumor, so they don’t have a huge bulk of tumor, but it’s just tracking into bad places. So those are really the 2 main people I see in terms of this locally advanced cohort of patients, so to speak. And then we also do see the occasional metastatic patient. DR LOVE: What other parts of the body, besides the face and head, do you see this on? DR LEWIS: It’s mainly the head and neck area, but we have seen some on the extremities. I have 1 patient who had a large squamous cell carcinoma on the arm, and he ended up actually being metastatic. A lot of these locally advanced are more head and neck cases. Pathophysiology and management of SCC of the skin DR LOVE: Before you go into what happened with him, could you just talk a little bit about what we have known in the past, or up until now, about systemic therapy for these kinds of tumors? DR LEWIS: Systemic therapy for advanced squamous cell carcinoma, especially in the skin, has really been extrapolated from head and neck cancers by and large. So the chemotherapeutic regimens would be the same, but from a nonchemotherapy standpoint, what we’ve used for years is monoclonal antibody cetuximab. And there’s data, a small series, in advanced cutaneous squamous cell carcinomas with reasonable response rates and reasonable toxicity profile. And so prior to PD-1 antibodies, really cetuximab was my go-to first therapy in these cases. DR LOVE: And from a pathophysiology or pathogenesis point of view, do you see these as similar to squamous cell cancer of the head and neck? DR LEWIS: They’re certainly, for a lack of good data, treated that way. But I do think that they’re somewhat different in the sense that if you look at these tumors, they have a very high mutation burden from chronic sun exposure. And they’re also very common in people who are immunocompromised. So people who, for example, are solid organ transplant patients on immunosuppression, they have a very high incidence of developing cutaneous squamous cell carcinomas. So they’re probably different. But that’s what makes them good candidates for immunotherapy — high mutation burden and likely have immune surveillance as some role in controlling the disease. Cemiplimab, a novel PD-1 antibody for locally advanced and metastatic SCC of the skin DR LOVE: So what do we know right now about the use of checkpoint inhibitors in patients with these kinds of tumors? DR LEWIS: These checkpoint inhibitors, there were some case reports and now there’s starting to be some prospective studies. And in The New England Journal of Medicine, there was a clinical trial of cemiplimab, a novel PD-1 antibody that was given to patients with metastatic squamous cell carcinoma, as well as a small cohort of — the published cohort was a small cohort of locally advanced. There’s a larger cohort that’s yet to be published of locally advanced patients, with very encouraging response rates. DR LOVE: Before you go on, is this an anti-PD-1? PD-L1? Or how does it compare to the agents that are out there right now? DR LEWIS: It’s an anti-PD-1. How it compares to the other agents, I think it’d probably be just in that class that’s certainly effective in this group of patients, as seen by the data from the New England Journal paper. But there’s not going to be any head-to-head comparisons with the other available agents like nivolumab or pembrolizumab. Activity and tolerability of cemiplimab observed in Phase I/II studies DR LOVE: In terms of tolerability or toxicity or particularly immune-related toxicity, same basic story? DR LEWIS: Same basic story. There didn’t seem to be any increased immune toxicity with this agent. It was fairly well tolerated, so there wasn’t concerning signals with this drug. DR LOVE: So in that New England Journal paper, it looks like there was only 26 patients, but half of them responded, which, I mean, it is a small number but kind of higher than you see with a lot of solid tumors. Any thoughts about that and the depth and duration of response? DR LEWIS: Yes, and that’s what’s exciting about it is the depth of response was very encouraging, and it seems to follow what we see in other tumor types with PD-1 antibodies in cutaneous malignancies at least, which is, you get a very good early response and that response tends to be durable. So I think this is going to pan out to be an effective therapy for this patient population. Pathogenesis of SCC of the skin and potential role of cemiplimab in management of this disease DR LOVE: Anything about the pathogenesis of this cancer? In Hodgkin lymphomas it’s very, very sensitive to checkpoint inhibitors, and they have this PD-L1 amplification, MSI, maybe seems like it’s a little bit high. Anything about this that might account for the high response rate? DR LEWIS: There’s some data out there looking at cancers and looking at their so-called tumor mutation burden. And tumors with higher mutation burdens, in theory, have increased neoantigens and, therefore, are more visible to the immune system. And certainly if you look at that spectrum of tumor mutation burden, cutaneous squamous cell carcinoma is up there with a very high mutation burden. And, theoretically, that might be a reason they’re good candidates for immunotherapies like this. DR LOVE: Interesting. What about other checkpoint inhibitors? What kind of clinical or anecdotal case report data do we have? DR LEWIS: There’s a small study with pembrolizumab in cutaneous squamous cell carcinomas that showed a response rate as well. So I think this is probably a class effect of the immunotherapies. DR LOVE: I want to ask you more about this patient, but anything else you want to say about this agent cemiplimab and the New England Journal paper that you published there on it? Anything else about it that you want to comment on? DR LEWIS: What I would say is that I think there was great need for this trial, because this is a patient population that really, there’s no standard effective therapy. Most of the data that we have is extrapolated from other tumor types, so getting some prospective data was very important. And I think in that regard, it was a very important and useful trial to do. Durable responses to pembrolizumab in patients with SCC of the skin DR LOVE: This patient got pembrolizumab. What happened? DR LEWIS: We got pembrolizumab for him and — off of protocol — he was not a clinical trial, but we were able to obtain it, fortunately. And his response was dramatic and immediate. And within a few months, that large ulcerated tumor mass on his cheek had nearly resolved. And he had a clinical complete response within a matter of about 4 or 5 months. He was interesting as we followed him and continued to treat him on pembrolizumab. Almost a year into therapy, he had a severe accident that was unrelated to the treatment. And he was hospitalized for an extended period of time. Had some rehabilitation. And so we held therapy during that hospitalization. He recovered. He’s back to his baseline. But since he clinically had a complete response, we decided to just stop therapy. And we’ve been following him now, coming up on a year without recurrence. So he went from looking at a disfiguring surgery to a complete response with the immunotherapy and is now followed off treatment. DR LOVE: Really amazing. Yes, it’s interesting, you started to hear, as the checkpoint inhibitors came out, I would hear stories, cases of people who got stopped because of toxicity. In this case an unrelated problem, motor vehicle accident. And I guess we saw that maybe you don’t need necessarily to give checkpoint inhibitors indefinitely, particularly to people who are having such responses. The other thing that’s interesting about your paper is the waterfall plot. It looks like almost everyone is kind of going down. DR LEWIS: Yes. So even though they didn’t meet criteria for response, there’s clearly clinical benefit. And that’s one of the difficulties with these cutaneous malignancies, these locally advanced, is trying to assess response, because you still see things. And that’s true with the basal cell carcinoma studies too, where you still see defects on the skin, but whether that’s viable tumor or just scar is often hard to say. So the response rate may be even underreported. DR LOVE: Did this man have any tolerability problems with pembrolizumab? DR LEWIS: Absolutely none. He did very, very well. DR LOVE: Interesting. Emerging data with anti-PD-1 checkpoint inhibitors in combination with anti-LAG-3/TIM-3 antibodies DR LOVE: What’s the next step in terms of treatment of this disease? Any new trials out there, new strategies being looked at? DR LEWIS: There’s a new trial that we’re starting at our institution that’ll look at adding additional monoclonal antibodies to this therapy, just as there are in a lot of the other tumor types that are, for some reason, not responding to single-agent PD-1 therapy. So looking at other checkpoints, like LAG3 or TIM3 — the study that we have that will open for this cohort of patients will be a PD-1 plus a LAG3 antibody. DR LOVE: Can you talk more about that whole concept of other checkpoints and what LAG3 is? And also, I guess at least one agent is relatlimab, what we know about these. DR LEWIS: The data is still emerging, so we don’t know a whole lot about them. But when you give patients with cutaneous malignancies, in particular, is what I specialize in, you see patients who are either primarily refractory to the treatment — that is, they never respond and they continue to grow — or they’re secondary refractory — that is, they have an initial response and then their tumor starts to grow for some reason. And it’s really not clear why that is. In terms of these secondary refractory, there is some T-cell exhaustion developing with other checkpoints then becoming overexpressed, like TIM3 and LAG3. And, therefore, if you can add blocking antibodies against those agents, you may be able to, again, induce the immuno response. So that’s where the studies are going. Whether they’ll pan out or not in the long run remains to be seen, but certainly very encouraging. DR LOVE: Do these agents have any single-agent activity? DR LEWIS: Not a whole lot. I think really the PD-1 backbone is going to be the key. Case: A 78-year-old man with recurrent, locally advanced basal cell carcinoma (BCC) receives the hedgehog inhibitor sonidegib DR LOVE: So let’s talk a little bit about your 78-year-old man. Listening to this incredible story of your patient with squamous cell, it made me think about basal cell, because it seems like that issue of locally advanced disease is a big issue with basal cell. So can you talk about your 78-year-old man? DR LEWIS: Yes. So this is a patient with locally advanced basal cell carcinoma. And again, these nonmelanoma skin cancers are very similar, basal cell and squamous cell. But he has some comorbidities. He’s diabetic, hypertensive, coronary artery disease. And he came to us a number of years ago, and he had this infiltrated mass, the right nasolabial fold, and it was extending inferiorly, involving his upper lip. And you can feel the tumor underneath into the gum. So it was quite extensive local involvement. He had a basal cell carcinoma resected from that area previously, and so this looked like a local recurrence of that previous resection site. And biopsy was consistent with basal cell carcinoma. DR LOVE: Was it your impression that he delayed seeking medical care? Or how did this get so advanced? DR LEWIS: Yes. This was another case of delayed medical care to some extent, not as bad as we’ve seen others, but he did delay somewhat. But there was also a component that a lot of this was subcutaneous and infiltrative and probably once he felt that nodularity, it was still quite extensive. So it was probably creeping up on him over the years. DR LOVE: Just out of curiosity, it almost reminds me a little bit of some of the discussion I had with breast cancer investigators about these occasional women who come in with very advanced locally advanced breast cancer. When you have patients like that, these kinds of tumors, I’m just kind of curious how you find them in terms of adherence to your treatment regimen. DR LEWIS: Oh, by and large, people with these locally advanced tumors are very adherent, in my experience. Now, a lot of these patients do delay seeking medical care and have avoided the medical community for a number of years. But I think once they reach that point where it’s decided, “Yes, I need to take care of this,” then they become quite compliant. And that hasn’t been an issue for us. DR LOVE: And again, that’s the same thing you hear in breast cancer. Efficacy and tolerability of sonidegib DR LOVE: So what happened with this man? How did you manage him? DR LEWIS: So initially when I saw him, we put him on sonidegib, actually, as part of a clinical trial. And he had a fantastic response. He had an ulcerated area by the nasolabial fold that cleared up. And the infiltrative mass got much better. There was still some scar there from the previous resection, so it wasn’t completely normal skin that we can see, but there clearly wasn’t any tumor that we can identify. And we kept him on the hedgehog inhibitor for quite some time. He received it for over 2 years. He did pretty well, but hedgehog inhibitors can be tough. They have an expected toxicity profile that’s relatively low grade. But if you have these chronic toxicities, even low-grade chronic toxicities for a long period of time, they become quite problematic from a lifestyle perspective. So he had hair loss. He lost his taste. And he was having muscle cramps. So he ultimately decided that he was going to stop the drug. So we lost him to follow-up after he stopped. And he recently came back, after 2 or so years off of therapy, almost with the exact presentation again — an ulcerated mass, an ulcerated area at the nasolabial fold, with clear tumor extension inferiorly, involving his upper lip. So he clearly had microscopic residual disease when he stopped the therapy a couple of years ago. And then over the course of a few years, it returned to basically where he started from at the beginning. DR LOVE: I was kind of curious. When he was taking the sonidegib, what was it about it that bothered him the most? DR LEWIS: Unlike some medications where there’s relatively severe what we call Grade 3/Grade 4 toxicities, there really isn’t that with the hedgehog inhibitors. They’re really Grade 1/Grade 2. But again, Grade 1/Grade 2 toxicities over 2 years get to be a problem from a quality-of-life standpoint. And so the muscle cramps were bothering him. The fact that he really wasn’t able to enjoy food for a couple of years period, it just got to the point where he said, “I’ve had enough of the therapy. I want to take…” he didn’t say he wanted to take a break, but he said, “I just want to stop.” DR LOVE: So what did you do when he came back with this recurrent disease? DR LEWIS: Restarted him on the hedgehog inhibitor and he had a very rapid response. And he continues on that now. He’s been on it for a few months, and the disease, visit by visit, is getting better and better. So he’s reresponding to the therapy, very similar to the way he did in the first place. DR LOVE: Same side effects? DR LEWIS: It’s still early, but the side effects are coming. He still has his taste a few months into therapy, but it’s decreasing, and the muscle cramps are returning. So the side effects, I think, will come at some point. Management of side effects associated with hedgehog inhibitors DR LOVE: Any supportive or even complementary strategies to deal with these side effects? DR LEWIS: No, that’s been looked at without really great success, as far as I’m concerned. Some things, like trying calcium channel blockers for the muscle cramps and those sorts of things. Not a whole lot of data and not too impressive, in my opinion. Because you also get into that issue that a lot of these patients have comorbidities, and do you want to start adding medicines on top of their other medications? What I find is the best way to manage these patients is to just give them drug holidays. There’s no dose reductions for these medications, either vismodegib or sonidegib — the dose is the dose. Some people have tried strategies of staggering doses. I don’t think that you get good steady-stage drug levels in those situations. So what I tend to do with patients on hedgehog inhibitors is treat them to side effect tolerability, and then I just have them take a break. And I have them stop for 2 to 3 months and let the side effects kind of get back more towards their baseline and then restart them and treat them again to tolerability. And patients are able to, I think, stay on the therapy for longer periods of time. Their quality of life is pretty reasonable. And in most of those patients, we control the tumor just fine. DR LOVE: So I’m guessing a lot of these patients you treat and then they go to surgery as opposed to being treated indefinitely. DR LEWIS: You’re using these hedgehog inhibitors sort of neoadjuvantly, trying to shrink the tumor and then make it amenable to resection. That’s a good thought. The studies haven’t clearly demonstrated that we can do that, because it’s not clear that you can change the surgery margins. We don’t know that the tumor shrinks uniformly and, therefore, if you do limit your resection size, you maybe leave — at higher risk of leaving microscopic tumor behind. But I think that that’s an individual case-by-case basis now. We certainly put it out there to try to do that, but the data is not convincing. DR LOVE: That’s interesting. So is that something you generally don’t do? Neoadjuvant therapy? DR LEWIS: I don’t see a patient and immediately plan that to say, “Okay, I’m going to give you this drug, try to shrink this and then you’re going to go for surgery.” It doesn’t mean we haven’t done it, but it’s something that we do after the fact, depending on their response and tolerability of the agent. DR LOVE: What about radiation therapy for a patient like this? DR LEWIS: Radiation therapy can be very effective for basal cell carcinomas. And we get our radiation oncologists involved in almost all of our basal cell patients, because it’s a radiosensitive tumor and something that we could have potential for good disease control. DR LOVE: Would that be something you’d consider in a patient like this, like a consolidation? DR LEWIS: Yes. Yes. DR LOVE: I mean, for example, when he stopped the therapy, I mean, was that something you thought about, radiate him? DR LEWIS: Yes. At the time, to be honest, I can’t recall why at the initial — we didn’t radiate him afterwards. But radiation is an option for these patients in general. Comparison of the efficacy and side-effect profiles of vismodegib and sonidegib DR LOVE: Can you talk a little bit — you mentioned vismodegib — about the clinical research data we have with the 2 agents? Whether there are other agents being looked at? Is there any difference in either efficacy or tolerability? DR LEWIS: Both of the trials, for the most part, were single-agent trials. There wasn’t a control group for the most. The vismodegib study, which came out first, was basically a straight Phase II trial of treating patients with vismodegib and looking at their overall response. The sonidegib trial was a little bit different. It was a randomized trial between 2 doses of drug, 200 mg versus 800 mg, which was more the minimally effective dose versus the maximum tolerated dose. It was felt that the 800 mg was probably going to be more effective, but it turned out that efficacy was equal between the two and toxicity was increased in the 800 mg. So the lower dose was actually the dose that was approved. The side-effect profile seems to be class effect, hedgehog inhibitor-induced side effects, which is the big one being taste loss, hair loss and muscle cramping. And then fatigue and weight loss along those lines. DR LOVE: When you choose between the two approved hedgehog inhibitors, is it kind of like an aromatase thing in breast cancer, that it’s a coin flip? Is there any way to differentiate them? DR LEWIS: I don’t think there is, because the trials weren’t head-to-head trials, so I can’t say that one is better than the other in terms of efficacy and tolerability. So it makes it difficult. And I think either choice is a correct choice. Case: A 67-year-old man with metastatic Merkel cell carcinoma experiences a complete response to the anti-PD-L1 antibody avelumab DR LOVE: So let’s talk a little bit about Merkel cell carcinoma. It’s amazing some of the tumors that you’re seeing have approvals in, nowadays, some really unusual ones. Can you talk a little bit about your 67-year-old man? DR LEWIS: Yes. So this is a patient who presented to us with metastatic Merkel cell cancer. He’s 67, and he did have a primary Merkel cell carcinoma on the buttock in July of 2015, and he had that excised. And he had a sentinel lymph node biopsy done at the same time as the excision, and that sentinel lymph node was positive — 1 of 3 lymph nodes on that were positive. And he followed that with radiation therapy to the lymph node basin. And then he was followed expectantly. And in April of 2016, so about a year after that initial diagnosis, he had new FDG-avid mediastinal and hilar lymph nodes, one of which was biopsied and pathologically consistent with metastatic Merkel cell carcinoma. So he was seen by us and we started him on avelumab, which is a PD-L1 antibody, and he had a complete response to therapy. The mediastinal and hilar lymph nodes are now normal in size. And he continues on that therapy in a radiographic complete response and doing well from a toxicity standpoint. Biology and clinical presentation of Merkel cell carcinoma DR LOVE: Can we kind of backtrack a little bit and talk about the pathogenesis and clinical presentation and management of Merkel cell? DR LEWIS: It tends to be a disease of older patients. It is increasing in frequency. And there is good data now that it’s likely, by and large, virus driven. There’s a Merkel cell polyomavirus has been identified. And that’s probably pathogenic in most of the cases. And there is an unfortunate very high propensity to develop metastatic diseases. DR LOVE: Where do you see these on the body? And how do patients present? DR LEWIS: They’re often in chronically sun-exposed areas, so head and neck. We’ve seen a number on the arms. This guy happened to be on the buttock. But they can present anywhere. So you just have be — these growing flesh-colored nodules have to be taken seriously. JAVELIN Merkel 200 trial: Efficacy of avelumab in patients with metastatic Merkel cell carcinoma and disease progression on chemotherapy DR LOVE: So what’s this man’s current situation? DR LEWIS: This man is in complete response, radiographically. And he stays on the avelumab infusions. DR LOVE: How long has it been since he started treatment? DR LEWIS: So he was diagnosed with metastatic disease in April of 2016, so we’re just about 2 years now. DR LOVE: And overall, what do you see with checkpoint inhibitors in terms of response, complete response? DR LEWIS: He’s a first-line patient, and most of the data that’s been published on avelumab has been in the second line, so chemotherapy-refractory patients. There is a little bit of data on the first-line population of avelumab. And the response rates in that situation, with an article published in JAMA, was about 60%. So a very high response rate to the immunotherapy. And there was just an update presented at ASCO 2018 in the second-line setting looking at overall survival. And there’s a plateau of that survival curve of around 40% for these patients. And looking historically at second-line therapy for Merkel cell carcinoma, these patients died in short order. So this was a game-changer for these patients and certainly the first-line option, in my opinion. Perspective on the duration of immune checkpoint inhibitor therapy DR LOVE: So again, it’s natural people want to present cases of patients doing well, really is great when we hear about these. What about the issue of stopping therapy? This comes up every tumor that checkpoint inhibitors are utilized. Any general thoughts and more specific thoughts about this man? DR LEWIS: The data for Merkel cell is not out there for stopping therapy, so I’m kind of apprehensive to do that in somebody who has such a deadly disease and such a great response. So right now there isn’t any immediate plans to stop. And it’s not clear, if it’s tumor dependent, whether we can stop these immunotherapies. The data for stopping immunotherapies is emerging in melanoma. And I think most cutaneous oncologists are starting to feel more and more comfortable with stopping immunotherapies for these patients who have great response, but whether that applies to other tumor types, other cutaneous malignancies, we don’t know yet. DR LOVE: You hear people in lung cancer talking about stopping therapy also. I guess it really gets into the heart of what you visualize as happening in these patients and whether you think there’s the potential for cure. DR LEWIS: Yes. And I think in melanoma that that’s certainly the case. If we go back to immunotherapy with melanoma with high-dose interleukin-2, those patients were cured. It was 5% or so of patients, but they received immunotherapy. They had a complete response to that therapy. And we were comfortable that we could stop the treatment and they would have long-term survival. So I think patients who get these great responses to immunotherapy with metastatic melanoma, a lot of those probably are cured and the good portion of them can stop immunotherapy at some point. Ongoing investigation of immune checkpoint inhibitors for Merkel cell carcinoma in the (neo)adjuvant setting DR LOVE: Any ongoing trials in Merkel cell right now that you want to comment on, new strategies to improve the outcomes? DR LEWIS: One of the things that is of interest is, now that we see immunotherapies effective in this disease, looking at these checkpoint inhibitors adjuvantly. Because these patients who have lymph node involvement or have very deep risky primary tumors are at such a high risk of developing metastasis that they’re good candidates for adjuvant therapy. And up until now, there really hasn’t been data that supports the use of adjuvant treatment for these high-risk patients. There was the data in chemotherapy but probably doesn’t affect outcome. So there’s some exciting trials going on looking at this adjuvantly. DR LOVE: So there are a couple of papers presented at ASCO on Merkel cell I was curious about, one looking at nivolumab as neoadjuvant therapy and another looking at pembrolizumab in patients with advanced Merkel cell. Any comments on those? DR LEWIS: So the nivolumab neoadjuvant data was interesting. I think, if I recall right, they just gave a couple of doses of therapy and looked at response. And they saw very early responses, which is what you do tend to see in these nonmelanoma skin cancers. They can respond quickly, and then they were resected. DR LOVE: You see people with, like, locally advanced disease with Merkel cell where surgery’s a problem? DR LEWIS: Occasionally. There can be bulky nodal disease. It’s not as much of a problem, but it certainly can happen. There can also be intransient metastasis and so, therefore, they don’t have quite systemic disease, but they have unresectable disease. And then a small study with pembrolizumab that showed very good response rates and durability. So again, I don’t think this is drug specific. I think it’s immunotherapy sensitive. So avelumab being a PD-L1 antibody, pembrolizumab being a PD-1 antibody, is one better than the other? I can’t say. I just think that these tumors respond to immunotherapy. DR LOVE: It’s interesting. I see that the response rate in the neoadjuvant trial was, it was almost 50%. We’re starting to see more and more neoadjuvant trials. I saw one in bladder that was, like, 40% path CR. I mean, amazing. Any other comments about trials in the neoadjuvant setting, not just in terms of trying to improve the patient clinically but maybe to learn more about this therapy in pre- and postbiopsies, predictors, et cetera? Is that something there’s a lot of interest in in cutaneous oncology? DR LEWIS: Oh, yes. And I think that’s where the excitement of the neoadjuvant trials, so to speak, lies, which is, can we identify markers that help us predict which patients are most likely to benefit? If somebody has resectable lymph node in Merkel cell carcinoma, you don't necessarily need to treat them neoadjuvantly, but the advantage of doing that on study is, you can get some valuable information. DR LOVE: It might be interesting, too, to look at what happens in a checkpoint inhibitor, with or without — you were mentioning the LAG3 inhibitor — or some other immunologic alteration to see, again, in the neoadjuvant setting, what’s going on. DR LEWIS: Yes. And I think that’s where we’ll go with combination checkpoint therapies and improving response rates beyond what we have. So it’s an exciting time. Case: A 30-year-old woman with Stage IIIB melanoma and a BRAF tumor mutation receives adjuvant pembrolizumab DR LOVE: So let’s spend the rest of our time talking about melanoma. And I want to begin by talking about the issue of adjuvant therapy. And to get into that, I’d like to hear about your 30-year-old lady. DR LEWIS: So this is a 30-year-old female. She came into a dermatologist, and she had a bleeding pigmented lesion on her lower abdomen. And a biopsy showed that it was a fairly deep melanoma, 3.2 millimeters. Had some poor prognostic features. It was ulcerated. And she had resection and sentinel lymph node biopsy done. And 2 lymph nodes were positive microscopically for metastatic melanoma. So I think we staged her as Stage IIIB. She had no other evidence of metastatic disease on imaging. We did molecular analysis of her tumor, and she did have a BRAF mutation, but she ultimately decided to be treated adjuvantly on a clinical trial of ipilimumab versus pembrolizumab, and she was randomized to receive pembrolizumab. She completed a year of adjuvant therapy. And she did well, minimal to no toxicity. And she remains without disease in follow-up. Efficacy of immune checkpoint inhibitors and BRAF/MEK inhibitor combinations as adjuvant therapy for Stage III/IV melanoma DR LOVE: So I guess it’s been almost a year now since the ESMO meeting in the fall of 2017, where we really saw some spectacular adjuvant data. Can you go through where we are today in terms of adjuvant therapy of melanoma, particularly checkpoint inhibitors, but also, obviously, BRAF-positive disease and BRAF inhibitors and how you go about approaching patients like this lady? DR LEWIS: Yes. So we went from a situation where we had really ineffective adjuvant therapies for melanoma to now having 2 approved, what appear to be, highly effective adjuvant options — immunotherapy and targeted therapy. So the first therapy that was approved was the CTLA-4 antibody ipilimumab. And in a randomized study against placebo, high-dose ipilimumab at 10 milligrams per kilogram was superior to placebo. And so it was approved a few years ago. But what was presented at ESMO in 2017 was a randomized study of nivolumab versus ipilimumab. And the results show that nivolumab resulted in superior outcome in terms of relapse-free survival. It’s also much better tolerated. So it’s a more effective therapy — appears to be a more effective therapy, as well as a better-tolerated therapy. And this trial was done in patients with Stage III and resected Stage IV melanoma. And it’s now approved as an adjuvant treatment option. And what was also presented at ESMO 2017 was looking at adjuvant BRAF inhibition. And there was a trial presented that was Stage III patients only harboring BRAF mutations, randomized between dabrafenib/trametinib combination, a BRAF inhibitor plus a MEK inhibitor compared to placebo. And patients who received dabrafenib/trametinib did better in terms of relapse-free survival and even overall survival compared to the placebo group. There was also a study that looked at single-agent BRAF inhibitor vemurafenib compared to placebo. And the results of that study were not statistically positive. And that study was driven more by looking at the highest risk groups, the Stage IIIc patients. And there was a significant benefit in relapse-free survival at 1 year for those who received single-agent BRAF inhibitor — and I should say that the therapies were 1 year of targeted therapy. Once they stopped vemurafenib, within the next 12 months those curves came together. So it appears that with single-agent vemurafenib 1 year may not be enough of therapy. So based on the results of those data, nivolumab is now approved adjuvantly, as well as dabrafenib and trametinib. DR LOVE: What about nivolumab/ipilimumab in the adjuvant setting? DR LEWIS: There are some studies that are being looked at with that. We don’t have any of the data yet. And that’s going to be interesting to see. Certainly with standard-dose ipilimumab/nivolumab, you do worry about toxicity, as that’s a very toxic regimen. But if you can attenuate that dose and attenuate the toxicities, is it superior to PD-1 by itself? And the studies will see. DR LOVE: Any subsequent data that have come out in the adjuvant setting, either from those trials or other trials, particularly in terms of trying to identify which patients benefit from which therapies more? DR LEWIS: No, not a whole lot of additional data has come out. But I think the key to, particularly the targeted therapy data, the dabrafenib and trametinib, for me, I think we need longer follow-up of that data to ensure that a year of therapy is enough, particularly in the highest-risk group patients, those Stage IIIC patients and probably Stage IIIB patients as well. The follow-up, after the ESMO presentation, was published in The New England Journal of Medicine, and the follow-up was decent. I think it was about 3 years, 2.8 years. But I am interested to see if those curves do come together with longer follow-up. So is the targeted therapy just delaying relapse, or is it really preventing relapse in those higher-risk patients? That’s the one concern I have, and what I look forward to is seeing that longer follow-up data. Choosing between a BRAK/MEK inhibitor combination and immune checkpoint blockade as adjuvant therapy for melanoma with a BRAF tumor mutation DR LOVE: When you compare indirectly these two strategies, which nobody is supposed to do but everybody has to do, what do you see? From your point of view, is there any way to separate them out, at least in terms of efficacy? DR LEWIS: Yes. As you try to compare across those — and I don’t think we’ll ever have data looking at BRAF/MEK inhibition versus PD-1 inhibition in the adjuvant setting, so how do you compare the two and how do you choose? My bias tends to be, honestly, I favor giving immunotherapy to those patients, because I do think that there’s more data out there that shows that immunotherapy responses can be durable. And, therefore, if a patient is otherwise a candidate for immunotherapy, I would tend to lean that way over targeted therapy. Again, getting to that point of the higher-risk patients, is a year of treatment enough of targeted therapy? Are we just going to delay that recurrence? But certainly, there are situations where patients aren’t good candidates for immunotherapy. And, therefore, it’s nice to have the targeted agents as a backup in my practice. DR LOVE: Yes, that’s a really good point, particularly in terms of prior autoimmune disease, transplants, et cetera. How often do you, in general, run into that situation, either in the adjuvant or metastatic setting? DR LEWIS: It’s fortunately the exception rather than the rule DR LOVE: Maybe that’s because the patients are younger. DR LEWIS: Yes. But it still comes up. DR LOVE: That’s an interesting point, though. When you bring up this issue of durability, is that more from the experience of metastatic disease, or is there more follow-up in the adjuvant setting that makes you say that? DR LEWIS: No, that’s a good point. It’s really extrapolating from the metastatic disease. We still need longer follow-up of the immunotherapy adjuvant trials. DR LOVE: And I’m kind of curious, have you used BRAF/MEK adjuvant therapy at all, either in the clinical trial setting or outside a trial setting? DR LEWIS: Yes. We did a lot of the clinical trials with the targeted therapies, so we put a number of patients on those adjuvant trials. But once they become approved, we’ve occasionally used BRAF/MEK. In Colorado, we have a large catchment area, so some patients come from quite a distance away, and coming in for frequent infusions is not conducive to their lifestyle. So if they have a BRAF mutation, we’ll talk about the targeted therapy and they’ll often choose that as an option. So you have to take it on a case-by-case basis. DR LOVE: I’m kind of curious how the BRAF strategy works out in the adjuvant setting. I mean, some of these patients are cured. It kind of seems more likely that they’re going to have side effects or quality-of-life issues than with checkpoint inhibitors. What’s it like trying to get somebody through a year of adjuvant therapy of a BRAF/MEK? DR LEWIS: Our experience has been, they do okay. I think these patients are highly motivated, knowing the disease that they have. And we’ve learned, over the years, how to manage toxicities of the targeted agents, with dose delays and dose reductions. And we can get patients through their therapy, by and large. Case: A 65-year-old man with a long-standing nevus on his back is diagnosed with metastatic melanoma with a BRAF V600E mutation and receives dabrafenib/trametinib DR LOVE: Maybe we can get into that more by talking about some cases in the metastatic setting. And we were talking about BRAF/MEK combinations. Let’s hear about your 65-year-old man. DR LEWIS: So this is a 65-year-old gentleman who really avoided medical care. He hadn’t seen a doctor in probably 30 years. And he had a longstanding mole on his back that about a year prior to presentation started to change. It grew significantly. It started to bleed. It was bleeding through his shirts. And he presented to an outside physician and had a biopsy that pathologically just came back as melanoma. And he subsequently came to see us, and he had about a 5-cm raised bleeding mass on the midback. And he clearly, on exam, had in-transit disease that was palpable at that time. So we imaged him with a CT scan of the chest, abdomen and pelvis and an MRI of the brain, and he had widespread metastatic disease with liver involvement, spleen involvement and multiple brain metastases. And the brain metastases were borderline for stereotactic radiation per our neurosurgery group. So we had initially planned to treat him with combination ipilimumab/nivolumab immunotherapy, but his molecular testing subsequently came back and he had BRAF V600 mutation. So we started him on dabrafenib and trametinib therapy. DR LOVE: Before you go on, can you just take a breath and talk a little bit about how you thought that decision through and why it was in this patient you decided to go with the BRAF/MEK rather than immunotherapy? DR LEWIS: We don’t know the right or wrong answer because we can’t select the patients who are going to respond to the immunotherapy. But we do know that the combination immunotherapy with ipilimumab and nivolumab has very high response rates in general. And there’s recent data that showed that those response rates systemically seem to equal response rates in the brain, and that was recently published in The New England Journal of Medicine. So giving that combination immunotherapy, although aggressive, because there’s a high chance of autoimmune toxicity, response rates are quite high. The fact that he had a BRAF mutation gave us an option. In this case, there was some social issues involved, which is why he hadn’t seen a doctor for so long. So you get a little bit worried about the immunotherapy toxicity. And we also have data that BRAF/MEK gets into the brain and there are responses in the brain. So ultimately, we decided in him, as we needed a rapid response, and for immunotherapy toxicity concerns, to do the targeted approach. First-line therapeutic options for patients with metastatic melanoma and BRAF tumor mutations DR LOVE: What’s your general strategy with first-line therapy of metastatic disease with a patient with a BRAF mutation, BRAF or immune? DR LEWIS: We’ve really, over the last few years, have changed. And my general strategy is, if they’re a candidate for immune therapy, generally we will give immune therapy first. And it’s to the point now where oftentimes if somebody presents, we won’t even wait for the mutational analysis. We’ll go straight to immunotherapy, whether it’s ipilimumab/nivolumab combination or whether it’s PD-1 by itself. DR LOVE: This discussion has gone back and forth now for a while, and I think at one point there was the perception that the chance of long-term benefit, even cure, was going to be greater with immunotherapy. And then I started to hear people, and I started to see data suggesting that even with BRAF therapy you see long-term survival and maybe cure. At this point, any kind of gut feeling or scientific basis to estimate what the likelihood is of that in both of these strategies? DR LEWIS: Yes. And it’s really interesting, and things have changed in the last couple of years. So when these targeted therapies initially came out, it was my thinking, and a lot of people’s thinking as well, I believe, that somebody who comes in with bulky, symptomatic disease that has a BRAF mutation, they should have targeted therapy. Shrink that disease down. They’re the best candidates for that. Some data has come out that’s really changed that thinking, one of which is that patients with heavy burden of disease, elevated LDH, can respond quite nicely to combination immunotherapy. So we don’t necessarily need that targeted therapy up front. The other issue is resistance to BRAF inhibitors. And the data is fairly convincing that patients with a heavy burden of disease are the patients who are most likely to develop resistance quickly. And as we look at the long-term follow-up of patients treated on BRAF/MEK combinations, the ones that do the best long term are actually the patients with low burden of disease — normal LDH, few sites of metastasis, less than 3. They’re the patients who are less likely to develop resistance and more likely to get long-term benefit out of it. So it's really a reversal of where we were. Results of the Phase III COLUMBUS trial: Efficacy and tolerability of encorafenib/binimetinib versus vemurafenib or encorafenib for unresectable or metastatic melanoma with a BRAF V600 mutation DR LOVE: So we now have a third BRAF/MEK combination now approved, encorafenib and binimetinib. Can you talk a little bit about what we know about that combination and how the data and your clinical experience compare to the other two? DR LEWIS: Yes, the trial that led to the approval, the so-called COLMBUS trial of encorafenib and binimetinib was, by and large, a randomized trial of the combination versus single-agent BRAF inhibitor, and the BRAF inhibitor they chose was vemurafenib. So the data is largely a study of BRAF/MEK, encorafenib/binimetinib versus vemurafenib. The progression-free survival and the overall survival is quite encouraging with that combination. But we know that single-agent BRAF inhibitor is not the standard of care. So you’re really not comparing it to what we would otherwise treat patients with off protocol. So that makes it a little bit more difficult to interpret those data. I think we’re going to have a hard time comparing them to the other combinations, dabrafenib/trametinib, vemurafenib, cobimetinib, but it certainly looks like it’s just as effective, if not slightly more effective. I think where it’s really interesting is maybe comparing across studies the toxicity profile. Because the previous approved combinations did have issues with toxicity, vemurafenib with photosensitivity, dabrafenib with fevers. And the toxicity profile of the new combination, encorafenib and binimetinib, seems to be favorable compared to that. So I think that’s important to take into consideration. DR LOVE: Is it your clinical impression, have you used this combination, that it is less toxic? The thing I think that I hear the most about is fevers. Do you have any experience with this combination? DR LEWIS: We are just getting experience with it. We didn’t participate in the trials that led to the approval, so we don’t have a lot of experience with that yet. But we have a handful of patients now on that combination, and they seem to be doing pretty well. But that’s anecdotal, and it’s still early in terms of our patient numbers. Management of dabrafenib/trametinib-associated side effects DR LOVE: Can you talk a little bit about your experience with fevers? Is it the most common issue that you have to deal with dabrafenib/trametinib? When do you see it? And how do you manage it? DR LEWIS: It’s probably not the most common toxicity, but it’s the toxicity that probably most commonly leads to stopping the medication or dose adjusting the medication. It could be quite problematic for patients. They just feel really bad when those fevers come in. Predicting who’s going to get the fevers? I don’t have the ability to do that. We just have to watch out, patient by patient. I tend to manage it with dose delay. So if patients develop fever, just hold the medication until they start feeling better. I usually try to rechallenge them at full dose and see if we can get back on full dose without recurrence of the toxicity. And that actually does work in some patients. They take a break. They restart. They have some sort of tolerability, and they’re able to tolerate full dose. But a lot of patients have to end up being dose reduced. And you do a dose reduction, and they can usually get through that and continue on the medication. Occasionally I’ve used low-dose steroids in patients as a preventative measure, but most of the patients get by with dose delay and dose reduction. DR LOVE: What are the other common problems that you run into? And how do you manage them in patients on dabrafenib/trametinib? DR LEWIS: Fatigue is an issue. They can get myalgias and arthralgias. And it’s really just symptomatic care. But if it becomes severe enough, it’s the same as the fevers, which is dose delays and dose reductions. DR LOVE: And in terms of what’s been seen in terms of the need for — I know it’s indirect, but in the COLUMBUS trial, the use of encorafenib/binimetinib. Did is seem like they had less need for dose reductions or holding treatment? DR LEWIS: The drug seemed more tolerable overall, yes. Certainly the fever’s much less and the photosensitivity is much less, which are the 2 biggest problems with the vemurafenib/cobimetinib and dabrafenib/trametinib combinations. DR LOVE: How often do you see patients just cruising through these combinations, no problems whatsoever? Or do they usually have something? DR LEWIS: They usually have something. It’s very uncommon, in my experience, that they have no toxicity. But most patients do not require dose reduction or dose delays. The toxicities are generally mild that they can tolerate them, get through and maintain on the therapy. But almost everyone has some kind of toxicity. DR LOVE: And what are some of the new research strategies that are being looked at in BRAF-positive metastatic disease? One obvious thought is combining immunotherapy and BRAF therapy. DR LEWIS: Yes. I think that’s really where the interest lies currently, is to say, can we combine these targeted agents with immune checkpoint agents? And it’s a very interesting area of research right now. And it turns out that, actually, the targeted agents, not only is there antitumor effects, but there’s probably proimmunologic effects and that we see that the tumor microenvironment becomes more inflamed microenvironment, so to speak. And so really, the targeted agents may set up patients to respond better to immunotherapy and really that’s where the interest lies at that point. And the trials are ongoing. And certainly much anticipated, the results. Case: A 53-year-old man receives first-line ipilimumab/nivolumab for metastatic melanoma DR LOVE: So let’s finish out hearing about your 53-year-old man. DR LEWIS: Okay. So I had a 53-year-old gentleman who has a history of intermediate-thickness primary melanoma on the shoulder in 2012 that was treated with wide excision, and he was found to have 2 positive sentinel lymph nodes at that time. And we treated him with adjuvant biochemotherapy. That’s where we were with adjuvant treatment back in 2012. He did well until mid-2015, when he presented with neurologic symptoms and he was found to have a solitary brain mass that was resected and was consistent with metastatic melanoma. Body imaging at that time showed a solitary pulmonary nodule that was new from his previous scans and consistent with metastatic disease. And we started him on combination immunotherapy with ipilimumab/nivolumab and he had a complete resolution of his lung lesion and then he, after the 4 cycles of ipilimumab/nivolumab combination, but then we continued him on single-agent nivolumab and he did well, with minimal toxicity. And he subsequently stopped infusions in October of 2017, and we’ve been following him now for a year off therapy without recurrence. DR LOVE: What are some of the teaching points that you think came up from this case? DR LEWIS: This case I think is interesting because it’s a patient where we’ve stopped immunotherapy and he’s maintaining that response without active therapy. And that’s really what we go for in these situations is, we want to be able to stimulate the immune response, have it deal with the melanoma effectively and then these patients have that chance of long-term survival without being active cancer patients coming in frequently for infusions. Also, this demonstrates the recent New England Journal paper that looked at the ipilimumab/nivolumab combination being active with brain metastasis. So this is a patient where you really worry, is he going maintain that immunologic response? Is he going to relapse, and is he going to relapse in the brain? It does make you worry. But this patient seems to be maintaining that. Perspective on combination therapy versus monotherapy with immune checkpoint inhibitors for metastatic melanoma DR LOVE: Can you talk a little bit in general about how you approach the decision between using single-agent PD-1 and combination ipi/nivo? DR LEWIS: In general, what I do is, it’s really patient based. So it’s individual patient based. I think that if somebody has metastatic melanoma and they’re young, healthy, not even necessarily have to be young, so to speak, just have to be healthy and able to tolerate combination immunotherapy, that’s my preference at this point, is the ipilimumab/nivolumab combination. If there is some comorbidities that make me a little bit worried about their ability to tolerate that, then I would favor single-agent PD-1. DR LOVE: Are there any situations, let’s say, where the patient’s kind of on the borderline in terms of age or performance status where you’ll start PD-1, see how the patient does and maybe start nivo and add in ipi? DR LEWIS: Yes, that gets to the issue of, okay, we’re starting somebody on single-agent PD-1 and they’re not responding. What’s our second-line therapy? And we’ll often go to ipilimumab/nivolumab combination. But there’s not a whole lot of data out there that tells us what the expected response rates and benefits overall would be in that situation. But we do it. DR LOVE: Do you obtain or look at or consider PD-L1 levels or any other tumor mutation burden? Any other assay in making the decision? DR LEWIS: The question about PD-L1 status is a good question. I do not check PD-L1 status. I do not think it helps in my decision-making process. There was a paper published in New England Journal of Medicine that was a 3-arm trial of ipilimumab/nivolumab, nivolumab by itself and ipilimumab by itself. I think it’s important when you look at the data from that trial that the study was not designed to look at ipilimumab versus nivolumab. It was designed to look at either of those nivolumab-containing regimens compared to ipilimumab. So you can’t make statistical considerations of the combination arm versus single-agent nivolumab. But there’s been subsequent looks at it that shows if PD-L1 expression is low, then the combination may be better, but I just don’t know that that data is truly ironclad. So I don’t use that to make my decisions, personally. DR LOVE: Anything that we haven't talked about today, whether melanoma or the other tumors that we talked about, that you want to bring up that you think would be of interest to a general medical oncologist? DR LEWIS: I think we’ve hit most of them. I mean, the emergent data in melanoma with immunotherapy continues to be exciting, the combinations, immunotherapy approach outside of ipilimumab/nivolumab with LAG3 and TIM3 or other antibodies. And then what’s very important was, I think, to look at advanced cutaneous squamous cell carcinoma as an immunosensitive tumor. These are emerging data, but I think they’re very exciting and will change the way we think about treating those patients. DR LOVE: Subsequent to this interview, on September 28, 2018, the FDA approved cemiplimab for patients with locally advanced or metastatic cutaneous squamous cell carcinoma who are not candidates for curative surgery or radiation therapy. In addition, on December 19, 2018, the FDA approved pembrolizumab for patients with recurrent locally advanced or metastatic Merkel cell carcinoma. Choosing between nivolumab and dabrafenib/trametinib as adjuvant therapy for melanoma with a BRAF tumor mutation DR SZNOL: We used interferon for years. And interferon was really replaced by ipilimumab some years ago. The problem with ipilimumab is that the initial trials used 10 mg per kilogram, which is a really toxic dose. There were data presented that showed that, at least for relapse-free survival, 3 mg/kg of ipi was as good as 10 mg/kg. And so I think at that point people switched over to 3 mg/kg, but that didn’t last very long, because it was not too long ago, not soon after we started using ipilimumab we learned that nivolumab actually was a much better drug. So in the adjuvant trial, nivolumab versus ipilimumab — and in this trial they did use the 10 mg/kg dose of ipilimumab. Clearly, nivolumab was better for progression-free survival. There’s no survival data yet in that Phase III trial. And the other trial that changed the way we treat adjuvant patients involved the dabrafenib/trametinib, which also showed a marked improvement in progression-free survival, in that case over control, over no treatment. DR LOVE: And kind of the global thought I’ve been hearing from investigators since that time was that it’s kind of hard indirectly to separate out the benefit that you see with adjuvant BRAF as opposed to adjuvant nivolumab in BRAF patients. Is that still your take? DR SZNOL: Yes. First of all, a lot of us didn’t really think that the BRAF/MEK inhibitors would have such a profound effect in the adjuvant setting. We all knew the curves would separate for progression-free survival. We thought that at the end of the year, when you finish treatment, those curves would come right back together. That didn’t happen. Those curves stayed apart. And they’re still apart. And, in fact, there’s an early trend toward an improvement in survival in that study. Although the problem with interpreting that survival data is that if you really look very carefully, at the time of relapse not all patients got dabrafenib/trametinib. In fact, they probably didn’t get the 2 drugs together, and that’s better therapy. So had patients been treated I think optimally in the metastatic setting, we might not be seeing any difference in overall survival. We don’t know. So if you have a patient who has a BRAF mutation and you’re to decide whether to give a BRAF/MEK inhibitor versus nivolumab, I’m not sure that we know which approach is going to be better. I favor giving adjuvant nivolumab. I think in the long run that’s going to end up being better, but I don’t know that we have data to prove that. DR LOVE: That’s interesting. When you talked about that, I was flashing on the adjuvant tamoxifen studies in breast cancer. Tamoxifen was thought to be a cytostatic agent. It wasn’t going to work, and then they saw this survival benefit. There was even a trial — I don’t know if you know this — that compared tamoxifen — literally, tamoxifen adjuvant to tamoxifen in first relapse, and they still saw a survival benefit. So the other thing is, what about ipi/nivo in the adjuvant setting? DR SZNOL: That trial was done and completed. It used a lower dose of ipi in a less frequent schedule. It may be better than nivolumab in the adjuvant setting, but we don’t have the data. And until it’s analyzed, I don’t think it should be used as standard of care in the adjuvant setting. There are some patients where I consider using adjuvant ipi/nivo, and that would be, for example, in patients with mucosal melanomas. For those patients, I’m hesitant to use nivolumab alone, because nivolumab in the advanced disease setting has a very low response rate in the mucosal melanoma patients. DR LOVE: Interesting. Efficacy, tolerability and quality of life with adjuvant dabrafenib/trametinib and adjuvant nivolumab DR LOVE: So you mentioned that you tend to favor nivolumab in the BRAF patients. In your mind, as you sit down and discuss this patient, I’m assuming this comes down to complications, quality of life, et cetera, since it’s hard to separate out the benefit. In your mind, how do you compare these two courses? Or are you thinking, even though I don’t see it in the data, maybe the immunotherapy is going to be better efficacy wise? DR SZNOL: So in the first year, if you look at progression-free survival, the BRAF/MEK agents look better and the studies use slightly different patient populations. So I think in the adjuvant nivolumab trial, I think it started with IIIB, I think the BRAF/MEK trial started with IIIA patients. So you have to do a little back of the envelope calculation to figure out and to compare apples to apples. But I still think BRAF/MEK will probably, in terms of progression-free survival, be better in the first year. But when you start looking now at the 2-year data, the numbers start looking, in terms of relapse-free survival, very similar to each other. So nivolumab can, in a small number of patients, can have very serious toxicities. But it’s IV. It’s once a month. It’s for a year. And based on everything we know about advanced disease, nivo gives you durable responses. BRAF/MEK rarely does. And so the more high risk the patient is, the more likely I am to use nivolumab. Interestingly enough, if I remember the BRAF/MEK study correctly, a lot of the effect, or at least the most impressive effect, was really in the IIIA. So the earlier the disease, the bigger the effect. The later the disease, the smaller the effect. And again, I tend to favor nivo in that group. But I give patients the choice, and there are some patients who want the BRAF/MEK inhibitors. DR LOVE: Just putting aside your thought that maybe it’s going to be a better outcome from a tumor point of view, how do you compare, in terms of quality of life and the risk of serious complications? A lot of people kind of do view these as equivalent efficacy wise, and that’s when they start getting into these discussions about what people go through. How do you see it? DR SZNOL: Let me address the efficacy question. I mean, in advanced disease, I don’t think there’s any question that nivolumab is better. If you looked at median survivals, the median survivals for nivolumab are better than for BRAF/MEK inhibitors if you’re comparing, I think, apples to apples, the same group of patients. And I think that probably will translate into the adjuvant setting. In terms of quality of life, dabrafenib/trametinib actually is not so benign. Patients develop fevers, and there’s a higher discontinuation rate than you might imagine in that group. Probably because it’s in the adjuvant setting and people don’t want to push patients as hard. Overall, nivolumab is really very well tolerated. There’s a small group, maybe 10% that can develop severe toxicities. And, by the way, if you develop a toxicity from nivolumab and you want to put those patients in on a BRAF/MEK inhibitor, you can do that. I’ve never done that, but it’s still possible to do that. DR LOVE: You mean to switch them? DR SZNOL: Yes, you can just switch over if you think that that’s necessary. I’ve not done that, though. If somebody is on nivo and they develop toxicity, I take them off. I don’t switch them over to a year of BRAF/MEK inhibitors. DR LOVE: So one of the things I’ve heard people say is that they think it’s safer to go the BRAF route. Yes you have to deal with fevers, whatever, but that the risk of permanent complications is not as great as with immune therapy. And I was like, hmm. I had never thought about it that way. What are they talking about? DR SZNOL: I think that’s true. We’ve had an occasional patient with either pembrolizumab or nivolumab that’s developed a really severe toxicity. I had a patient who developed myasthenia gravis after a single dose of pembrolizumab. And that patient was on a respirator for 3 months. DR LOVE: Wow! DR SZNOL: So you can get really bad, bad toxicities. Interestingly enough, I’ve given in a rare patient with BRAF MEK inhibitors, they’ve been admitted to the hospital and they’ve gotten severely ill. Although in all those cases, the toxicities have been completely reversible. There’s an occasional patient who gets anti-PD-1 that develops arthralgias. And those arthralgias don’t go away. I mean — and they take a very long time for them to go away, or they just don’t go away at all. And they become dependent on some form of treatment for those arthralgias, at least for a very long time. So you can develop long-term toxicities. Remember, there’s an occasional patient who gets anti-PD-1 that gets pneumonitis and fatal pneumonitis. So I think there are rare, irreversible or at least very long-term toxicities with the immune checkpoint inhibitors that you may not see with the BRAF/MEK inhibitors. DR LOVE: That’s interesting. And I want to talk more with you about toxicity later. But I have to ask you about this myasthenia case. In that case, if you look back historically, were there any experiences, symptoms, anything that would suggest maybe there was preclinical myasthenia? DR SZNOL: In this patient, there was nothing. Nothing. We were surprised. I guess retrospectively we would have gone back and said, “Oh, he had a little droopy eye.” But we didn’t make anything of it at the time that we offered in the treatment. And he denied any prior autoimmunity when we started him on the pembrolizumab. DR LOVE: When you think about this kind of a case with this neurologic toxicity, in your mind, is it likely that it’s magnifying a preexisting situation or really just de novo, creating an autoimmune complication? DR SZNOL: I think it could be both. I think in some people they may have some predisposition and you may be exacerbating something. It may be subclinical. In some, you may be generating — those T-cells have to be there, right? So it’s some subclinical autoimmunity that you bring out by giving the anti-PD-1. DR LOVE: Once I start going on toxicity, we could just go all day long. But I’ve got to ask you one other thing, and then we’re going to get back to the adjuvant thing. But I have to ask you this, because actually, we’re getting ready to go to the SITC meeting. And we’re talking about lung cancer and immunotherapy and small cell now has kind of come on the — it has been for a while, but now maybe first line looks like they’re going to be using checkpoint inhibitors. And I started thinking about people with small cell and paraneoplastic syndromes. I can’t find anything, almost anything in the literature, just, like, a little bit. Do you know anything about that? Whether or not checkpoint inhibitors make — of course, if the tumor gets better, you would expect the paraneoplastic syndrome. But do you think that it would make the paraneoplastic syndrome worse in and of itself? DR SZNOL: So I don’t treat small cell, but I did take care of a patient in the hospital where I thought that the immune checkpoint inhibitor, the anti-PD-1, had made some paraneoplastic syndrome worse. And it was a neurologic paraneoplastic syndrome. DR LOVE: Wow! DR SZNOL: It’s a pretty bad syndrome. And it wasn’t at a time when their disease was actually progressing, which you thought might have correlated. DR LOVE: Wow. DR SZNOL: So yes. It’s possible that in some people the paraneoplastic syndrome might become worse, assuming you’re not making their disease better. DR LOVE: Was this a myopathy, or what? DR SZNOL: It was some sort of encephalopathy. DR LOVE: Wow. Updated results from the Phase III CheckMate 238 trial: Adjuvant nivolumab versus ipilimumab after complete resection of Stage III/IV melanoma DR LOVE: So let’s get back to adjuvant. I wanted to ask you, there was a presentation by Jeff Weber at ASCO with more data from the adjuvant nivo. In particular, the BRAF patients in the nivo trial. Maybe we can talk about what he presented there. DR SZNOL: So the effect was seen regardless of BRAF status. And the risk reduction, maybe a little bit less in the BRAF-mutant patients. But still, similar effect whether you had a BRAF mutation or not in terms of improvement in progression-free survival at the 24-month time point. And if you look carefully at the hazard ratios, again, maybe a little bit less, 0.73 in the BRAF mutants versus 0.61 in the BRAF wild type, but still a reasonable effect. Toxicity manifestations with anti-PD-1 antibodies alone and in combination with a CTLA-4 inhibitor DR LOVE: So let’s move on and talk about some really interesting issues about toxicity. There’s a lot to talk about. Maybe you can kind of go through it. DR SZNOL: If you’re just talking about anti-PD-1 alone, it’s really very well tolerated. If you look at most of the studies, the Grade 3, 4 AEs — all Grade 3/4 AEs are in the 15% range, 15% to 20% range. If you look at select AEs, it’s less than 10%. When you start combining anti-PD-1 with any anti-CTLA-4, the rate of Grade 3/4 AEs goes up substantially, right? If you combine it with ipilimumab, depending on the dose of ipilimumab, that rate of Grade 3/4 AE could go from 15% with nivo alone — not much higher, really, with ipi alone — all the way up to anywhere from 30% to 60% Grade 3/4 AEs. DR LOVE: So as long as you bring that up about combined therapy, kind of the global kind of feeling I’ve had talking to people has been is that it’s more toxicity but the same types of toxicity. Is that the case, or is it a different toxicity profile qualitatively? DR SZNOL: It’s the same type of toxicity, but really compared to ipilimumab more so than nivolumab. So you really just see the ipilimumab toxicities. They’re actually very similar, regardless whether it’s ipilimumab or nivolumab. They’re very similar adverse effects. I think with anti-PD-1 you see more arthralgias. You see more pneumonitis. You don’t see that as much with anti-CTLA-4. But when you combine the two, what you’re really doing is, you’re taking the ipilimumab toxicities and you’re seeing higher incidence, more severity. So the most common toxicity manifestations involve the skin, the GI tract, the endocrine organs, liver and then a little bit further down, lung and kidney. Really, the most common thing you see is skin. You rarely see severe skin toxicities, but you can see skin toxicities. The ones that we worry most about are gastrointestinal toxicities. Those are the ones that you really need to act on right away. As far as symptoms are concerned, patients really present with fatigue, rash/pruritis, arthralgias, diarrhea/nausea, mucositis. Some patients with dry eye, dry mouth. You really have to have very close communication with the patients. You really have to have your nurses call them. You have to have them call your nurses. And you really have to treat them early for these adverse events, otherwise they can get really very ill. So those are the most common toxicities. But what we're seeing now, and we’re treating many more patients, are really rare adverse events that can be potentially life threatening. So we see in some patients inflammatory syndromes, for example. HLH — hemophagocytic lymphohistiocytosis — which is often fatal. Of the gastrointestinal toxicities, you see colitis. That’s common. But sometimes you can see a bowel perforation. Pancreatitis is rare, but we’ve seen it. Rare but potentially life-threatening complications associated with immune checkpoint inhibitors DR SZNOL: Cardiovascular toxicities are now being seen, and there are patients that have died from myocarditis and CHF. A rare Stevens-Johnson syndrome. If you develop Stevens-Johnson syndrome, it’s likely going to be fatal. It’s a bad toxicity. DR LOVE: That’s kind of interesting. Remind us about the pathophysiology of Stevens-Johnson syndrome and why it would get worse. I mean, I know it’s immune based, right? DR SZNOL: Yes. I mean, it’s just a very severe form of skin toxicity. It’s probably T cell mediated. I don’t think it’s antibody mediated. We’ve seen bullous pemphigoid, which I think is antibody mediated. DR LOVE: Right. DR SZNOL: But I think Stevens-Johnson is a more T-cell mediated toxicity. But when we see it, it’s rapidly progressive. It’s hard to turn off. And I’ve seen 1 case myself. I’m not even sure that it was the immune checkpoint inhibitor, because the patient was also on sulfamethoxazole at the same time. But had a very rapid progressive course and died in the ICU. So the other things that we’re seeing now, a little bit more commonly that used to be rare — in fact, we were among the first to report it — we’ve seen patients develop insulin-dependent diabetes mellitus, and that diabetes mellitus can be very brittle. You have to have an endocrinologist help you to manage that, because they’re very hard to manage. DR LOVE: So presumably it’s Type 1 diabetes? DR SZNOL: Yes, it is. It’s Type 1 diabetes. They’ll lose their pancreas. They’re insulin dependent. But they’re very brittle. You need somebody who really knows how to manage the toxicity to get these patients through. We’ve seen really debilitating rheumatologic disorders, arthralgias. They are often responsive to steroids. The myositis can be reversed relatively quickly, but some patients who develop myositis can also develop myocarditis. And you have to — if you see somebody — they come in weak or with myalgias — you get a CK. The CK is high, you should check for troponins. If they have myocarditis, they should be in the hospital. And I’ve seen myositis, myocarditis and myasthenia gravis in the same patient. DR LOVE: Wow! When you see myositis and myocarditis, what’s going on? I mean, it’s different types of muscles. DR SZNOL: Again, we don’t know the pathophysiology. But it’s, again, a potentially life-threatening complication. And then for the CNS toxicities, in addition to myasthenia, which is already bad enough, we’ve seen an optic neuritis that led to blindness in 1 patient. We’ve seen uveitis. The uveitis is most often reversible. And then this ascending or multifocal neuropathy. That I’ve seen more with ipilimumab alone, where you’re using 10 mg/kg — I can’t say I’ve seen it very often. But it’s a very scary adverse effect. People come in with leg weakness. We bring them into the hospital. We give the IVIG. We give them steroids. And that weakness doesn’t go away right away — it can take months before they recover function. DR LOVE: So this is an MS kind of syndrome? DR SZNOL: No, it’s different. It’s actually probably a demyelinating neuropathy. DR LOVE: Hmm. Wow. DR SZNOL: So it’s very important, I think, if you’re going to give these drugs — it’s one of the problems about giving any oncologic agent outside of a major center. That’s true for any agent. You have to have a multidisciplinary team that really knows how to manage these adverse events. The gastrointestinal toxicities and endocrine toxicities I can manage mostly by myself. When people start developing severe neurologic toxicities, insulin-dependent diabetes mellitus, you really need that expertise from other subspecialties in order to get these patients through the agents safely. The good news is that these toxicities that are to manage are very rare, and the vast majority of patients have reversible toxicity. DR LOVE: What about the radiation necrosis? I hadn’t heard that before. DR SZNOL: So we don’t know for sure that that’s tied into the immune checkpoint inhibitors, but many patients with melanoma, as you know, present with brain metastases. These days, we don’t necessarily give them Gamma Knife® radiation or stereotactic radiation, because they can respond to ipilimumab/nivolumab. In fact, that’s a subgroup of patients that probably should get the combination versus nivo alone. So if you’re not treating them with radiation, you don’t have to worry about radiation necrosis. But some patients you do treat with Gamma Knife radiation or stereotactic radiation, they either get the immune checkpoint inhibitors after or concurrently. And in those patients, even though the brain lesion goes away, anywhere from months to years later you can start seeing enhancement at the area that you treated before. And sometimes the edema around that can become enormous and can cause almost as many symptoms as the lesion itself. And our sense is that the immune checkpoint inhibitors may exacerbate the development of the radiation necrosis. DR LOVE: That’s really interesting. In general, how well do these CNS problems respond to corticosteroids? DR SZNOL: They do respond to corticosteroids fairly well, but in some patients you have to resect the area of radiation necrosis. There’s a method called LIT, which is a laser that you can actually insert into the lesion. And that can sometimes take care of the necrosis. And in some patients, we’ve had to give them bevacizumab in order to manage the edema around the radiation necrosis. DR LOVE: How much of an issue is having ER docs, primary care docs, tuned into these problems? And what are you doing at Yale about approaching it? How important do you think it is? DR SZNOL: I think it’s very important. And I can’t say we’re doing at Yale as much about this as we should, but certainly through the SITC we’ve created educational programs, including programs with community oncology centers, so that we can educate the emergency room physicians who, in some cases, see these toxicities before we do. So you’re right, it’s a very important thing to do, but the programs are being put out there through larger organizations. I think the last toxicity here that we should mention is, an occasional patient will develop hematologic toxicity — so develop severe, for example, hemolytic anemias or pancytopenias. I had one patient who came in with a white count of zero, low platelet counts, pancytopenic. She responded to steroids, but it was a very scary situation. DR LOVE: Wow. DR SZNOL: And now she’s back to normal. Her counts eventually recovered. This is another person that we gave steroids and IVIG. So you need to be on the lookout for this. And it’s a very good idea to have your hematologic consultant available when these bizarre toxicities occur. ITP has been seen, for example. Monitoring and management of rare complications with immune checkpoint inhibitors DR SZNOL: So the question is, how do you manage these toxicities? And it’s really important to understand that there are algorithms already developed that you can access online that help you to manage the toxicities. But primarily, the first thing is, you need to monitor these patients very closely. You need to establish a relationship with them. You need to be sure that they can come into the center if they have toxicities. If it’s somebody that lives far away and can’t get transportation into a major center, you have to worry about that person. We often encourage them to call frequently, and once we know that they’re having — for example, if they start developing diarrhea, we don’t depend on them specifically to call us. We actually proactively check on them to make sure that their symptoms aren’t worse. At the very first visit, I have to tell patients, “Don’t be stoic. Don’t be afraid to call us, even if you have 1 or 2 episodes of diarrhea. If you have something that’s unusual, call us. Let us manage it.” And if we have to, we bring them into an emergency room. Major centers often have oncology emergency centers that are better for patients. So I think that’s the first thing. The nursing staff is really critical in terms of managing these adverse effects. Supportive care is very important. People will develop, in some cases, nausea. For the diarrhea, sometimes we give loperamide, although we try to stay away from loperamide because we want to understand the severity of the symptoms. The single most important thing that you can do is start steroids when steroids are indicated and to start the right dose of steroids. For most of these toxicities, especially severe toxicities, we usually want to start at somewhere around 2 mg/kg of methylprednisolone. And for some toxicities, for example, if somebody comes in with uveitis and loss of vision, we put them in the hospital and we give them a gram of methylprednisolone daily for 3 days. So I would not be scared about going too high on the steroids when you need it. And also not to stop the steroids too quickly. Some people will give the high dose of steroids. People were averse right away, and they’ll want to stop the steroids right away. For Grade 3 or Grade 4 colitis, you don’t want to do that. You want to give them a week of high-dose steroids and then taper them over 35 to 40 days. We have a very low threshold to put somebody in the hospital. A lot of these can be managed as an outpatient, but I think if somebody comes in with Grade 3 or 4 diarrhea, I’d rather have them in the hospital, control it in a hospital. One of the things that fools us often is that they get better. We’ll give them 2 mg/kg of methylprednisolone. They’ll look better after 24 hours. We send them home. Twenty-four hours later, they’re on oral prednisone. They’re symptoms come back, sometimes even worse. So I may keep them an extra day, and if I see their symptoms come back, those are the people that I might actually give a secondary immunosuppressive, like infliximab, for example. And the other things are just common sense medicine. Most of the time you treat these things to drug, but you have to be sure that you rule out other causes. I’ve had people who’ve come in with diarrhea we thought was related to the immune checkpoint inhibitors, but it turned out to be C diff. DR LOVE: Wow. DR SZNOL: It was an infectious cause. And sometimes they can be present in the same patient. So that’s another twist. Some people have recurrent colitis. You think that you’re dealing with recurrent colitis, when actually you’re dealing with CMV colitis that occurred because of the immune suppression. And then the last thing is, really stay alert for severe symptoms. I mean, if somebody is weak — now when somebody calls and they’re weak, we get cortisol levels, we get TSH, we get free T4. I get a CK, because I’ve been fooled. And in somebody who has symptoms that look like myasthenia — droopy eyes, they can’t lift their neck — we’ll get neurology involved and get an antibody, antiacetylcholine receptor to make sure that they haven’t developed myasthenia. DR LOVE: And when you were talking about trying to find out early on that these people have problems, I was flashing on the fact that we had the pleasure of working with Ethan Basch a couple of times this year who did that ASCO plenary presentation on the automated preemptive approach where patients are — now I think they’re using phones and apps and all, putting their symptoms in on a regular basis. That gets triaged to a nurse. And it seems like that would really be ideal for these kinds of patients. DR SZNOL: It is. I personally prefer for them to call my nurse. I’m worried that there’s somebody on the other end of that phone reading the text. And a lot of people have Epic now, and Epic has a MyChart feature. And sometimes people will send in their symptoms on MyChart. I don’t know who’s looking at that and when they’re looking at it. If I’m on a plane, I can’t look at it. And so I tell people to call our staff and to keep calling until they get a call back from either our fellow or our nurse. So I haven’t gotten to the point yet where I trust those systems. DR LOVE: That’s a good point. Clinical experience with immune checkpoint inhibitor-associated colitis, pneumonitis and hepatitis DR LOVE: What are some of the main questions involved in AE management right now? DR SZNOL: It’s like anything else in medicine. It’s a differential diagnosis — is it disease? Or is it drug related? Or is it something else? I think that for a lot of us, we’ve been doing this so long that we skip the additional diagnostic test. So I don’t really do colonoscopies anymore for colitis. I get the stool cultures. I get fecal leukocytes, C diff. But I don’t even get a CT anymore, often. I’ll just go ahead and treat with steroids. Although, CTs are helpful because if you see stranding or inflammation around the bowel, it can tell you that things are little bit more serious than you might have imagined. If somebody comes in and has pneumonitis, do you do a bronchoscopy or not? Or do you go ahead and treat with steroids? Because sometimes these people have secondary infections. Another really difficult part of the clinic is when to start steroids. So it’s an easy decision when somebody comes in with severe symptoms. But somebody who calls in and says, “I have a little bit of diarrhea and a little bit of nausea,” do you start steroids then, or do you wait a day or two to start them? I tend to not start people with very mild symptoms right away. But those are the people that I have my nurses call every day or every other day to make sure that their symptoms aren’t escalating rapidly. And when you start steroids — again, if it’s severe, it’s easy. You start high-dose steroids. But there are people that you can manage with a methylprednisolone pack or with 30 mg or 40 mg of prednisone for a couple of weeks. And so knowing when to back off a little bit on the steroids I think is important. And you’re going to ask, why do you need to back off on the steroids? It turns out that giving steroids often doesn’t affect the efficacy of the treatment, but we’re always a little bit concerned. So if I can avoid giving high-dose steroids for prolonged periods, I’ll try and do so. But if somebody is really sick or if I think they’re going to get sick, I don't hold back. I give them what they need. The other problem in the clinic has to do with if somebody is already on steroid but they’re not responding to the steroids, do you go up on the steroids or do you add a second agent? The most common agents are mycophenolate, which we use for liver toxicity. We don’t use infliximab for liver toxicity, because that can cause liver toxicity in and of itself. For other adverse events outside of the liver, the second agent is usually infliximab or an anti-TNF. And then the question is whether to use the standard dose of infliximab, which is 5 mg/kg, or go right to 10 mg/kg. And then how often to give? Some people don’t respond to the first dose of infliximab, so 2 weeks later or a week later you might give, instead of 5, you might give the 10 mg/kg. I think when somebody starts to develop a toxicity, the other simple things are, how often do you monitor them in clinic? And one of the questions that often comes up is, if somebody develops a severe adverse event, can you rechallenge them after the toxicity resolves? And that’s a really difficult problem. I can tell you that in one patient of mine that had uveitis, terrible uveitis, 2 years later she had a recurrence. We rechallenged her with ipilimumab/nivolumab. And interestingly enough, she did not develop recurrent uveitis. DR LOVE: Wow. DR SZNOL: On the other hand, almost everybody that I’ve rechallenged who had colitis in the past had recurrent colitis. But I have rechallenged patients. If they’ve had very, very severe colitis, I tend not to do it. But it’s a clinical question, and if you think that rechallenging them at some point would give you benefit and you have nothing else for those patients, you might consider then giving them the ipilimumab/nivolumab in this case and then just saying, “I’ll just manage the toxicity when it comes up.” General principles for managing adverse events in patients receiving immune checkpoint inhibitors DR LOVE: So we’ve covered a lot of ground with regard to management of common and rare adverse events associated with checkpoint inhibitors. Maybe you could just distill it down to the general principles you use in managing these problems? DR SZNOL: So these are just general principles that we use. For any severe event, once we start them on high-dose steroids, we usually maintain the high-dose steroids until the event resolves. And then we taper slowly over about 30 to 40 days. When somebody has a really bad adverse event, it’s okay to think about using pulse high-dose methylprednisolone for a gram a day for 3 days and then bringing them back down. If the AEs persist or recur, sometimes you have to use a second agent, and that would be an anti-TNF or mycophenolate, as we’ve talked about before. Not all toxicities can be managed with steroids or mycophenolate, infliximab. Sometimes you have to use other things like IVIG or plasma exchange, especially for events like myasthenia, neurologic toxicities. And some people have used an anti-IL6 receptor for HLH, for hemophagocytic lymphohistiocytosis, which is a very bad toxicity. And then again, if you’re going to consider rechallenging patients, generally it’s okay if their immune adverse event was less than Grade 3 and if it resolved rapidly and if you think that if they get the recurrence of the toxicity that they won’t have major morbidity. But if it was a very severe event, if it didn’t resolve very rapidly or if it’s an organ that you might be afraid, for example, it was a bad neurologic toxicity before, you may not want to expose them again to these agents. Those are really the general principles. You can rechallenge if you’ve given ipilimumab before — ipilimumab/nivolumab — and they’ve had a bad toxicity. You can go back to nivolumab again. Often the toxicities are related to the ipilimumab component and not to the anti-PD-1 component. So you can go to anti-PD-1 when you’ve had a severe toxicity to the combination. But in some of those patients you may not ever want to give them ipilimumab again. I tell my colleagues to use consultants but also to beware of some of the recommendations, because the management of these toxicities are not identical to the same entity that would have occurred without the immune checkpoints. So just become familiar with it and question the consultant about whether — some consultants, for example, might come in and see pneumonitis and say, “Oh, you only need to give them 40 mg of prednisone.” You might want to question that approach. If somebody has been on immune checkpoints or immunosuppressives for more than 4 weeks, especially if they’re on dual immunosuppressives, have to start thinking about giving them sulfamethoxazole to prevent pneumocystis and maybe acyclovir and watch carefully for secondary infections. We’ve seen, for example, CMV colitis in people who have been on steroids for a while for immune-related colitis. And I tend to reassure patients that their AEs will resolve over time. So it’s really important to prepare people up front to say, “If you have one of these events, it may take months before you get over this. We may take a while to get you off the steroids.” And so that’s why they’re mentally prepared for a long slog, especially if they’re benefiting. You want them to not to put up with it but to preserve until you could get them through this. The adverse events that don’t resolve are the endocrinopathies. And it’s very important to tell people up front, “You may lose your thyroid. You may lose your pituitary. And you might have to be on hormone replacements for the rest of your life.” DR LOVE: I’m just kind of curious. How many patients would you guess you have in your practice right now who are on checkpoint inhibitors? DR SZNOL: Since I treat melanoma and kidney cancer, it would be —- DR LOVE: Like hundreds? DR SZNOL: They’re not on forever. DR LOVE: Yes. No, I understand. DR SZNOL: We do stop the drugs. But yes, I would say. DR LOVE: That’s interesting. What tends to be the time course when you see this spectrum of problems? DR SZNOL: Those data have been published. It’s very variable — some patients will develop toxicities within the first week. Some people can develop it weeks after their last dose. And some things like arthralgias can occur very late. Renal toxicities can occur very late after patients have been on treatment for a while. It’s not true that if you’ve been on treatment for 6 or 8 months that you won’t develop an adverse event. We’ve seen toxicities — for example, when we were giving ipilimumab, we’ve seen toxicities develop — GI toxicities a year, a year and a half later. DR LOVE: While they’re still on the drug though? DR SZNOL: Yes, they were still on the drug. Therapeutic approach to disease progression on an immune checkpoint inhibitor DR LOVE: So one other related issue is the management — and again, you hear this in many different cancers now and I know everything is different — but I’m curious about the management of the patient who’s progressing, at least by imaging, on a checkpoint inhibitor. Kind of the global approach that I’ve heard, in general, for a long time is, if the patient’s clinically stable, keep going. Is that still the way you approach these patients? DR SZNOL: Yes. There are different kinds of progression, and it also depends when they’re progressing. If they’re progressing early, that could be pseudoprogression. You may want to continue for a while longer to make sure they don't have late responses. If they’ve been on treatment for a while and they just have 1 site of progression, or a limited number of sites of progression, you can treat those locally with radiation or surgery and you’re still getting substantial benefit. But the person who’s getting multiple sites of progression while they’re on treatment, those patients you shouldn’t continue on therapy. There is a caveat, though. Let’s say that you’ve received ipilimumab/nivolumab and you’ve responded. You then go on nivolumab. You’re on nivolumab for a while, and on nivolumab you develop disease progression, even in multiple sites. Some of those people you can reinduce with ipilimumab/nivolumab again and see a second response. So those people you can go back to the original induction regimen and get additional benefit. If somebody has received treatment, has gone off treatment for some period of time and then starts to progress, those people can be reinduced, often with the same drug. And in some percentage of those patients, they will have a second response. DR LOVE: What about the so-called phenomena of hyperprogression? Do you believe it exists? Have you seen it? DR SZNOL: I haven’t seen it in melanoma, and I haven’t seen it in kidney cancer. My colleagues who treat lung cancer or head and neck cancer are convinced that it exists. DR LOVE: Those are bad diseases too, but we’ll see. Duration of immune checkpoint inhibitor therapy for melanoma DR LOVE: Another thought, as you were chatting, I was thinking about another issue that I hear a lot about from oncologists, and, of course, this goes way outside melanoma, which is the issue of stopping or how long you can use a checkpoint inhibitor. And that kind of gets woven in, because once you see toxicities, that thought comes into your mind. So how do you think through, in melanoma, the issue of stopping checkpoint inhibitor therapy or duration of treatment? DR SZNOL: Melanoma may be different in other diseases, because we really do see CRs that are durable when we stop the drug. So if somebody develops Grade 3 or 4 toxicity from ipilimumab/nivolumab but they have a complete response when we finish their steroids, I don’t often give them nivolumab maintenance. If they don’t have a complete response, I may give them nivolumab maintenance for 6 months or 8 months. But if they’ve been in a continuous CR for 2 sets of scans, for example, I’ll stop the drug. One of the things that we like to do is — and this is work done by Georgina Long, but we’ve been doing it outside of study data for a while — after about a year, when I think somebody has had a stable response for a long period, I’ll do a PET scan. And if the PET scan is clean, if there’s no uptake — even in things that look like disease, I feel much more comfortable about stopping them. If there’s still active disease there, I probably will continue the drug for a while longer. It’s really important in melanoma when you have residual activity. I take those things out. We’ve had a couple of patients that have stopped drug that have had lymph nodes or other areas with very low-level FDG activity. In some cases we take that out and it’s just necrotic mess. But in some cases, it’s still melanoma. And so we take those out, and in some cases we see very long progression-free survival after doing that. But the short answer, to your answer to your question, is, we tend not to go much beyond a year, year and a half. We tend to do those PETs when we think they’ve had their best responses. They are not responding additionally to drug. And if those PETS are negative, we feel very comfortable taking them off drug at that point. Perspective on the utility of immune checkpoint inhibitors for patients with preexisting autoimmune diseases DR LOVE: So I want to ask you about a couple of related issues. One is the issue of patients with known prior immune disease at the time you’re considering checkpoint therapy. Of course, it’s different in the adjuvant setting versus metastatic. It’s, of course, the type of cancer, et cetera. But globally, what do we know about the effect of checkpoint inhibitors on prior autoimmune disease? DR SZNOL: We don’t know a lot. There’s a retrospective paper that’s put together from the experience of a number of institutions. And it looks like sometimes you exacerbate the prior disease, and sometimes you don’t. Or sometimes you get a different autoimmunity than the one they had when they started the treatment. I think it’s a risk-benefit question for most patients. And if you have metastatic cancer your metastatic cancer is much more dangerous to you than any autoimmune toxicity — than most autoimmune toxicities that you might develop as a result of the drugs. And for most autoimmune toxicities, you can reverse the course with steroids. You can treat the autoimmunity with steroids and reverse the toxicity. So I think in patients who come to me with a prior autoimmunity, I go through the risk-benefit with them, but in most of the cases we decide to go ahead and treat them with the checkpoint inhibitors. Now, it could be that in some patients I may not give them ipi/nivo. I might just give them nivolumab alone. Although, that’s not based on any data, because it could be that the nivolumab is worse for exacerbating their autoimmune toxicity than the ipilimumab. We have a patient who had multiple sclerosis. And we’ve given her ipilimumab. We’ve given her pembrolizumab. We’ve given her the combination of ipilimumab/nivolumab. And her multiple sclerosis has not changed. We’ve never made it worse, it seems like, with these drugs. You would have expected that she would have been very debilitated by this, but that’s really not the case. DR LOVE: I mean, of course the severity, the threat, how much treatment, all these things come into play. But just inherently, are there some autoimmune problems that because of the biology you’re more concerned about? I’ve heard people say isolate multiple sclerosis as an example of a type of pathophysiology that should be worse. And then you could look at psoriasis, rheumatoid arthritis. When you look at the different pathophysiologies, are there some that jump out at you as being more likely to get worse? Or do we have any clinical data on it? DR SZNOL: We don’t have clinical data. There’s a study that’s being put together by the NCI to look at this. They’re going to take groups of patients with each of those different autoimmunities, treat them with nivolumab and then in a very careful, premeditated way, study their outcomes. And I think that’s going to be very valuable. At the moment, it’s all anecdotal. Obviously we’re not going to be treating patients with these agents if they have autoimmune toxicities and they’re requiring large doses of immunosuppression. Because if they’re on large doses of immunosuppression up front, it’s very unlikely that the checkpoint inhibitors are going to work. So the patients that we’re really talking about are patients who have a history of autoimmunity, but they’re really not currently on major immunosuppression. They’d be on very low-dose steroids, for example, and those are the ones that we’re considering for these drugs. Use of immune checkpoint inhibitor therapy after organ or allogeneic transplant DR LOVE: What about a patient with solid organ transplants — kidney, liver, heart? DR SZNOL: There’s data in the literature. I would be very hesitant to give to those patients immune checkpoint inhibitors. I think some patients have been reported to have rejected their transplants because of this. But again, it’s a risk-benefit ratio. So there is a set of allotransplants that I would consider. So we published on patients who had allogeneic bone marrow transplants in the remote past. And we’ve given those patients ipilimumab and nivolumab, a small number of patients, and we’ve gotten away with it. They haven’t rejected their graft. But if you’ve had a kidney transplant or a heart transplant, that may be a different risk-benefit. DR LOVE: These patients, are they on chronic immunosuppressives? DR SZNOL: No. So the allo bone marrow transplant patients that we treated with checkpoint inhibitors were not on any immunosuppressives at the time. DR LOVE: What do we know right now about the relationship, if any, of autoimmune toxicity treatment benefit? DR SZNOL: It’s a hard question. Let’s take nivolumab, for example. If you look at the nivolumab data in melanoma, and these data have been published by Jeff Weber et al, there is a correlation between response rate and development of any toxicity or any select toxicity. So for nivolumab, patients who develop toxicity appear to have at least better objective response rates. For ipilimumab/nivolumab that’s also true, but almost everybody who receives ipilimumab/nivolumab develops an adverse event. The ones that don’t do much worse. We’ve shown that. But a lot of those patients never made it beyond 1 or 2 cycles, because their disease was progressing so rapidly. So we can’t tell whether that’s because they’re not responding to the drug or because they just got a couple of cycles. They didn’t get enough drug to develop an autoimmune toxicity. We’ve also looked for ipilimumab/nivolumab, whether it makes a difference whether you get your adverse event early, before 12 weeks or after 12 weeks. And it doesn’t seem to make a difference in terms of overall outcome. Association between the gut microbiome and response to anti-PD-1 antibody therapy in metastatic melanoma DR LOVE: One final thing I want to ask you about is, I’ve always been fascinated by the emerging research looking at the biome, the microbiome. And there have been some reports coming out maybe linking some of that to checkpoint inhibitors. Also the issue of antibiotics in checkpoint inhibitors. DR SZNOL: So there are a couple of groups that have looked at use of antibiotics in the period immediately before getting checkpoint inhibitors and in the 2 or 3 months after receiving checkpoint inhibitors. And they’ve reported primarily on invalidated cohorts, although modest-sized cohorts, that patients who received the antibiotics will have a worse outcome, presumably because you’re modulating the microbiome in those patients. I think we have to wait to see bigger sets of data before we believe that that’s true. I obviously try and avoid antibiotics if I have to in those patients based on that data. But if somebody needs antibiotics, they need antibiotics. The data from Jen Wargo and Tom Gajewski and a few others, Laurence Zitvogel, I think, also suggest that there are certain bacterial species that are associated with better response or with worse response to the immune checkpoint inhibitors. The question is, again, will those validate in larger cohorts and in second cohorts? I suspect there is a role. I just don’t know how big of a role. And it seems hard for me to believe that your gut bacteria will be the ultimate determinant of whether you respond to an immune checkpoint inhibitor or not. Maybe, maybe there are people who can’t prime because they don’t have the right gut microbiome. Maybe that’s the reason why they don’t have T-cells in the microenvironment. But the more direct biomarker might be to look at the microenvironment. The other question is, can you modulate their microbiome and make them sensitive to immune checkpoints? And those trials are ongoing now. DR LOVE: Just flashing on a conversation I had with Jeff Weber. I guess he’s interested in it. They collect stool specimens on these patients. Interesting to see how that goes. DR SZNOL: I’ve tried to get our people to collect and store stool specimens. They haven’t been too enthusiastic about it. Case: A 64-year-old man with newly diagnosed, symptomatic metastatic melanoma and a BRAF V600E mutation receives dabrafenib/trametinib DR LOVE: So let’s talk a little bit about your cases, beginning with your 64-year-old patient. DR SZNOL: So this is a 64-year-old who had rapidly progressive, newly diagnosed BRAF-positive — V600E BRAF mutation — metastatic melanoma. Probably don’t have to go through the primary presentation. But about January of 2018 — he had had a prior melanoma years ago. He was doing very well, but around May of this year, he started developing nonspecific abdominal pain, increasing abdominal girth and early satiety. And his symptoms increased. He went to see his primary care physician. Was started on antibiotics. They thought he had diverticulitis. His symptoms persisted, and then he came to us. They did the scans before he came to us, and he had this huge mass in the abdomen. It was in the middle of the abdomen. It was extending down in the pelvis. It was at least 20 centimeters by the time I saw him but had been growing very rapidly over a period of around 4 weeks. He had a laparoscopy. They did a biopsy. It turned out to be, as I said, melanoma. It was BRAF V600E mutation-positive. And when I saw him in the clinic, he was really in bad shape. He had bilateral peripheral edema. He had ascites. He was symptomatic. And we had to make a decision at that point, after we fully staged him, whether to give him a BRAF/MEK inhibitor or ipi/nivo. DR LOVE: I was just flashing on the fact that this is the 1 case that I can see you’re going to use BRAF therapy. This is the classic Dr Sznol BRAF case, I guess. Does it have to be this bad, incidentally? DR SZNOL: It does, actually. So anything short of somebody this sick, I would use ipilimumab/nivolumab first. But in this gentleman, his disease was growing really as I was talking to him. And he was very symptomatic. We had a very long discussion internally whether to give him ipilimumab/nivolumab first or BRAF/MEK inhibitor first. And I can’t say all of my colleagues that I work with agreed. They’re very much believers in that you can see very rapid responses to ipilimumab/nivolumab. And if this person had had a wild-type tumor, no question, we would have given him ipilimumab/nivolumab. And we’ve seen benefit in those patients. Response rates with dabrafenib/trametinib and nivolumab/ipilimumab as first-line therapy for metastatic melanoma with a BRAF tumor mutation DR LOVE: What would you say is the response rate in this situation to both strategies? DR SZNOL: To the BRAF/MEK inhibitors, the response rate’s very high. You’ll see improvement. The problem is, in somebody who comes with this kind of rapidly progressive disease, high LDH, huge masses, they’ll respond for some period of time and then they’ll progress. And if you look at the survival of patients who start with high LDH and get BRAF/MEK inhibitors, it’s not very good. On the other hand, if you look at ipilimumab/nivolumab and you look at patients who come in with high LDH, you cure some of those people. DR LOVE: What about response rate with ipi/nivo? DR SZNOL: Hmm. That data is known. I have to actually look at the forest plot from 067 to tell you what that is. DR LOVE: But what about globally? You’re talking about in terms of LDH high. But I mean in general. DR SZNOL: Oh, in general. The response rate to ipi/nivo is somewhere between 55% to 60%, patients with metastatic melanoma who present without prior treatment. DR LOVE: So what happened with this patient? DR SZNOL: So I gave him a BRAF/MEK inhibitor. Within days, he started to feel better. Within days. And I saw him every 2 weeks to make sure that he could tolerate the drugs. He did not get the fevers from the dabrafenib/trametinib. We scanned him at 6 weeks. He had an amazing response. Still had a huge mass, but it was now, instead of a solid mass, it was mostly internally necrotic. His lung lesions disappeared. His peripheral edema went away. His performance status went from 0 to 100% — not 0. It went from a 2 or 3 to a 0. So he was now eating well. Energy level was normal. No pain. Switching to nivolumab/ipilimumab for patients experiencing a response to dabrafenib/trametinib DR SZNOL: And at this point, we ordered the ipilimumab/nivolumab. We’re going to keep the BRAF/MEK inhibitor going until the day that we start ipi/nivo. And then we're going to follow him weekly in the clinic to make sure that his disease doesn’t come roaring back when he’s off the drugs. DR LOVE: So I’m going to guess that if I presented this to a whole bunch of melanoma investigators, not all of them would do exactly what you did, at least in terms of adding the ipi/nivo. Oh, this is this one, progression. You’re going to keep him — DR SZNOL: No. No. No. DR LOVE: That’s what I thought. You’re just going to go ahead and switch him. DR SZNOL: I’m switching him at 6 weeks. DR LOVE: So am I correct in saying not necessarily all investigators would do that? Or do you think most people would? DR SZNOL: Not all investigators would do that. DR LOVE: I haven’t even heard people talk about this, to be honest with you. DR SZNOL: You mean switching over? DR LOVE: Yes, just switching over in somebody who’s responding well to BRAF/MEK in therapy. DR SZNOL: We know from the data that these people will do horribly if you just keep them on the BRAF/MEK inhibitors. Now, if you switch them at the very first sign of progression, that might be okay. If you give them ipi/nivo at the very first sign of progression off the BRAF/MEK inhibitors, they might do just as well. But when they start to progress, sometimes they can progress very rapidly. DR LOVE: I mean, aren’t there patients, though, with very, very prolonged responses to BRAF/MEK inhibition? I mean, like, 5 years or more? DR SZNOL: In this subgroup? Not many. DR LOVE: Because of this subgroup. I see. So you wouldn’t necessarily do this for some other patient? DR SZNOL: Oh, no. DR LOVE: I see. It’s because of the tumor burden. Management of dabrafenib/trametinib-associated fevers DR LOVE: Can you talk a little bit about what happened in terms of — now, this patient actually got dabrafenib/trametinib, and I’m going to ask you about other options, but what happened in terms of side effects also? DR SZNOL: He had a little bit of fever in the first few days. He didn’t actually stop the drug, even though that’s what we usually tell them to do. About 4 or 5 weeks in, he also developed a fever. He stopped the drug for a couple of days and then restarted but had no other adverse effects from the BRAF/MEK inhibitor. DR LOVE: Any comments on what your experience has been with fevers and other side effects? DR SZNOL: With dabrafenib/trametinib, a reasonable number of patients get troublesome fevers. You can manage it. Most people, you manage it by just stopping the drug. They go on ibuprofen. You wait 24 hours until the last fever and then restart the drugs. Usually, the fevers occur early on during treatment. They get to a point where they rarely get fevers. But I’ve also had patients who — the fevers actually get worse, and I’ve had to take them off drug. And they get bad fevers — 102. They feel terrible. So we can manage it, but it is a little bit of a problem in management. I don’t think it decreases efficacy. And as I’ve said, we’ve only had to take off a few patients because of fevers. Incidence of treatment-associated fevers with dabrafenib/trametinib and encorafenib/binimetinib DR LOVE: I’m curious. We now have encorafenib/binimetinib available. Can you talk about the data we have there and how it compares to, for example, dabrafenib/trametinib?DR SZNOL: So the incidence of fevers is much lower with encorafenib/binimetinib — that’s the only problem with those drugs. DR LOVE: Bini. DR SZNOL: Yes. I have to learn how to pronounce those drugs. So now we're starting people more on encorafenib/bini than we have on dabrafenib/trametinib. Just simply because the fevers are easier to manage — there’s less incidence of fever. DR LOVE: Any sense about efficacy? I’ve heard people say that they think maybe it’s more efficacious, but not for sure. DR SZNOL: I myself have not examined the data in great detail. But the people who have suggested that the difference in efficacy may be just due to the patient selection and that it may not really be more efficacious than the other combinations. Case: A 68-year-old man with metastatic melanoma and PD-L1 expression greater than 5% receives ipilimumab/nivolumab DR LOVE: Why don't we go on to your next case, the 68-year-old patient? DR SZNOL: This is a 68-year-old we saw around Christmastime of 2015 — he scraped his skin and, actually, he started bleeding. It’s a lesion on the skin. He went to see his dermatologist. Basically, the dermatologist found 2 suspicious lesions, both of which turned out to be deep melanoma, IIC melanomas, very deep and ulcerated melanomas. I do believe he underwent sentinel node biopsies, which were negative. So the final stage was IIC, but 2 separate lesions. What was interesting is that when we did the molecular analysis of these lesions, one turned out to have an NRAS mutation. The other one turned out to have a BRAF mutation. So he went to his dermatologist. The dermatologist noted a mass in his left forearm. And we biopsied that mass. It turned out to be melanoma. And when we did staging, he had extensive metastatic disease. So we also did a molecular analysis on that recurrent tumor, and it turned you to be the NRAS mutant, not the BRAF mutant. DR LOVE: Wow! DR SZNOL: He had other mutations. It’s not terribly important for this discussion. We also, not uniformly at our institution, but I do PD-L1 assays on these patients. We use not the C-223 antibody. We use another antibody. But in any event, the results are very similar. David Rimm has shown that the results using any of these antibodies are very similar. It turned out that he had 5% to 10% of tumor cells that stained positive for PD-L1. We decided, even though he had a PD-L1-positive tumor, because of the extensive disease that he had that we would go ahead and give him the standard dose of ipilimumab/nivolumab. We gave him 3 mg/kg of ipi together with a mg/kg of nivo every 3 weeks, starting, actually, in mid-June. And again, this is a patient who had a high LDH and extensive disease. At the time that he started treatment, although the difference between this patient and the last patient is that the rate of progression wasn’t nearly as rapid and the extent of disease wasn’t as large. And he wasn’t nearly as symptomatic when he came in. So after the second cycle, he developed diarrhea, but it self resolved. So we didn’t have to give him any treatment. DR LOVE: Before you go on, could I just to kind of stop at this point? You mentioned he had an NRAS mutation. Where are we today in terms of targeted therapy of these patients? DR SZNOL: Nothing that’s approved. There was a study with bini, which is a MEK inhibitor, which was really a negative study but may have shown a little bit of activity in patients who have had prior immune checkpoint inhibitors, a slight improvement in progression-free survival. But I would say we have nothing effective for this group of patients at this point. Testing for PD-L1 expression in patients with metastatic melanoma DR LOVE: The other question is — it’s funny, because every time I ask somebody about PD-1 levels in melanoma, they all say they don’t do them. So I’m glad you do. Because I was going to ask you, like, what’s a high level? What’s like the highest you would see? DR SZNOL: I’ve seen people with 50%, 60%, 90%. DR LOVE: So, for example, if this patient had had a 50%, 60%, still ipi/nivo, or would you have thought about nivo alone? DR SZNOL: That’s a really hard question. If you look at the 067 data, in a retrospective, post-hoc analysis looking at PD-L1 of less than 1% versus 1% or greater, in the ones who had PD-L1 1% or greater, there seemed to be no difference in overall survival between the combination and anti-PD-1 alone, ipi/nivo and anti-PD-1 alone. It was in the group that was PD-L1-negative by that assay that appeared to have most of the benefit from addition of ipi. But there are other data that suggest that — and again, it’s not broken down by PD-L1 status — that the patients who get ipi/nivo, even though their survival may be the same, even when you look out at 3 or 4 years, fewer of the patients who had ipi/nivo required additional treatment beyond the anti-PD-1. So you got them into remission and they stayed in remission. DR LOVE: I mean, that initial data sounds like it was kind of an interesting algorithm. Made a lot of sense to me, anyhow. And then started asking people and, kind of, people weren’t — it seemed like more the way people look at it — I don’t know whether you do — is, if the patient can tolerate ipi/nivo and they have metastatic disease, give it to them. DR SZNOL: That’s mostly the way we approach it. But we really need more data, because you’re buying a whole lot more toxicity, as you’ll see in this patient, by adding the ipilimumab. And if, in fact, at the end of the day survival is going to be the same, that might not be the right strategy for a subgroup of patients. DR LOVE: Any other biomarkers that might be helpful or being looked at? Of course, in other cancers tumor mutation burdens is being looked at. Is that helpful at all? DR SZNOL: Not in melanoma. Not as far as I know, in terms of differentiating between using the combination versus anti-PD-1 alone. So I get it. Whether I actually act on the assay or not depends on the individual patient. If you have somebody who’s 75 years old and wants treatment and has a lot of other comorbidities and they’re PD-L1-positive, that patient I might give anti-PD-1 to. On the other hand, if they’re PD-L1-negative, they can still benefit from anti-PD-1 alone. We know that from the trials. But because they may get added benefit from adding ipi, if I think I can get away with it, I’ll add ipi to that individual. DR LOVE: I was just flashing. I work with your colleague Dan Petrylak in GU, and I was thinking about this strategy some of the people do there where they’ll start LHRH agonist, see how people do and then add in whatever, chemo or hormones. What about that strategy in your 70-year-old, starting with PD-1 alone and then adding ipi if the patient doesn’t respond? DR SZNOL: So I always leave that option open. If I start with anti-PD-1 alone and they don’t respond, I’ll then go and give them the combination if I think they can tolerate it. But I don’t do that in every patient, because I’m always concerned that things have changed when you’ve had some progression. And even though you might salvage some of those patients with the combination, at that point the tumor has grown. The tumor is bigger. Who knows how you’ve modified the immune microenvironment by giving anti-PD-1 alone? And so I’m not a fan of that approach in patients in whom I think I should give the combination to up front. Clinical experience with immunotherapy-associated uveitis and vitiligo DR LOVE: So what happened when you started the ipi/nivo? DR SZNOL: So he tolerated relatively well. He had a rash at the beginning, after cycle 3. He had a little bit of diarrhea, but I didn’t need to give him steroids. It resolved on its own. He came in after that third cycle, and he was complaining of myalgias, fatigue. He had weight loss. His free T4 was elevated. So as commonly in these patients you see a thyroiditis, you don’t really need to do very much about it. We went ahead and gave him the fourth cycle. And in a few days, after giving him the fourth cycle, he called with blurry vision. He said it was intermittently severe. He couldn’t read very well. He had more nausea, early satiety. He actually didn’t feel very well. He also was a little bit orthostatic. So we brought him in. We had ophthalmology see him, and they confirmed that he had bilateral anterior uveitis. But also, anterior uveitis doesn’t make me as nervous as posterior uveitis. He also had posterior uveitis. So we immediately brought him into the hospital, gave him methylprednisolone, a gram a day for 3 days. And then went down to prednisone, 100 mg daily, with a very slow tapering. Obviously, the ophthalmologist gave him steroid eye drops. He didn’t really call us. In fact, he missed one of his subsequent appointments but eventually came back in. We did an MRI of the brain and of the orbits. It looked like things were getting better. We continued the steroid taper. And then on September the 18th, which was about 3 months after we had started the ipi/nivo, we did the scans and he had amazing resolution of disease. All of these masses, the subcutaneous masses, were gone. I couldn’t find them on examine anymore. It was an amazing response. So we’ve continued the prednisone eye drops. We’ve tapered the steroid. He still has some visual. He’s 80% back, 90% back. Not 100% back. We’re not going to give him nivolumab maintenance, because he had such a wonderful response and we’re just going to scan him about every 3 months in the clinic. He’s also, by the way, developed vitiligo. DR LOVE: Really? DR SZNOL: Which we see. Yes. It’s not uncommon. DR LOVE: Is that associated with benefit? DR SZNOL: I don’t know if it’s associated with benefit, but we see it in a lot of people who develop these great responses. DR LOVE: Huh. He’s Caucasian? DR SZNOL: Yes. DR LOVE: And, hmm. He just lost pigment everywhere? DR SZNOL: Yeah, and that’s not uncommon in our patients with melanoma. We don’t see it in kidney cancer very often. In fact, almost never. But in melanoma, we see development of vitiligo quite often. DR LOVE: Why do you think that is? DR SZNOL: What’s probably happening is that there’s epitope spreading. So when you attack the melanoma cells, the T cells that are really important are the ones that are attacking the neoantigens. But at the same time, when you’re releasing all these antigens, you’re probably developing an immune response against normal melanocyte antigens. DR LOVE: Hmm. Wow! DR SZNOL: And that does go on and attack your normal melanocytes and cause vitiligo. DR LOVE: Do you see anything in the mucosa or the eyes? DR SZNOL: This is an example of the uveitis is probably an on-target effect. DR LOVE: Right. DR SZNOL: It’s probably the same immune response that was causing the vitiligo is probably causing the uveitis. DR LOVE: Huh! Wow. Interesting. DR SZNOL: So we see this quite commonly. And my favorite story is of our first patient who developed uveitis. She developed it after the first dose. She had a mucosal melanoma that had spread to multiple sites. Within a couple of weeks of the first dose, she developed terrible uveitis. We put her on high-dose steroids. When she came back, she had an amazing response. And then the drug monitor — this was on a trial — had asked me, he said, “This is a toxicity.” And I said, “It’s not a toxicity. It’s an on-target effect.” And he asked me why I thought that. I said, “Because, basically, you’re attacking melanocytes in the eyes, so this must be an immune response that’s attacking the melanoma and it’s cross-reacting with normal melanocytes.” But at the time, when I saw the patient in clinic, she came back in and I hadn’t undressed her yet, and she still had black hair and black eyebrows. And I said to her, “I’m wondering whether you’ve noticed any patches of whitening of your skin, lightening of your skin.” And she goes, “Actually, a couple of weeks ago all my hair turned white. But I dyed it.” DR LOVE: Wow! Wow! DR SZNOL: So she actually had an on-target effect, very similar to this patient. DR LOVE: Wow. Anyhow, it sounds like even though it was pretty scary, but great outcome. DR SZNOL: It’s a great outcome. And we’ve had other people with visual problems. They’ve recovered their vision. And I think he’ll go on with a very long, durable CR. We don’t know yet, of course, because it’s only been a few months. But I think he’ll do very well. Case: A 47-year-old man with metastatic melanoma experiences dermatologic toxicity with ipilimumab/nivolumab DR LOVE: You had some patients with toxicity I thought we could just maybe — rather than go through the whole case, really just focus more on the toxicity, beginning with the 47-year-old patient with the dermatologic toxicity. DR SZNOL: So this is a gentleman who had a primary ocular melanoma to liver and bone. As you know, we don’t have a lot to treat these patients with. We have a rare patient who has benefit from ipilimumab/nivolumab with an ocular metastatic melanoma. We’ve had several over the course of the years. We went ahead and treated him with ipilimumab/nivolumab. He developed a rash initially, was treated normally. We didn’t do anything about it, because it looked like a routine rash. He then went on and developed colitis, which we treated with steroids, our standard high-dose steroid regimen. But while he was on steroids, he called and reported bleeding from hemorrhoids. He wasn’t eating very well. He was orthostatic. And then he said, at about the same time — again, he was on steroids at the time — that he had developed this new rash on the back on the legs that was spreading very rapidly. He was on sulfamethoxazole at the time, because we had put him on sulfamethoxazole as prophylaxis for the prolonged steroids that he had been on. He came into the ER, was seen by a dermatology resident in the ER and the dermatology resident immediately said, this is Stevens-Johnson syndrome. So he was admitted to the hospital. Was put on very high-dose steroids. DR LOVE: Could you describe kind of his physical exam at that point? DR SZNOL: The dermatologist described it as hundreds of red, dusky, irregularly shaped macules and thin papules coalescing into large patches and thin plaques on the trunk. And at that time, there were some erosions on the skin but no mucosal involvement. DR LOVE: Did they biopsy the skin? DR SZNOL: They did. It showed epidermal necrosis and lymphoplasmacytic infiltrate consistent with, essentially, Stevens-Johnson syndrome. So he was stable for a couple of days, but you could see the rash spreading into his palms and his soles, with a lot more blisters and erosions, eventually just sheets of skin starting coming off. DR LOVE: Wow. DR SZNOL: Still had diarrhea. He went to the ICU. We tried to give him infliximab. I think the management of Stevens-Johnson syndrome is a little bit controversial. When I spoke with the dermatologist, they weren’t all that interested in giving him steroids or the infliximab, as I recall. But we did. Continued to get worse. He was transferred to a burn unit for management and died while he was in the burn unit. DR LOVE: Wow, that’s terrible. DR SZNOL: Yes, it’s a terrible story. DR LOVE: Can you talk a little bit about the spectrum of dermatologic problems? I know that low-grade problems are pretty common. What kinds of low-grade problems do you see? And what are some of the severe problems? DR SZNOL: The low-grade problems really are things that we mostly ignore. It’s pruritis and then patches of erythema. Sometimes punctate lesions. Not anything that we pay much attention to, and people tolerate it very well. We give triamcinolone cream or whatever is that they want to use, more than just the hydrocortisone creams that you can get over the counter to treat the rash. They do very well. We often have them seen by the dermatologist, but it’s not necessary because they do okay. There’s a group of patients that can develop very severe rashes. I’m not a dermatologist, but we’ve seen bullous pemphigoid, and we’ve had to stop treatment because of the bullous pemphigoid. So they develop blistering lesions on their skin. So far, nothing life threatening. The most severe skin reactions, there are patients who come in with very Grade 4 rashes. It’s not Stevens-Johnson syndrome. It’s not bullous pemphigoid. But they’re covered head to toe in what — it’s almost — it’s an erythema, but it’s almost an indurated erythema. You could almost feel it. It looks very scary. But again, you can treat them with oral steroids and topical steroids, and they tend to do quite well. And then the one Stevens-Johnson syndrome that I just discussed a few minutes ago. Case: A 31-year-old man is diagnosed with metastatic mucosal melanoma DR LOVE: Why don’t we finish out with the last patient, your 31-year-old man with the mucosal melanoma? DR SZNOL: So this is a gentleman, he’s black, and he presented to an urgent care clinic with 2 weeks of constipation. He said he felt something pop while he was having a bowel movement, and he felt a soft mass in his rectum. And, of course, when most of these people get to the doctor, it’s diagnosed initially as hemorrhoids. But eventually he was seen by a surgeon. They did a biopsy, and it turned out to be a rectal melanoma, a mucosal melanoma. They did an excisional biopsy, but they also scanned him at the time. And it turned out that he had not only multiple subcentimeter pulmonary nodules — not skin lesions — he also had lymphadenopathy, and he had a liver lesion that was clearly suspicious for metastatic disease. He also had retroperitoneal implants. So he had a fairly disseminated disease. We sent an OncomineTM panel, because about 10% to 15% of these people will have c-Kit mutations. And, although we wouldn’t treat them with c-KIT inhibitors first, if they don’t respond to the immune therapy, it gives us second line of treatment. Therapeutic options for patients with metastatic mucosal melanoma and a c-KIT mutation DR LOVE: As long as you brought that up, what is your second line? What would you do? DR SZNOL: If you had a c-KIT mutation? Believe it or not, we start most people on sorafenib. DR LOVE: That’s interesting. What do we know about the response in that situation? DR SZNOL: We don’t have any organized data. A lot of people use imatinib and dasatinib. I use sorafenib, because some of the KIT mutations that we see are only sensitive to sorafenib or are mostly sensitive to sorafenib, not as sensitive to imatinib and dasatinib. The other thing I like about sorafenib is that it’s a VEGF receptor inhibitor. And I’ve treated patients who have been on imatinib/dasatinib. They’ve progressed on that. And then put them on sorafenib and seen wonderful responses to second-line treatment with sorafenib. So we started doing this many, many years ago. And sorafenib has just been our go-to drug for patients who we treat with c-KIT inhibitors. Activity and tolerability of ipilimumab/nivolumab in patients with metastatic mucosal melanoma DR LOVE: So what happened with this patient? DR SZNOL: I did the PD-L1 staining, and as we would expect with a mucosal melanoma, he had no staining. Less than 1% of tumor cells, zero immune cell staining. That’s true for the majority of patients with mucosal melanomas. If they are PD-L1-positive, by the way, the data would suggest that they can respond to single-agent anti-PD-1 fairly well, as well as a cutaneous melanoma. But most of them don’t have PD-L1 staining. They don’t respond to anti-PD-1. So we gave him ipi/nivo. He tolerated the therapy really well. Very soon after we started treating him, we noticed vitiligo. And in a black male, it’s very striking. I mean, really striking. DR LOVE: Complete vitiligo? DR SZNOL: No. No. DR LOVE: Like patchy? DR SZNOL: Very patchy, but there were areas that were complete vitiligo. But overall, if you look across all his skin surface, it was patchy. He did very well. He developed a little bit of arthralgias but tolerated the treatment very well. And when we did his first CAT scan, everything was better. He had a great response. About a third of patients who have metastatic mucosal melanomas will have great responses to ipi/nivo. We had to hold his treatment once. I gave him a short course of steroids for the arthralgias. He came off. Then we put him on nivolumab maintenance again. And we’ve been scanning him ever since. He’s a near-CR. He’s had not progression now for months. I actually have a PET scheduled for him this coming month, and if his PET is negative, I’m going to stop his treatment. DR LOVE: How many cases like this do you see every year? DR SZNOL: I’d say we see somewhere between 5 to 10 metastatic mucosal melanomas per year. More like 5, I guess. DR LOVE: Where is the primary usually? DR SZNOL: We see the sinonasal primaries, vaginal primaries and rectal primaries, primarily. We see them across all sites. So the vaginal primaries, obviously, come to us from the gyn oncologist. The rectal primaries, like him, came from a surgeon. But we also get a number of these from our head and neck surgeons. DR LOVE: Do these patients usually present with metastatic disease? Or do you see some that are localized, surgically cured? DR SZNOL: We see some that are localized. And some of them are cured. But our impression is that these patients have, for the same depth and stage, if you use the standard staging for cutaneous melanoma, that they have a higher risk for recurrence. DR LOVE: And do you see BRAF positivity with mucosal melanoma? DR SZNOL: The answer is yes, but much, much lower frequency. And I’m trying to remember whether I’ve ever actually seen one, and I can’t remember whether I’ve actually seen one that’s BRAF mutant. Conjunctival melanomas can be BRAF mutant-positive. I can’t remember seeing a rectal, a vaginal or a sinonasal one, but I think they’re reported in the literature. DR LOVE: Do you ever see cases where you consult about adjuvant therapy with mucosal melanoma? DR SZNOL: So yes. And it’s a problem, because not enough of these patients have gone on to the randomized trials to know whether nivo would work. My own suspicion is that nivolumab, as a single agent in these patients, would not work well, because it doesn’t work very well in the metastatic disease setting. So I have tried to get ipilimumab/nivolumab for a couple of patients who I wanted to treat in the adjuvant setting because I thought they were very high risk, but it’s actually not approved by insurance companies. There is a trial ongoing, I think at the Mayo Clinic, a single-arm study. But, in fact, I just saw a patient within the last 2 weeks that had a vulvar melanoma that was very deep, 8 millimeters deep, was ulcerated. She had bilateral sentinel nodes. The sentinel nodes were negative. But I think she’s at very high risk. She had perineural invasion. She had many — so 17 mitoses per mm2. I realize those are not used in the staging system, but when you look overall at this lesion, you just get a sense that she’s at very high risk for recurrence. So I wanted to offer her ipi/nivo, and at this point I haven’t yet been able to get the insurance companies to approve it. Selection of adjuvant therapy for patients with melanoma and a BRAF tumor mutation PROF ROBERT: Of course the question is addressed when you have the patient with a BRAF-mutant melanoma, because then you have 2 possibilities. Are you going to give this patient an anti-BRAF/anti-MEK combination? We have the COMBI-AD trial, which is quite a major trial. We already have overall survival, and we already know that we have a benefit with this combination as compared with placebo. And on the other hand, we have patients who have been treated with pembrolizumab or nivolumab. So the trial for nivolumab was just with ipilimumab, which was already an effective treatment, and the trial with pembrolizumab was versus placebo. That is to say that we have also the anti-PD-1 monotherapy that can be given to this particular patient. So how are you going to choose? Honestly now, you cannot choose based on the clinical trial, because we don’t have the same line of evidence. The trials with anti-PD-1 are less mature. They’re also positive, but we don’t have the overall survival, and we don’t have a head-to-head comparison. But if you have the choice — I mean, this is a very debated question. I was recently at a meeting where we had with one of my colleagues who is also a key opinion leader, each of us had to defend one position. And honestly, I mean, I could have defended the other position, because right now we don’t know. I can give you an example. I had a very young patient, 18-year-old. This is rare, but we can have patients very, very young with melanoma. And this young woman had a very, very pejorative melanoma with a relapse with multiple lymph nodes in the cervical area after a melanoma on the head. And this patient was BRAF mutant — the melanoma was BRAF mutant. But she had Crohn’s disease. So this is a quite easy situation. We went and give her anti-BRAF/anti-MEK. If she did not have Crohn’s disease, I think you have — I mean, maybe some physicians are very convinced that one’s better than the other, but honestly, right now you cannot really be very convinced because of what I just said: We don’t have the comparison head to head. And I think what is extremely important is to give your patients the most clear information, because you also have patients who would be very afraid to have an adverse event that would be permanent. Like, you know that if you give an anti-PD-1, you might have your thyroid stopping functioning. And that you would have to take hormones all your life. On the other hand, with anti-BRAF/anti-MEK, you have twice as many adverse events, but it’s very rare that they can persist. So you have to explain all that to your patients. And today, there is no right or good solution. Both treatments are acceptable. Clinical benefit associated with adjuvant therapy with BRAF/MEK inhibitors and with immune checkpoint inhibitors for melanoma with a BRAF tumor mutation DR LOVE: So maybe I could dissect out with you a little bit some of the things that you say to your patients, including this patient. Or a patient who — I’m sure some of your patients — maybe you even have physicians who are patients who are pretty sophisticated. So first question would be, indirectly how do you compare the benefit of adjuvant therapy in a BRAF-positive patient, comparing BRAF/MEK to checkpoint inhibitor? PROF ROBERT: Right now, as I said, you have 2 categories of trials. You have much more data and more major trials with anti-BRAF/anti-MEK, and I must say I’m quite surprised of the outcome of this trial. It seems really good. Because we could be afraid, because these treatments maybe we should say that up front, it’s given for 1 year. And we know that in the metastatic setting, if we stop anti-BRAF/anti-MEK, most of the patients will relapse. So we could be afraid that the patient after stopping after 1 year would relapse, and we don’t seem to see that. It suggests that maybe if you have a very small disease, like subclinical disease, maybe you can eliminate all the cells. So you know also that you can stop immunotherapy in the metastatic disease. Because all we know — I mean, the most knowledge comes from what we know from the metastatic disease. We know we can stop immunotherapy. We know we cannot stop targeted therapy. So intuitively, you would think that maybe it would be better to give anti-PD-1. But I must say that with the data that we have, the solid clinical trial-based data, we cannot really say that anti-BRAF/anti-MEK are less good than anti-PD-1 in the adjuvant setting. DR LOVE: So what would you estimate in this particular patient, your 18-year-old woman, what would you estimate her baseline risk of relapse was without adjuvant therapy? And what would it be with either a checkpoint inhibitor or anti-BRAF/MEK? PROF ROBERT: She had the multiple lymph nodes microscopically involved. So the risk of relapse for her was really, really high, something like 70% in the very — I mean, very soon in the 2 or 3 years. So we know that with these treatments, we decrease the risk of death. We know that with anti-BRAF/anti-MEK, significantly more than 50% at 3 years. And with anti-PD-1, we don’t yet know the risk of death. But the risk of relapse, we know we decrease it for close to 50% also. So it’s a huge effect statistically for this patient. DR LOVE: And you mentioned the fact that she had Crohn’s disease. What kind of treatment was she on for the Crohn’s disease, and how bad of a problem was it? PROF ROBERT: She was not in a big crisis. This was not in an acute crisis. But you know you are not — in the adjuvant setting, if you have the choice, it’s better not to really, we say, to give the evil a temptation. She had a local treatment. She has steroid, but locally administered. And she’s doing not so bad, but she clearly has Crohn’s disease. It’s not like patients who have had an inflammatory bowel disease years ago with several years of not 1 single symptom. She had some crisis, like, 1 or 2 years ago. So we felt it was really more reasonable. Use of immune checkpoint inhibitors for patients with preexisting autoimmune disease DR LOVE: What in general — I mean, I’m sure you have much more experience with patients with metastatic disease. But what’s been your experience with people with prior autoimmune disease like Crohn’s disease — rheumatoid arthritis, psoriasis? There are times I think when these patients end up getting treated because there’s really no other option. What’s your experience been? PROF ROBERT: My experience is not so bad. In fact, it happened quite a lot. So it depends on the disease. If you have somebody who had, for example, a Guillain-Barré syndrome, I would not. Even I would be too afraid. But it really happened to me, for example, inflammatory bowel disease, and this is not so rare in the population. What do we do? We work closely with a colleague, with specialists in this disease. And we do a colonoscopy to see how active is the disease at the time we might give the treatment. And then if it’s okay, we give the treatment. And I can tell you that I cannot really tell exactly how many patients I treated like that. But I think that in the majority of the cases it was okay. It did not come back. And also, you talked about arthritis. We have some patients with polyarthritis rheumatoid. And we had some patients that we treated. And I would say that maybe half of them increased their symptoms, and sometimes we could give a treatment like symptomatic treatment. But it is true, you are right, when you are in the metastatic setting, we don’t really have the choice. This is a little bit different for the adjuvant setting. DR LOVE: So what happened to this woman, and what’s her current situation? PROF ROBERT: She’s doing very well. She has — I mean, this young patient who received anti-BRAF/anti-MEK, she’s doing well. She has almost no adverse events, and she’s a little bit tired. You can have some fatigue. But she’s really doing well. DR LOVE: So she’s getting dabrafenib/trametinib? PROF ROBERT: Now she will finish the first year very soon, and we are going to stop. And we are going to really follow up closely, yes. Safety profiles and duration of therapy with BRAF/MEK inhibitor combinations DR LOVE: Could you reflect on what your experience is in using this combination or other combinations in the adjuvant setting, as well as metastatic disease? What are the common problems that occur, and what do you do about it? PROF ROBERT: So you have different combinations. One is dabrafenib/trametinib, where I think we give this one much more often than cobimetinib/vemurafenib. Why? Because we think that it’s easier to manage in terms of adverse events. These 2 combinations, now we have a third combination that is very soon going to be available on the market in France, binimetinib/encorafenib, but the price is not yet fixed. It might be very interesting. But with dabrafenib/trametinib, in addition to the class effect, the adverse events that are linked to the class effect, we have fever, chills, and this can be a problem for some patients. With vemurafenib/cobimetinib, the adverse event that can be really difficult to manage is photosensitivity. So it can seem not such a big problem, but I can tell you it’s a very, very strong photosensitivity if people do not take care, do not exert a very strict photoprotection. They go out, even if they don’t live on an island, on a tropical island — they are red like a lobster and a surfer. So that can be very tricky to manage. But we can go from one combination to the other, depending on the adverse events. But usually I think it’s quite tolerable. It’s very rare that we have to stop definitely this one or the other of these regimens because of adverse events. It’s quite rare. It’s good, because as I said before, we know that if we stop, the probability to relapse is very high. DR LOVE: So we’ll talk a little bit more about checkpoint inhibitors and also checkpoint inhibitors in the adjuvant setting in a second. But I’m kind of curious, what are some of the trials that are going on right now to move the adjuvant therapy forward? And one issue you brought up was the duration of therapy, specifically with BRAF/MEK. I was kind of thinking about the tamoxifen trials in breast cancer. They started out 1 year, then it was 2 years, 5 years, 10 years. Is there an interest in extending the duration of this therapy? PROF ROBERT: That’s a very good question. And if the curves are beginning to fall down after 3 years, I think it is really something that we should evaluate. But presently with the data that we have, we’ll see the follow-up, but right now it’s quite encouraging. And surprisingly, 1 year seems to have a very long-term effect. How to improve this adjuvant treatment we have now, we are evaluating different combinations of immunotherapy, like ipilimumab plus nivolumab. We try also to optimize the tolerance, because, as you know, this is very, very rich in adverse events. So we tried to find a way to decrease the toxicity. But maybe, because we know that this regimen gives a slightly better effect than anti-PD-1 single agent, it is now evaluated in the adjuvant setting also. Tumor mutation burden and other potential biomarkers of response to adjuvant targeted therapy or immune checkpoint inhibition DR LOVE: What about predictors of treatment benefit in the adjuvant setting? Any way to maybe shade it towards immunotherapy versus BRAF, PD-1 levels, tumor mutation burdens? Any factors related to BRAF mutations that push you one way or the other? PROF ROBERT: We would love that. Right now, I must say we do not. We begin to have presentations, and we will have publication very soon about the biomarker analysis that are done in these trials. Right now what do we know? We know that this combination, dabra/treme, or these anti-PD-1 treatments, give good results regardless of PD-L1 expression and most of the biomarkers that we have analyzed until now. But I hope that with more analysis and more follow-up, maybe we will see something coming out and helping us to decide. Right now we don’t have that, unfortunately. DR LOVE: It’s interesting in lung cancer — and even other cancers — we hear more and more about tumor mutation burden. Yet I don’t hear too much about that in melanoma. Has it been looked at? PROF ROBERT: Sure, it has been looked at. Actually, it was the first publication in The New England Journal of Medicine. Patient treated with ipilimumab, and it was shown that patients with high tumor mutational burden responded better. This was a paper by Chung and collaborators. So then there was paper about lung disease. But finally, what do we see in melanoma patients? It is true that if you have a lot of mutations, your probability to respond is higher. However, what we know also from the work that has been done in a very detailed fashion about really trying to know which neoantigens were involved in the response, because what we have to remind everybody is that the tumor mutational burden, why do we think it helps to respond? It’s because at the favor of this genetic instability, the tumor might express some neoantigens, meaning some antigen that the immune system has not seen, so it’s very good to respond. But in reality, what has been very well shown — for example by the team in NKI, the team of Ton Schumacher working really trying to dissect the molecular events at the neoantigen level. In fact, what do we see? We see that you have very few neoantigens that really do the job, that really induce the T-cell response that is going to kill the tumor. And plus, not only there are very few of them, but they are very private. Meaning if you have a patient A with a melanoma and the patient B, the neoantigens are not going to be the same even though it’s melanoma in both cases. But, of course, if you have a lot of mutations, your probability to have this neoantigen is higher. So, in fact, it’s true that in general, it’s better to have a lot of mutations. But sometimes you might respond very well in spite of having very few mutations, if only you have the good ones. So I’m very happy that we are not limited by the tumor mutational burden. But sometimes it can help us. I remember there was a very interesting paper. We have this subgroup of melanoma, very rare, it’s melanoma of the eye, uveal melanoma. This is a different story molecularly. And these melanomas are not linked to the sun, because it’s not sun exposed. It’s a very different story. They have different mutations, and they have very few mutations, very few — 1, 2 per megabase. And they do not respond to immunotherapy. But there was this patient reported in a recent issue of Nature Comm. It was Rodrigues et al. And this patient responded beautifully to anti-PD-1. And you know what? This patient, by chance, had a constitutional mutation that was inducing a lot of mutations. And that was by chance that this melanoma of the eye was highly mutated. So in some cases where you are in the tumor that usually does not have a lot of mutations, it might help us. But routinely it’s not helping us a lot. But just also to really insist on this point, shown by our colleague Paul Nghiem at the last ASCO meeting, that Merkel cell carcinoma either have a lot of mutations or have very few mutations when they have the polyomavirus in their genome. And in both cases, they respond very well to immunotherapy. It means that either they have a lot of mutations, so they might have the good neoantigens that are expressed, or they have, with the virus, very few but very well-poised neoantigens. Emerging data with the novel IDO inhibitor epacadostat and anti-LAG-3 and anti-TIM-3 antibodies for metastatic melanoma DR LOVE: So I’m going to move on now and talk about metastatic melanoma. But just kind of as a follow-up to what you were saying about new trial ideas and immunotherapy new ideas, just kind of curious about other checkpoint inhibitors that are being looked at in melanoma. The ones that I’ve heard about — and I think a lot of people have heard about — epacadostat, LAG-3 and TIM-3. To us, these are barely names, but at least we’ve heard the names. Could you kind of summarize what these other checkpoints are and what we know about them in terms of clinical research? PROF ROBERT: Yes. So epacadostat is a little bit a sad story for us. It’s not really a checkpoint, I would say, but it’s an enzyme that we know is immunosuppressive. It acts on the cycle of tryptophan. And there were some quite promising data in a few patients with a 55% response rate, no toxicity. I think we should have been a little bit more cautious about the fact that this drug does not induce any adverse events. Tell me about one drug that gives no adverse events and that is effective. I don’t know any. But anyway. We combined epacadostat, this IDO — so the enzyme is called IDO — IDO inhibitor epacadostat with pembrolizumab, and we compared with pembrolizumab single agent plus placebo. So a very well-conducted trial. If you look at the curves, you just cannot distinguish them, because they are really on top of the other, PFS, OS. Totally negative trial. At least we didn’t have a bad effect, a pejorative effect. But that was a big disappointment in the community. I must say that it was a long time that we didn’t have a negative trial in melanoma. We now become to be very, very ambitious. We want to do better each time. So it was a big disappointment, but also I think it’s a good lesson. It’s a good lesson, because we need to have more data in the Phase II, more preclinical data to go ahead with a big Phase III. And I hope it’s going to help us to design trials in a better way. Now, you talk about TIM-3 and LAG-3, they are a checkpoint inhibitors also interacting — either interaction between tumor cell/immune cells or immune/immune cells. And these are negative checkpoints also. So we try to block them. And we have clinical trials going on. We have some Phase II. We have some so-called platform trials, where we combine anti-PD-1 with such and such a drug because we look for a signal of efficacy. Because after you fail anti-PD-1 or anti-PD-1 plus ipilimumab, if you are wild-type BRAF, what do you give to your patients? You are in a very difficult situation. So any signal of efficacy — even 10%, 15% — we would try to go and to see who is going to respond. Also, we have some good, very promising results with stimulation of innate immunity, trying to induce a dangerous signal to activate the immune system with, for example, Toll-like receptor agonist. This receptor, they are the same family as TNF-alpha or IL-1. These are so-called primary cytokines. They activate the immune system but in a very unspecific way, innate. They are activated, for example, this Toll-like receptor, they are activated by bacteria or germs. So we can stimulate them with agonists, and we try to combine that with anti-PD-1 or with anti-CTLA-4. And we have some quite promising results. So now, we already have Phase III going. And more specifically, with TLR9 agonist. And there are several of them in development. Choosing between single-agent and combination immune checkpoint inhibitor therapy for metastatic melanoma DR LOVE: Let’s maybe talk about just clinical decision-making nowadays in terms of patients in the metastatic setting where you’re going to use immunotherapy and whether that’s BRAF wild type or not. And, of course, one key question is, are you going to use a checkpoint inhibitor alone or are you going to use a combination like ipi/nivo? Can you talk a little bit about, in your own mind, how you compared those 2 options? And how do you go about making a decision? Do you look at PD-L1 levels, for example? PROF ROBERT: That is a very, very important question. So first, I would like to say that even if the patient is BRAF mutant, I would go first with immunotherapy in the metastatic disease unless the patient has a very, very big tumor burden, very rapidly threatened by the metastasis, very symptomatic. But in the other situation, which is the majority of the cases, I would go with immunotherapy. So now your question is very important: Should we go with anti-PD-1 single agent, about 40% of response rate, about 15% of Grade III and IV adverse events? Or should we go with anti-CTLA-4/anti-PD-1, so nivolumab plus ipilimumab, with a response rate that is a little bit higher, close to 60%, but also 60% of Grade III and IV adverse events? Do I do PD-L1 staining? I do PD-L1 staining, but I do it in a research objective. Because I would not rely on that. The reason is simple. It really happened to me several times to do 2 biopsies in one patient of 2 different metastasis and to have a positive on one side and a negative PD-L1 on the other side. This is not stable. This is not homogeneous. The staining, it is not very well standardized. And here again, I’m happy that we can give the drugs regardless of the staining. It’s a little bit like what we said about the tumor mutation burden. If you have a high PD-L1 staining, you know the chances are high that you are going to respond to anti-PD-L1. But I can tell you I have a lot of patients who have responded with a negative PD-L1 also. So I don’t do it as decision-making. I know some people do it, but we don’t do it in most of the cases in our hospital and in our country. But what I think is going to be very important is to really explain to the patients. Because if you give anti-CTLA-4/anti-PD-1, you know that the patient has a high probability to have strong adverse events, severe adverse events. And he has or she has to be okay to take the risk. And that’s very important that you explain the risk to the patient, and you have really to spend a lot of time explaining it. What I would do, I would propose anti-CTLA-4/anti-PD-1, so nivo/ipi. I would propose that to patients who are in good shape. I would not propose that to a patient who is 90 years old. But patients who are in good shape who really want to fight, I would really propose them this option, because, I mean, even if it’s not much better than anti-PD-1, it seems to be a little bit better. And we know that even if we have to stop for toxicity in 40% of the cases, it’s huge. I mean, close to half of the patients do not receive the entire treatment with the 4 infusions. Why? Because of toxicity. But it does not impair the response. And I can tell you a story of a patient, 1 infusion. I remember, very young man, good shape. One infusion, big adverse events like hepatitis, colitis, fever. We stop. And before even considering coming back with a second infusion, he’s in complete response. He never received another infusion. So that’s great. DR LOVE: Was that ipi/nivo or a single agent? PROF ROBERT: Yes, ipi/nivo. DR LOVE: Wow. Perspective on the association between immune-related adverse events and benefit from immune checkpoint inhibitors DR LOVE: And I’m kind of curious since you brought up that case, do you think there generally is a correlation between autoimmune toxicity and treatment benefit? PROF ROBERT: That’s a good question also. In fact, it has been very well shown with vitiligo or vitiligo id reaction — the pigmentation of the skin — and which implies that your immune system is doing too good, like, is destroying even the melanocytes who are just benign melanocytes. And that we really clearly showed, and we and others showed that if you have a vitiligo, you’ll respond with a higher probability. But for other adverse events also, there is a tendency to show that if you have immune-related adverse events, the response rates are higher. But it’s difficult to do this kind of analysis. Why? Because you have a big risk of bias, because it’s called the time lead bias. If you have a long survival response — so a long survival — your probability to develop an adverse event is higher just because of the time. So you have to work with good statisticians who do the analysis, but getting rid of this time lead bias. But in doing so, yes, you can see in most of the cases in the retrospective study — we cannot do prospective studies for that, of course — we see that it seems to be associated with response. And the best thing, also reassuring thing, is that if you give steroids — because you know that if you have a strong adverse event, you give steroids, antidote to the activation of the immune system — it’s seen that you don’t decrease your response. So that’s quite reassuring. If you need the steroid you can give it, and you don’t impair the good outcome. DR LOVE: So you mentioned — and most melanoma investigators also tell me — that their inclination is to use the combination ipi/nivo when they can. What about the strategy of starting PD-1, starting nivo and waiting and adding in the ipi if the patient doesn’t have a good response? PROF ROBERT: It needs to be evaluated in a clinical trial. We have some data about that from clinical trials where we know that the patient failed, and then we have captured the data of what happened after. And we have about 13% of response after anti-PD-1 failure with ipi. But this is not very clear, clear data, because it was captured like that. It was not prospectively looked at. And we tried really hard to do this kind of trial, actually. It’s not so easy. We would like to do to patients who fail anti-PD-1 to compare ipi single agent or ipi/nivo. That would be very important for the community. But right now I know that, for example, in Europe with the EORTC, we are trying to mount this clinical trial, and it’s quite a lot of effort. I’m not discouraged for trying, but it would be very important. Today we don’t exactly know what’s the best. And also, it depends of the availability of this combination in your countries. But the data that we have available — it’s interesting, because ipilimumab first line, you know the response rate is about — is not really higher than 13%. Depends — it’s between 13%, 19%. So it seemed that you don’t really decrease the rate of response if you use it after anti-PD-1. It’s quite interesting. Duration of therapy and complete response rate with immunotherapy versus BRAF/MEK inhibitor combinations DR LOVE: So you alluded to your preference, I think, for immunotherapy in patients with metastatic disease who are BRAF-positive. What’s your own estimate of the chance of prolonged response, maybe even cure, immunotherapy versus BRAF/MEK? PROF ROBERT: So complete response, complete remission. It means you don’t see metastasis anymore. So if there are some, they are microscopic, so you don’t see anything with your eyes, you don’t see anything with your CT scan, your PET scan. If you are in this situation with anti-PD-1 or anti-PD-1 plus anti-CTLA-4, we now have some data — even some hindsight to be quite confident to stop this immunotherapy. And this is absolutely big, big news. We never could stop before we have this drug in the context of metastatic melanoma. So this is something very new. We stop. But we had to decide on some criteria. So we decided that we need to treat the patient for at least 6 months. Practically, I think most of the patients receive 1 year of treatment by the time you decide, you discuss, et cetera, about 1 year so — between 6 months and 12 months. And then you have to confirm your complete response with 2 CT scans that show the same complete remission. And then you tell your patient, “You know, we have some data now. We have more and more data, and we have more and more follow-up after stopping. And we know that more than 90% of the patients who stop treatment in this situation do not relapse, are still in complete response after more than 5 years.” So this gives us some comfort to tell the patient that we can stop. We cannot oblige them to stop. We have some patients who are very, very afraid to stop. This week I saw a gentleman. He’s a doctor, actually. And he told me, “I’m so afraid to stop. I don’t want to stop.” So in France, we can continue the treatment. We are not limited. But I also can tell you that I had a patient who developed type 1 diabetes after 3 years of treatment. Was it related to the drug? Was it not? I don’t know. But that’s the question. I think if you are in complete response and the patient is not too afraid to stop, I think it’s very reasonable to stop. Now, what happened with BRAF/MEK and also, excuse me, the percentage of complete response with anti-PD-1? About 20%. The percentage of complete response with combination ipi/nivo? The same, not more complete response. That’s interesting. We have more partial response but the same — you know what? I’m looking forward to having a treatment with more than 20% of complete response. With dabrafenib we also have 20% of complete response. But here, you don’t want to stop. Because if you stop, most of them will relapse. DR LOVE: Do you have patients who have a partial response but yet are stable and live 5 years? PROF ROBERT: Yes, we have that also. We have that also, and we also have some results of one trial, the KEYNOTE-006. We had 2 years of treatment in the trial. And we had to stop after 2 years. So we had some patients who stopped in partial response and some even in stable disease. And we also have some follow-up. And after 2 years of median follow-up, most of them were still partial responders or stable disease. But a little bit less patients remain in response than the complete responders. Case: A 27-year-old man with metastatic melanoma and a BRAF tumor mutation receives nivolumab/ipilimumab after experiencing disease progression on a BRAF/MEK inhibitor combination DR LOVE: So I want to hear about your patient here, a really interesting case of BRAF-positive disease, this patient with a brain met. PROF ROBERT: Yes, it’s a young patient who had BRAF-mutant melanoma. He had a lymph node involvement, axillary lymph node, and he also had a bone metastasis. And he was treated with BRAF/MEK. He was treated not in my center. When we saw him, he relapsed. He progressed. But he progressed badly, because he progressed also in the brain. So he was very young. He is also a colleague, a doctor, a young doctor. And so 1 brain met and a relapse in the lymph node. And so we decided to do stereotactic radiosurgery and the unique brain met that was asymptomatic. And to switch to anti-PD-1 plus anti-CTLA-4. So he was informed of the risk. And then he responded in the lymph node, but he had a big enlargement of the brain mets that was treated by radiotherapy, but we didn’t know if it was just not a good effect, inflammation due to the treatment, maybe increased by the radiotherapy or just the radiotherapy was not strong enough to prevent the metastasis to grow. And he became a little bit symptomatic with headaches, so we gave steroids, because he had a brain edema. And we worried a lot about this patient. And then we repeated MRI. And finally 1 month after, the met began to decrease, and he finally responded. It took some time. And this is important, because, of course, we were not going to biopsy his brain. It’s not easy — it’s not something that you can do. If it had been somewhere else, we could have done a biopsy to try to understand what was happening. But we know we can have these sorts of flares. And when you flare in the brain, sometimes it’s not very good. But the good thing is that this patient responded totally. He could stop therapy. And now he’s off therapy for maybe 1 year and is doing very well. DR LOVE: Can you talk a little bit more about what we know about flare? What the mechanism is and what happens clinically? PROF ROBERT: In fact, we have learned that with ipilimumab. We have seen some patients with an initial increase of the size of the metastasis and even sometimes new metastasis but then secondarily a decrease of the size of the metastasis. So we thought — I mean, everybody thought, maybe we attract lymphocytes. We induce a big inflammation with the immune system that is activated. And then the immune system does the job, and then afterwards it destroys the metastasis. So it’s not very well documented, because even if you do a biopsy, you are going to find some lymphocytes and some tumor cells. But if you don’t compare, it’s not so easy to really make sure that it’s a flare before — I mean, the response is given afterwards when you see that — the metastases eventually decrease in size or disappear. So this is happening more frequently with ipilimumab than with anti-PD-1, we think. It’s maybe 10% of the cases of the responses with ipilimumab that are preceded by an increase, but it’s much less frequent with anti-PD-1. It creates a difficult situation. The patients have to be explained, you have to tell the patient, it doesn’t seem to respond right now, but we don’t know if it’s because it just does not work, which happens quite a lot. Response rate 40% it means that in more than 50% of the cases, you just do not respond. Or if you are not in a too critical situation, you can wait a little bit more, do another CT scan and really hope that you will see a response a little bit later. DR LOVE: I guess there’s really no way to know whether it’s an actual flare or just the disease continuing to get worse. PROF ROBERT: No. Just the time will tell you that. DR LOVE: So I’m kind of curious. The one other place that I’ve heard stories of cases — it seems like may be more common — where you see people progressing before they get better is renal cell cancer. Do you think that’s true? PROF ROBERT: I’m not really an expert. I’m not an expert in renal cell cancer at all. I heard that also. And that puts a lot of similarity between renal and melanoma. These are the 2 cancers where we were fighting to try to treat with immunotherapy for a long period of time. So maybe this same kind of phenomenon can happen. DR LOVE: Yes, I’ve heard people say that. Efficacy and tolerability of BRAF/MEK inhibitor combinations DR LOVE: Let’s talk a little bit about the use of BRAF/MEK combinations. And you alluded to the fact that now there’s a new one out there, we now have 3. So can you talk a little bit about what we know about the 3 options and at this point how you sort through them from the point of view clinically? PROF ROBERT: We have, for example, in France, the third combination is not yet available. But let’s imagine it is. I think that most of the prescribers will maybe initially stick to their habits and continue to prescribe maybe dabraf/treme or cobi/vemu. I mean, in France, more people prescribe dabrafenib and trametinib because of the photosensitivity. But very soon, I think as soon as we will have adverse events, fever, we will try to change. And who knows if we find that finally it’s simpler to use the third combination? Maybe the habits are going to change. That we’ll see over the years. Also what I think — and that also will depend on the countries — is the price. Maybe one combination will be less expensive than another. So it depends on the public health system. DR LOVE: From your point of view, using indirect comparisons at this point, how do you compare the efficacy and tolerability of these 3 regimens? PROF ROBERT: Yes. In term of efficacy, we have the median overall survival is longer with the third combination, binimetinib/encorafenib. It’s 33 months. So it’s the longest that we have with the combination, a little bit longer than dabraf/treme and cobi/vemu, which were around 2 years, a little bit more. I’m not sure we can really say today, because we don’t have head-to-head comparisons that one is really more effective than the other. If you look really into the group of patients, you see that you have less patients with high LDH levels in the COLUMBUS trial of binimetinib/encorafenib. So maybe less patients with a bad prognosis. So we are not sure we have exactly the same populations of patients. So maybe, I think, these are 3 very effective combinations. Now, I think, the difference is going to be played on the tolerance profile, honestly. DR LOVE: And how do you assess the tolerance profile of the three? PROF ROBERT: So as we said, fever/chills with dabraf/treme. If your patient complains about being uncomfortable because of the fever, we are going to try and give another combination and probably binimetinib/encorafenib. And then if you have cobi/vemu and the patient who develop a strong photosensitivity, you’ll try another combination. I think it will take some time, and then maybe doctors will take some habits, and maybe some of them will give first line one combination. But I think that will come after using the combination and seeing what happens to the patients. DR LOVE: You were part of a paper at ASCO, a poster, that looked at adverse events in the Phase III COLUMBUS study that looked at encorafenib and binimetinib. What did you report there? It looked like there was less fever than what you see with dabrafenib/trametinib. PROF ROBERT: Of course. Yes, yes. We have all the class effects of anti-BRAF or anti-MEK and the combination, so that is common to the 3 combinations. Like, you have some skin adverse events, some fatigue, some diarrhea. But as we said, dabraf/treme, what is specific is a fever, chills. Cobi/vemu is the photosensitivity. And bini/enco we don’t really have something special popping out. So it’s possible that it’s the best ratio between the adverse events and the benefit. It’s possible. Recent advances in the management of melanoma with metastases to the brain DR LOVE: So I noticed when looking at some of your work that you had a number of publications related to brain mets with melanoma. And you had an education presentation at the last ASCO meeting, “New Era in the Management of Melanoma Brain Metastases.” Can you kind of provide an update in terms of what’s new in terms of melanoma and brain mets? PROF ROBERT: What’s new is that the drugs that work in the metastatic setting outside of the brain — because the clinical trial initially did not allow patients with brain mets. What’s new is that we know that combination of dabraf/treme, even cobi/vemu, but it came later we know that it can act in the brain, but we were disappointed, because the response rate is lower than in the cutaneous melanoma metastasis elsewhere, and also the duration of response is less durable. It’s 6 months in median PFS instead of 1 year. So we were a little bit disappointed. And the good thing is that the combination of ipilimumab plus nivolumab at the ASCO meeting 2 years ago gave some interesting results. With 2 presentations, one with a very small group of patients, so I think it’s mostly the other one, the US team with the combination. It’s Tawbi, the first author, who showed a response rate in the brain of about — I mean, more than 50% in the brain. So that was quite spectacular. And the Australian trial that was presented by Georgina Long, very few patients but still the same ratio. So it was comforting to see these 2 trials, meaning that ipi/nivo can give rise to good responses in the brain. But what we said when we did this education session with my colleague Tawbi, we need to have trial that incorporates the stereotactic radiotherapy, because this can be a big help. So we need to build trials where we evaluate combination of systemic treatment plus stereotactic radiosurgery. This remains a very critical situation. And patients with brain mets have a very pejorative prognosis, unfortunately, especially if they have a lot of brain mets. But you also have some patients who survive a long time with brain mets. And I saw in my clinic 2 days ago a man, now he’s 67. He was diagnosed 4 years ago with a metastatic melanoma, multiple lung and brain metastasis. And he had ipilimumab initially and then anti-PD-1. He’s NRAS mutant, so could not have BRAF/MEK treatment. And he responded in a disassociated way to anti-PD-1, but he responded, and some metastasis. The other ones were either surgically removed — he had 2 surgeries of the brain, 2 surgeries of the lung metastasis. And he had a lot of radio stereotactic sessions, more than 10. Just to show you that we have really changed our way of treating the patients, because now our radiotherapist colleague, they know that we have good drugs. So they really changed their standard. Before that, they were telling us only 3 metastasis, only 3 centimeters. Now, I mean, this patient had more than 10 radio stereotactic. And he’s in complete response. It’s more than 4 years ago. This man, undoubtedly, would have been dead if it was before these treatments and this combination of treatments. So things have changed. But it’s still much more difficult, and we need to optimize. Clinical experience with hedgehog inhibitors for BCC DR LOVE: So for the remainder of our time discussing things, I thought we could move beyond melanoma and get your take on some of the new developments in other dermatologic cancers, in some cases maybe not that new. But I’m curious starting out with basal cell cancer. At least in the United States, there are now 2 Hedgehog inhibitors approved. Can you talk about your clinical experience with these types of agents? PROF ROBERT: So Hedgehog inhibitor, they really help us. They help us in some difficult situations. It’s hard to believe, but we all know that sometimes we see patients arriving with huge basal cell carcinomas. You think, but how can this person have waited so long? But it happens. It happens, because the psychology’s a big mystery, and some people just are afraid to see the doctors or consider that if they don’t have pain, it’s okay. And you see not stupid people. You see people very intelligent coming with huge tumors. And sometimes the surgery would be even sometimes technically feasible, but it would really be very disfiguring. So you have these drugs that are very effective. The adverse events are there, but they are not very strong. They are chronic adverse events. So sometimes it’s a little bit difficult at the end, but it really happens a lot. And I have a lot of patients who really beneficiate from that. And we now begin to use these drugs in the neoadjuvant setting, meaning that we give sonic hedgehog inhibitor, and then we do surgery a second time, and we can have very good outcome. But I would say the situation where it helps me a lot also are these patients with Gorlin syndrome. These patients, they have a constitutional mutation of PTCH gene or any gene on the pathway, on this sonic hedgehog pathway, and they are full of basal cell carcinomas. Full of them. So they are not metastatic, because, as we know, these cancers very rarely give rise to metastasis. But they are full on their skin. They are full of basal cell carcinomas. And if they take the drug, it just stops. I mean, they don’t have any one occurring on their skin for the period they take the drug. It works very well. But now what I do with my experience is that because these patients sometimes are tired of taking the drug every day, and it’s very different, the adverse events. It’s not very severe but chronic like. They have this dysgeusia, meaning that they don’t really feel the same taste of what they eat. And that, I mean, it’s a pain after some time not to be able to be very happy with what you eat, especially because these patients sometimes are old. So, I mean, maybe they don’t have a lot of distractions, and they like to eat good things, and they begin to eat less and less and to lose weight, and they have cramps also. So cramps doesn’t seem to be very important, but if you have cramps every night in your bed, these kinds of adverse events can be really difficult after some months. So what we do is that we try to find a regimen that is acceptable for the patient. And what I do now currently with my patients is, I give the drug for 6 months, and then we stop for whatever time. And we try to evaluate the time where the cancer comes back for the Gorlin syndrome. And now for some patients, we know. We know that we can stop for 2 months, and then they are in good shape. I say, “Okay, let’s take the drug again.” So this is not something that has been validated by clinical trials, but that’s something that we do in practice, clinical practice. Activity and side-effect profiles of sonidegib and vismodegib DR LOVE: Any difference between the Hedgehog inhibitors in terms of either their efficacy or tolerability? PROF ROBERT: So we have much less experience in France with sonidegib. We use vismodegib. And from what we read, I’m not sure there are a lot of differences. For example, there is something in the trial with sonidegib. They looked at the CPK enzymes. So it seemed that the CPK enzymes are elevated in a high fraction of the patients. But in the vismodegib trial, we just did not look at that in the blood. So it doesn’t mean that it does not go up. It does go up also. And we have some differences in term of efficacy, but here again, it has not been tested head to head. And in Europe, we have much less experience with sonidegib. So from what I see, I think these drugs are both very effective, and they have the same kinds of adverse events. DR LOVE: So you mentioned the muscle cramps and the dysgeusia. Other than dose reduction or stopping the treatment, are there any complementary strategies that work? Do you advise the patients to eat certain types of food, any medications, anything that helps with these problems? PROF ROBERT: Yes, we can help. We can help with work with dieticians, and they can try to help the patient to find the food that they like and that is not destroyed by the drug. Also, we advise the patient to do some exercise but not too violent, and they have some physical therapy. And that can really help. That can really help, symptomatic treatment. DR LOVE: You mentioned the genetic Gorlin syndrome, and I was trying to remember back to medical school what Gorlin syndrome is. What else do you see besides basal cell? Is that the only manifestation? PROF ROBERT: Yes, some of the patients also might have some meningioma. So we do a CT scan of the brain. And they also have cystic lesions in the bone. They might have some bone dystrophy. And besides that, they are doing well. Activity of the PD-1 antibody cemiplimab in metastatic SCC of the skin DR LOVE: So what do we know about checkpoint inhibitors in squamous cell cancers? Who gets these tumors? And what is cemiplimab? PROF ROBERT: Okay, so cemiplimab is an anti-PD-1 antibody. It has been tested in these squamous cell carcinomas of the skin, which are actually very frequent. It’s the second in frequency, in incidence, after basal cell carcinoma. It’s due to the sun exposure, so it’s mostly sun-exposed areas that are involved. It also can be induced by some viruses, like papilloma virus, especially in the genital areas. And these cancers, I can tell you, they are very frequent. Most of the time, they are small. We can just excise them surgically, and that’s fine. But in some patients, especially now — I think if this recording can help you to diffuse the information that men, when they lose their hair, they have to really protect their scalp from the sun. That would be a very good point. Because I see every day patients with cancer on their scalp because of the sun exposure. And if you take the sun every day, even if it’s not very, very strong sun exposure, they will get some keratosis, and eventually they will get squamous cell carcinoma. So we have a lot of patients — and sometimes old patients — with squamous cell carcinoma of the head and neck and lymph node metastasis. Because it’s not like basal cell carcinoma, which very rarely gives rise to metastasis. Here, in squamous cell, we can have metastasis when the tumors are aggressive. DR LOVE: Can I clarify, though, when you talk about the typical patient who gets a sun-induced squamous cell, are those the people who end up with metastatic disease, or is it, like, more people with immune deficiencies, transplant, et cetera? PROF ROBERT: You also have that. You have patients just because of the sun exposure who will come with a squamous cell carcinoma and sometimes becoming metastatic. But I was going to this point, the immunocompromised patients, for example, transplanted patients, they are at high risk. And in these patients, it can become very aggressive, and it’s difficult. It’s very difficult, because in this patient we are a little bit afraid to use anti-PD-1 to stimulate a lot the immune system of a patient who is transplanted — it’s another problem. But most of the patients have no immunosuppression. I mean, not obvious immunosuppression. And they have these tumors that are sometimes very big, sometimes very aggressive, sometimes metastatic, regionally metastatic or distant metastasis. And in this case, we have these clinical trials that have been published recently. It was not randomized, and it was single arm, so it’s a very impressive response rate of this squamous cell carcinoma and close to 60%. And I could see that in some of my patients very good efficacy of this anti-PD-1. DR LOVE: Yes, I’ve seen some amazing pictures of before and after where you have these huge tumors that just go away. Response to immune checkpoint inhibitors in patients with SCC of the skin DR LOVE: What do we know about the biology? Again, PD-L1 levels, tumor mutation burden, whatever? It sounds like they’re, like, fitting into the category of, like, MSI or Hodgkin or something, more responsive. Is that the case? PROF ROBERT: Yes, exactly. But they are usually not MSI, but they have a lot of mutations because of the sun also, probably. So it’s probably for this reason that they respond so well. And it’s true that it’s amazing. And it’s very good, because this anti-PD-1 — what did we do before with these patients? We gave them chemotherapy — 5-FU, platinum-based, sometimes we gave anti-EGF receptor. But first of all, it did not work that well. But plus, giving chemotherapy to old patients, not so easy. And we are much more comfortable with anti-PD-1, so it has really 2 advantages — more efficacy, less toxicity. It’s a great, great advance. DR LOVE: So you mentioned that those patients have had transplant. I was going to ask you before when we were talking about autoimmune diseases and checkpoint inhibitors what your experience is and what we know from the literature about patients who’ve had various kinds of transplants. Solid organ transplants, bone marrow, allo and the various types of immune suppression that they might receiving. What about checkpoint inhibitors within that type of situation? PROF ROBERT: So it’s always a very difficult decision to take if you have a renal transplant. You know that if the graft is injected, you are going to go back to dialysis, but you are not going to die. If you have a heart transplant, if you reject your heart, you are going to die. So it’s difficult. It’s a very different situation. I personally would not do this kind of treatment in a person with a heart and lung transplant. I would be too afraid. And I heard of some patients receiving these drugs in this situation, and I heard of some deaths, but it’s very isolated cases. In my hospital, we treated some patients with renal transplant, and it’s few patients, so I cannot give you a percentage, but I can tell you that some of them lost their transplant and some did not. But I’m not sure exactly of the percentage. Overview of Merkel cell carcinoma DR LOVE: So let’s finish out talking about Merkel cell carcinoma. Maybe first you can kind of just provide a little brief review of what Merkel cell is and what’s new in terms of systemic therapy, again, particularly the issue of checkpoint inhibitors. PROF ROBERT: So Merkel cell carcinoma is a very rare disease, very rare. It’s orphan cancer. It looks like nothing like a nodule, a red nodule. It looks like spider bites, for example. But then it does not disappear. It grows, and it can be very aggressive. In fact, it looks like melanoma in the outcome. It can give rise to lymph node metastasis and then to any kind of location — any kind of organs can be touched. Less frequently the brain, in fact, than melanoma. But it can go almost everywhere. It’s rapidly fatal, and chemotherapy works but works for 2 months — so saying it does not work. And it touches people quite old. The median age is 10 years more than melanoma and most of the cancers. So we have old patients, so fragile patients. Sometimes resolve with immunosuppression, but not always. And in 80% of the cases, you have the presence of a virus. It was found not such a long time ago. This virus, which is from the family of the polyomavirus, a lot of us have this virus, but it does not do any harm, except in some patients it’s a defective virus. It’s integrated in the genome, and it gives rise to this Merkel cell carcinoma. And so what did we do before anti-PD-1 was tested? In case of metastatic disease, we gave chemotherapy with platin and etoposide. So about 30% of response, but the first CT scan or the second, they were already relapsing, so most of them. Response to PD-1/PD-L1 blockade in patients with Merkel cell carcinoma PROF ROBERT: And now we have anti-PD-L1 and anti-PD-1, which have been tried. Avelumab is an anti-PD-L1, and that gave a very good response rate. It has been tested in second line first, about 30% of response. But then in first line, it seems to be much higher. And it’s just been authorized in France, and it has been authorized in most of the countries. And we also have some very good results with anti-PD-1, the one that we know, pembro and nivo. And also, very interestingly, it was evaluated in the neoadjuvant setting. Neoadjuvant, it means that we give the drug before doing the surgery in a limited disease. Like, the tumor is resectable, but before doing the resection, we give the drug. And it was very, very promising, these first results. DR LOVE: Any patients you’ve taken care of with Merkel cell who’ve had good responses to checkpoint inhibitors? PROF ROBERT: Ah, yes. I remember, I showed a patient to my colleagues recently. So this patient was a little atypical, because he survived several lines of chemotherapy, which is rare. So he had 2 or 3 lines of chemotherapy, but he was relapsing. So we did surgery several times. We resected the metastasis. Maybe it doesn’t seem very logical, but if you don’t have any other solutions, sometimes you do that — doing a surgery for Stage IV disease if you don’t have to or if you have only 1 or 2 metastases. And then we could enroll the patient in the avelumab trial, and he received the drug. Complete response. We stopped the treatment after, I think, 15 months, and now it’s more than 2 years ago, and he’s free of relapse. This is wonderful. This was not seen before. In my center, which is a big center in France, we have about 15 patients with Merkel cell carcinoma. So it doesn’t seem a lot, but, in fact, it’s a lot compared to the total number, because it’s less than 100 in total in France. But half of them are metastatic. And we have some data with anti-PD-L1, and we have — I mean, with our small group of patients, we have close to 40% of response, and we have 2 patients who could stop the drug. So we look at that as a miracle. DR LOVE: Yes, you have a patient who got avelumab, which I guess is approved, at least in the United States, for Merkel cell. And this patient had a complete response, has now been off therapy a couple of years. PROF ROBERT: Exactly. DR LOVE: Where was the metastatic disease that patient had? PROF ROBERT: Oh, he had a lot of internal lymph nodes and liver. I mean, he had real Stage IV disease. |