Oncology Today with Dr Neil Love: Melanoma Edition (Video Program)
Oncology Today with Dr Neil Love: Melanoma Edition
Featuring a roundtable discussion with Drs Jason J Luke, Michael A Postow and Ryan J Sullivan.
3.5 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.
Evolving adjuvant treatment approaches for patients with localized melanoma DR SULLIVAN: Now I’ll talk a bit about adjuvant therapy for melanoma, where we stand in 2019. In 2018, 3 really important things happened. The first thing that happened is a new staging system came into effect. The AJCC 8th edition was adopted, and so everything that had preceded it in terms of every clinical trial launched prior to 2018 was with the AJCC 7th edition. We, virtually overnight, in June 2017, went from favoring completion lymph node dissections to not favoring completion lymph node dissections with the publication in the New England Journal of Mark Faries’ MSLT-II study. And then data from recent and past clinical trials that enrolled patients, again, with the older staging system, and who had completion lymph node dissections as part of the eligibility for those trials were reported and showed really promising data. I think it’s important to say that all of our interpretation of the data, though valid, is still a little unstable in terms of, how do we factor in what happened on those studies and how do we apply it to now. Here’s the landscape. In the ‘90s high-dose interferon alfa-2b was approved based on 3 cooperative group trials. A pegylated version of interferon alfa-2b was FDA approved in 2011 based on presented and published data mostly in Europe. And then ipilimumab came next, then nivolumab, and then dabrafenib/trametinib and pembrolizumab. The timeline has been an adoption of the therapies that were effective in the metastatic setting shifted about 2 or 3 years to the right for the adjuvant setting. Why don’t we talk about the data of adjuvant nivolumab. This is the CheckMate 238 study. This randomized patients in a 1:1 fashion to either nivolumab 3 mg/kg and an ipilimumab placebo versus ipilimumab 10 mg/kg and a nivo placebo, and treating patients for 1 year of therapy. And the data is pretty clear that nivolumab is associated with improved relapse-free survival compared to ipilimumab, and that’s the primary endpoint. And the obviously predicted endpoint, as well, was toxicity and that nivolumab was less toxic than ipilimumab, and in those red boxes are showing the treatment-related AE percentage, Grade III to IV, 14% versus 46% with adjuvant ipilimumab. And 4% treatment discontinuation rate versus 30% rate with ipilimumab. In addition, the toxicities were mostly reduced, again, across all the spectrum of immune-related toxicities, except for the endocrine toxicities. Again, a 20% rate of thyroid dysfunction, a 2% rate of pituitary disorder. The thyroid dysfunction’s a little bit more common with nivolumab. Pituitary dysfunction’s less common. And then rarely we see diabetes. To summarize the data with nivolumab, and this was the data from the 18-month cutpoint, but it’s now with additional follow-up I would say it stands. And so there’s a lower rate of relapse compared with nivolumab in the adjuvant setting versus ipilimumab. We don’t have data about whether nivolumab, as of yet, improves overall survival compared to ipi, but we’re hopeful that we’ll have that data at some point in the near future. And what is clear is nivolumab is better tolerated than ipi. And based on those 3 bullets, nivolumab clearly deserved to be FDA approved for this patient population. Important to note, 15% to 20% of patients develop Grade III or IV toxicity. About 10% of patients stop early due to side effects. And about 20% of patients develop some permanent side effect, whether that’s thyroid dysfunction, pituitary dysfunction, diabetes, vitiligo, et cetera. There were no deaths on this study, but based on all the data we have with single-agent PD-1 inhibitors, we would expect that there would be a fatality rate somewhere between 0.1% and 1.0%. What about the data with adjuvant pembrolizumab? The data supporting the approval of adjuvant pembrolizumab came from this study. This is a randomized Phase III trial that randomized patients to either pembrolizumab or placebo. And in those who received placebo and at time of relapse are offered pembrolizumab. This is the first part of this study, is pembrolizumab up-front associated with a better, or lower, relapse-free survival. And the second part of this study will be, is pembrolizumab now better than pembrolizumab later, in terms of treating patients who ultimately recur. What’s clear is that there’s an improvement in relapsed-free survival in patients who receive pembrolizumab versus those who receive placebo with a hazard ratio of 0.57. And from a toxicity standpoint, very similar data that was seen with nivolumab. We end up seeing, again, about a 20% rate of thyroid dysfunction, rare severe toxicities like the neurologic toxicities, and hepatitis and pneumonitis, et cetera. And to summarize here, quite similarly to nivolumab, except this study obviously compares pembrolizumab to placebo, but we see a significant improvement in relapse-free survival. It’s consistent across all subgroups. BRAF positive versus BRAF negative seemed to be a pretty consistent hazard ratio. PD-L1 positive versus PD-L1 negative, pretty consistent in the stages, Stage A, B, and C — Stage IIIA, B, and C. It was well tolerated, but again, just like nivolumab, about 15% of patients, a little bit higher in this data set, had to stop — or developed Grade III or greater AEs. About 14% stopped due to those AEs, early, and about 20% of patients developed permanent side effects. And then in this study, there were 2 treatment-related deaths. And to summarize adjuvant immunotherapy effectiveness, interferon is probably better than observation, although that’s debatable and certainly not worth discussing much here. Ipilimumab is definitely better than placebo in terms of relapse-free survival and overall survival advantage. Nivolumab is better than ipilimumab based on the CheckMate 238 study. And pembrolizumab is better than placebo. It’s perhaps not fair to have the blue rectangle shifted off because we would imagine that pembrolizumab would be equivalent to nivolumab. There are a few other studies that either have just been reported or will be reported soon. One of those is the ECOG-E1609, which randomized patients to ipilimumab or — at 3 mg/kg, ipilimumab 10 mg, and pegylated interferon. And basically low-dose ipilimumab is better than interferon. High-dose ipilimumab may be better than interferon. And high-dose ipilimumab is more toxic than ipilimumab. The SWOG-S1404 study has not been reported yet, but is pembrolizumab versus these. And then there’s another study comparing ipilimumab and nivolumab versus nivolumab in the adjuvant setting. And we’re waiting on that data. In terms of summarizing the immunotherapy toxicity, interferon is more toxic than placebo. Ipilimumab is way more toxic than placebo. Ipilimumab’s more toxic than nivolumab. And pembrolizumab is more toxic than placebo. But when you think about it from a permanent standpoint, interferon — occasionally patients develop vitiligo or have some joint stuff and thyroid dysfunction, but it’s not that common. Ipilimumab and pembrolizumab are definitely associated with more permanent side effects than placebo. And quite honestly, nivolumab and ipilimumab are associated with similar rates of permanent side effects. Moving on to BRAF-targeted therapy, the only approved BRAF-targeted therapy in the adjuvant setting is dabrafenib and trametinib, but there is some interesting data that I’ll just touch upon briefly with single-agent BRAF inhibitor therapy. So the BRIM8 study randomized patients to vemurafenib or placebo. And there were 2 cohorts. The first cohort was Stage IIC, IIIA, and IIIB. The second cohort was Stage IIIC. And the way this trial was designed is basically that the second cohort needed to be positive in order to trigger analysis of the entire study. I don’t understand why it was designed that way, but it was. And in the cohort of the Stage IIIC, this is the cohort that needed to be positive in order to study. What was seen is out of the gates, after the 1 year of vemurafenib, there was a clear separation of relapse-free survival. And after 1 year of therapy, when patients were discontinued, there was an immediate collapse of the relapse-free survival curve. So they came back together. The p-value was not significant, and because this part of the study was not significant, the trial failed to meet the criterion for analysis of the entire study. So negative trial. However, when you look at the other cohort, something interesting happens. There’s very few patients progress when on vemurafenib, and when vemurafenib is discontinued, those curves stay apart. There’s something still here about this single-agent BRAF inhibitor for these patients with lower-risk disease that’s associated with probable durable benefit in some patients treated with single agent. And it’s probably the patients with the least amount of disease. The summary here, randomized Phase III trial failed to meet its endpoint because it was designed weird. But the encouraging data in the earlier stages, Stage IIC to IIIB, predictable reversible toxicity, about a 20% discontinuation rate, and BRAF single agents associated with paradoxical activation of the MAP kinase pathway, which can be associated with secondary skin cancers. That was definitely seen, and about 20% of patients developed either keratoacanthoma, squamous cell carcinoma, or basal cell carcinoma. There is no difference in nonskin cancer malignancies, meaning there wasn’t a higher rate of colon cancer or pancreatic cancer, things like that. Finally, the COMBI-AD study with dabrafenib and trametinib versus placebo. This trial randomized over 850 patients, and the primary endpoint was relapse-free survival, and there’s a clear separation of relapse-free survival that stays apart, unlike that data with single-agent BRAF inhibitor. At 2 years it remains apart. At 3 years it remains apart. Importantly, as opposed to the single-agent BRAF inhibitor study — this is the forest plot. At each subgroup looked at, it maintained a pretty consistent hazard ratio, about 0.35 to 0.5. And the interim analysis of the overall survival suggested that there may be an overall survival advantage. In terms of adverse events, virtually everybody has side effects, and about 25% of patients discontinue because of those side effects. But they’re typically manageable and almost always get better within a few days of holding therapy. And so the summary here is a randomized Phase III trial of COMBI-AD shows superiority of adjuvant dabrafenib/trametinib versus double placebo. It met its primary endpoint, secondary endpoint. The initial analysis looks promising. Improvement was maintained in all subgroups, with consistent hazard ratios, predictable reversible toxicity. But again, over 25% discontinuation rate and no significant difference in the rate of additional malignancies. The kind of overall summary is that we have current data now with Stage III melanoma, that anti-PD-1 antibody can be used as standard of care therapy in Stage III melanoma, independent of BRAF status. And that BRAF/MEK inhibitor combination is an absolute alternative standard of care option for the BRAF-mutant patients. There are pending trials, nivolumab versus ipilimumab and nivolumab, and there is now a Stage II pembrolizumab versus placebo study that is accruing. I think there are unanswered questions. We don’t know which patient should get BRAF-targeted therapy versus which patients should get PD-1 inhibitor therapy in the BRAF-mutated status. And we also don’t know, ultimately, which patients should receive therapy versus close observation. And we don’t know whether immunotherapy now versus immunotherapy later is better. However, we are doing studies to hopefully answer all of these. With that I’ll stop. Contemporary management of metastatic melanoma with and without BRAF V600 mutations DR LUKE: I’m going to speak on the contemporary management of patients with metastatic melanoma, both with and without BRAF V600E mutations. This is a little bit of a big area, so I want to try to hit some highlights that I think could be relevant to clinical practice. We’ve had an enormous change in the management of melanoma over the last several years. If you date back to 2011, which was really when the first of our modern approvals came into clinical practice. At that time, patients were treated with chemotherapy, and then a small number of patients could get interleukin-2, which we know was associated with a long-term benefit. Now that’s very different now in 2019, where we have targeted therapies and 3 groups of BRAF and MEK combination therapies. We also have a compendium-associated treatment with KIT-mutant melanoma, and then immunotherapies that are quite commonly used for multiple different cancer types, and even approval of a viral therapy. And all of this can be somewhat overwhelming, but the really great thing for patients is that when we look at the outcomes of patients as they were described by AJCC 7 back in 2011, we can see that the 2-year survival for most of these patients was about 30% or less. And based on the treatments that we have now, just from the clinical trials that we’ve done, the 5-year survival is now on the order of about 50%. And so there’s been just an enormous change in the landscape and the number of treatments that we can offer to our patients. In that regard, it’s now well known that there’s improved overall survival with the use of both BRAF-targeted therapies, as well as immuno-oncology agents. We mentioned that 3 different double BRAF/MEK combinations have been approved, the first of which being dabrafenib/trametinib, followed by vemurafenib and cobimetinib, and most recently encorafenib and binimetinib. These are highly efficacious. They are easy to take and are associated with a reversible side-effect profile. The caveat always being that the median progression-free survival tends to be on the order of about a year, and we all worry about resistance. And those are in contrast with the PD-1 CTLA-4 agents. Well aware that pembrolizumab improves survival relative to ipilimumab. And we’ve now seen improvements both for nivolumab compared with ipilimumab, as well as nivolumab plus ipilimumab. And so all of these led to big papers, changes to the standard of care, but I think really what’s become obvious in the last year, that’s really so exciting, is the data over long-term outcomes for these patients. In the last year we’ve had reports now for 5-year survival rates in metastatic melanoma, first for the BRAF and MEK combination at 34%, and more recently with the PD-1 or combinations, where you can see in the upper right-hand side, the pembrolizumab data at 38%. And the 5-year overall survival for nivolumab or nivolumab plus ipilimumab now at about 52%. And so this is really an amazing change in the standard of care over a relatively short period of time. Before anyone gets too excited about comparing all of these curves, I think it’s important to point out quickly that these are post hoc analyses. These weren’t preplanned to analyze these data this far out, and that these were not necessarily contemporary with each other. In other words, the options available to patients at the time of progression were different across all these studies. It isn’t to say that we can use these as absolute thresholds about what the best treatment is, per se, but rather to make the note that there’re many options now available for our patients. And based on these data, we would think that a 50% 5-year survival is a very reasonable number to quote on the median to patients who are diagnosed with what used to be the cancer that made cancer bad. I will note there are a lot of open research questions that still need to be answered around the most efficient use of these agents. For the BRAF wild-type patient, should you start with targeted therapy? Or should you start with immunotherapy? And there is a clinical trial, it’s available in the community, the ECOG 6134 study [EA6134], now deemed the DREAMseq trial. I won’t go through all the details, but suffice it to say that this trial tries to answer the question that everyone does in clinical practice, which is do you start with targeted therapy? Do you start with immunotherapy and patients crossover? I would advocate for people to participate in that clinical trial. There are still a number of questions that are outstanding about the field, however, and 1 of them has to do with the right dose of giving ipilimumab. We have the data for BRAF and MEK and for PD-1 monotherapy. They actually look quite similar in terms of the outcomes for the patients both at 1 year and out through the landmarks. But the dose for ipilimumab is actually a little bit of a big deal because we know that the adjuvant approval was at 10 mg/kg. However, we usually use 3 mg/kg in practice. But some people have started using even a lower dose. There was a study reported out last year, the CheckMate 511 study, which looked at this inverted dosing. Half the patients got nivolumab at 1 mg/kg plus ipi at 3, or vice versa, nivo at 3 and ipi at 1. And the take-home point here is those curves don’t look very different, and so there was no difference in the outcomes of the patients treated on this trial. Now the trial wasn’t powered for the analysis of the outcomes in terms of efficacy response. Rather, it was to look at toxicity. And as no one would be surprised, the lower dose of ipi caused less side effects. But this has made some people think, well perhaps we would be able to flip the dose of ipilimumab to reduce toxicity, and maybe we should integrate that into clinical practice. And some people have begun to do that. Now that being said, there was also a clinical trial performed called the CA184-169 study, which was a randomized Phase III trial of ipi 10 versus ipi 3 as monotherapy. And I don’t have time to go through each 1 of these numbers, but you can see that the numbers for ipi 10 across the board are better in a randomized Phase III trial, suggesting that the higher dose of ipi may, in fact, be improved. And so what does this mean? I think it’s an open question, and it really goes to the individual patient. Most of the time when we use ipilimumab in a combination regimen, it’s because we’re so worried that their clinical factors are high risk, and they might not have the opportunity to get ipilimumab later. And therefore, in my practice, I’ve continued to use the higher doses of ipilimumab, mostly because the reason I was using it in the first place was I thought that these patients were at a higher risk of progression and death. When we compare and contrast these different treatment modalities, BRAF versus immunotherapy, there are pros and cons to think through. For BRAF, this is rapidly effective in almost all patients, at least for some period of time. You can palliate symptoms very rapidly. And it’s a convenient therapy taken orally by the patient at home. The side effects are most commonly fever, although there are a number of others that you have to be cognizant of, but they’re generally reversible with discontinuation. But treatment discontinuation is not recommended. Meaning if you have the excellent treatment response, this is a lifelong therapy in the metastatic setting. It’s not the case that we would recommend, say after a year or after 2 years, stopping treatment, because our general experience in that approach is that patients will progress and have a recurrence. It’s also important to note that where we used to think of using BRAF inhibitors the most aggressively, which was in high-volume and high-risk disease, we’ve actually now learned that, as should have been expected, it actually works least well in these settings. And so for patients with high LDH, for brain metastases, for more than 3 sites of melanoma, the efficacy of the BRAF inhibitor is actually attenuated, meaning the progression-free survival is much more likely to be on the order of 6 months as opposed to a year for those populations. In contrast then, immunotherapy is an intravenous treatment, but generally speaking it can be well tolerated and administered less frequently, on the order of every 3 to 4 or maybe even to 6 weeks in the near future, and has this potential for something we’re describing now as treatment-free survival. Meaning after you are treated you can stop being on treatment and survive. And so this was a new concept in oncology, but I think is important to think about, which is to say even with people with metastatic disease we can get them to a place where they no longer have to be treated. The toxicity profile is in contrast with BRAF inhibition in that it is generally very well tolerated, but does include some severe and some rare irreversible and even potentially fatal side effects that you have to be cognizant of. And, as we discussed, there may be a disproportionate benefit of immunotherapy in BRAF-mutant or more aggressive disease. And so all of these considerations of when to choose which therapy have to be taken under advisement for an individual patient. Quickly then, I want to discuss other combinations on the horizon. We won’t go through them in any detail, but I’ll just note that an obvious combination partner then would be to give targeted therapy and immunotherapy together. And 2 randomized Phase III clinical trials are now accrued and we’re waiting for the readouts, the names COMBI-i and TRILOGY, combining standard of care agents. There’s also a Phase III study of pembrolizumab with talimogene laherparepvec, the oncolytic virotherapy, the MASTERKEY-265 trial. There is a Phase III study of nivolumab plus an anti-LAG-3 agent known as relatlimab that is also now accruing patients. And there’s a Phase III trial of nivolumab plus bempegaldesleukin, or NKTR-214, as compared with nivolumab. So a number of different treatment options that will also be coming forward in the near future, and we’ll really have to try to work together to try to figure out what the best way to give all these therapies might be. To wrap up, what about what do you do for patients after the front-line setting? Now that we’ve moved our therapies into the adjuvant space, or patients have progressed either in first-line therapy or in third line after BRAF inhibition. What do you do in that situation? There’s giving ipilimumab for patients who had PD-1 monotherapy in that setting. Should you give ipi monotherapy, or should you give the combinations, even the second line? Should you recycle BRAF and MEK inhibitors? Meaning if you had a treatment holiday, could you go back to those kinds of therapies? And what about clinical trials? And there are a few coming forward. Surrounding ipilimumab, again, with this question of dose. It’s interesting to observe that from the KEYNOTE-006 study, which was the registrational study for pembrolizumab, a certain number of patients crossed over from pembrolizumab onto ipilimumab as part of their standard therapy. And importantly from that group there was approximately a 13% overall response rate. So that’s important to keep in context. There are a number of retrospective analyses now of giving pembrolizumab — or giving ipilimumab at the lower dose plus pembrolizumab in the second line. I highlighted 1 here that I was involved in. To note that in a small number of patients we report a much higher response rate relative to ipi monotherapy, suggesting that we may consider giving the combination even in the second line despite a lack of really robust data. What about rechallenging patients with BRAF and MEK inhibition? Say, for example, a patient got adjuvant BRAF and MEK inhibitor therapy and then progressed in the front-line, got immunotherapy, if they were to progress again, could you treat them with BRAF and MEK a second time? This group from Europe published a case series of patients where that was pursued, and you can see that this waterfall plot suggests upwards of a 50% response rate on rechallenge of BRAF and MEK inhibitors if patients had had at least a 6-month holiday. Certainly similar to other chemotherapies that are used for other diseases, perhaps we can return to BRAF and MEK inhibition at a later time in the patient’s treatment course. Those are standard of care options. What about clinical trials that are existing in this space? And again, we don’t have time to get into all of them, but I’ll highlight that patients can be referred for such studies. Two registrational approaches are ongoing at this time, 1 to combine ipilimumab with the Toll-like receptor 9 agonist tilsotolimod, and that’s for injectable lesions, and another of adoptive cell transfer or tumor infiltrating lymphocytes. Both of these are trials that hope to either readout or to get FDA evaluation within the next year or so. Much more still ongoing in the realm of metastatic melanoma. To conclude, I’ll summarize by saying that BRAF/MEK, PD-1, or PD-1 plus CTLA-4 are standard therapies in the front-line. And the choice really depends on multiple factors, some of them clinical, such as LDH, the performance status, and the number of metastases. The pace of progression is important, and the patient’s choice, obviously, about route of administration and side-effect profile. The optimal dose of ipilimumab, unfortunately remains unknown. As monotherapy we clearly give 3 mg/kg, but in combination there’s now data to support giving PD-1 with a 1 mg/kg ipilimumab. In my practice, I bias towards using the 3 mg/kg dose because that’s the dose that we have the most familiarity with. These BRAF, MEK and PD-1 triplets are near. And so they will be coming in the next 1 to 2 years. We don’t have the data to look at now. However, one wonders whether or not this is truly better than giving these therapies individually. And before this comes forward as the standard option for everyone, I think we need to take a step back and make sure we’ve adequately evaluated, for each patient, what the optimal therapy might be. And a number of front-line novel combinations are in Phase III and coming forward, so please be on the lookout for those, as they may represent further treatment options for your patients. Similarly, in later lines of therapy there are ongoing clinical trials and likely to be approvals in this space in the relatively near future. And I’ll just end by saying that despite major progress in this field, with 50% of patients at 5 years on the ipi/nivo trials, there still remains a major unmet need for refractory metastatic melanoma. As much as we can support clinical trial accruals, I hope that we can continue to do so. Thank you very much. Optimal clinical care algorithms for patients with melanoma and CNS metastases DR POSTOW: I’m going to talk about a big problem with melanoma, which is unfortunately that it has a high predilection to going to the brain. And unfortunately, as we all know from our clinical practice, a lot of patients with melanoma unfortunately will pass away due to complications of neurologic involvement of their disease. But as we’ve been hearing about, a lot of these treatments that have been very effective in treating patients with systemic melanoma have also now shown promise in treating brain metastases. We’ll go through some of that data in this discussion and how we think about it in treating our patients. The first consideration, as we always think about, for patients with melanoma, is whether we’re going with immune therapy or we’re going with BRAF/MEK-directed treatment for patients with BRAF mutations. And some of the data that we have for efficacy of these newer treatments in the brain come from studies of BRAF and MEK inhibitors in patients with brain metastases, and specifically dabrafenib and trametinib, I think, has the most robust prospective data among the BRAF/MEK-inhibitor targeted treatment options for patients with BRAF-mutant metastatic melanoma involving the brain. And the combination of dabrafenib and trametinib in this Phase II study, there were a number of different cohorts that were tested in this study, but this waterfall plot really, I think, sends the message that the response rate for BRAF-mutant melanoma in the brain to dabrafenib and trametinib in the COMBI-MB study is quite high at 58%. And this is much higher than we saw with single-agent PD-1 drugs that I’ll talk about, which has a response rate of about 20%. Much higher than ipilimumab as a single agent in the brain in asymptomatic patients, on the order of 10% to 15% response rates. So very, very profound responses to BRAF/MEK inhibition for BRAF-mutant metastatic melanoma involving the brain. But the problem that we had in the brain specifically is that unfortunately the median duration of response to this BRAF-mutant metastatic melanoma in the brain was short. We see on the left curve here the median duration of response only a little bit over 6 months, and the median progression-free survival approximately also about 6 months. Although the response rate in the brain is really, really high with BRAF/MEK inhibition. Unfortunately, those responses don’t last very long. And it still is an interesting question, well, how do these patients progress? Do they progress because the existing lesion shrunk and then they grew again? Or is it that these existing lesions shrunk and that new lesions developed? And I think that’s important data that hopefully will be forthcoming so that we can really tease apart what the role of these BRAF/MEK inhibitors are in treating the microscopic metastatic disease in the brain that we can’t see on scans. But I wanted to start with this example of BRAF/MEK inhibition in the brain because 1 thing that we’ve learned recently is that extracranial activity of these BRAF/MEK inhibitor treatments, now that we have 5-year data with dabrafenib and trametinib, can be very, very prolonged in certain subgroups of patients. But patients with brain metastases still remain to be an unmet need in need of better treatment options with BRAF/MEK inhibitors. And into that landscape we have now data to support combinations of immune checkpoint blocking antibodies from 2 prospective studies. Both of these are prospective Phase II studies. One is the ABC brain metastases study, this is a group of investigators led from Australia, that looked at patients with brain metastases that were less than 3 centimeters in size and at least 5 millimeters in size who had previously not had treatment with checkpoint inhibitor drugs. Prior treatment with BRAF/MEK inhibition was allowed in this study. And the other interesting eligibility criteria are included on this slide. And these patients were randomized 1:1 to the combination of nivolumab and ipilimumab at the current FDA-approved dose of ipi 3 mg/kg and nivolumab 1 mg/kg. And then the other randomized arm was nivolumab monotherapy. Keep in mind, these are relatively small groups of patients, 35 in 1 and 25 in the other group. And then there was another cohort that I won’t talk about too much in the context of this study, but this was a group of patients with symptomatic brain metastases, or leptomeningeal disease, a small group of patients treated with nivolumab monotherapy. But I think the key to take away with the ipi/nivo data, at least from the ABC brain mets study, was the result in the asymptomatic patients, and particularly comparing single-agent checkpoint inhibition with PD-1 alone, that being nivolumab, versus combination ipi/nivo in this study, with the primary endpoint of intracranial response rate. In this study, when you look at these different cohorts in the nivolumab in the red box. That was the group of symptomatic patients, so I’d really caution any comparisons with that compared to the other 2 groups. But what we saw was a very, very high response rate intracranially with ipi/nivo, 51% among these groups of patients, which is essentially approximate to the extracranial response rate in the brain. And so that tells you that the brain seems to be just like any other organ, you can get responses with ipi/nivo like you can get in the lung, the liver and other lesions, and that although it’s a different anatomic area that has different treatment considerations, this systemic treatment can get into the brain and can treat these patients just as well as it can in extracranial sites, at least these selected patients that were asymptomatic, not on steroids, in this study. And that really contrasted to the nivolumab efficacy that was seen. When we look at the intracranial progression-free survival, the gold curve here shows the ipi/nivo group. One can really see that there is a very strong tail on this curve of approximately 50% of patients with metastatic melanoma involving the CNS that have treatment with ipi/nivo and have long-term intracranial progression-free survival. Now unfortunately most of these progression events do happen within that first 6 months of treatment. Trying to monitor patients closely, at least during that first 6 months, is really imperative. But if patients make it for about 6 months without progression, they seem to enjoy this long-term progression-free survival with ipi/nivo, in the brain. Very effective treatment for this group of patients, and certainly better than nivolumab monotherapy. Overall survival, also, is very impressive in this study. Both PD-1-containing arms in this study. But one can know that now metastatic melanoma involving the brain, in this highly selective group of patients in this clinical trial, is not necessarily a death sentence, in that we have about half of the patients that are living long term with brain metastases, and this is follow-up. With a couple years follow-up, of course, the tail of the curve gets a little less stable at around 3 years, but we see pretty good long-term follow-up at least through 2 years of treatment. Into that context, we have another randomized — actually this is a single-arm study of ipi/nivo for patients with asymptomatic brain metastases and also a cohort of patients now with symptomatic brain metastases that were treated with the nivolumab/ipilimumab combination. And I’ll go through the asymptomatic group of patients first, because now we have a decent amount of follow-up for those patients, almost 2 years. And then we’ll review the symptomatic cohort, which are the first data that have come forward about how well does ipi/nivo work for brain metastases in patients that have neurologic symptoms or patients that require steroids. And the steroid dose was limited to a maximum of 4-mg equivalent of dexamethasone per day. So not higher doses of steroids than that, but some doses of steroids prior to treatment. Again, just like the ABC brain metastases study, this used the standard FDA-approved dose of nivolumab and ipilimumab, 3 mg/kg of ipi and 1 mg/kg of nivolumab. And the primary endpoints are all indicated on the slide here. The best objective overall response rate is, as shown, 54% intracranial responses with ipi plus nivo, which is very, very impressive and very similar to the 49% objective response rate with extracranial ipi/nivo use. Again, very similar results from the ABC brain mets trial. Great concordance between responders in the brain and responders outside of the brain. The notion of the blood-brain barrier is not the classical notion. We know that immune cells can get into these tumors and destroy them. When we follow these asymptomatic patients now with almost 2 years of median follow-up, one can again see this tail of the curve emerging with ipi/nivo treatment, both intracranially, as well as extracranially. And that the median time of follow-up here in terms of progression-free survival has not been reached, so those bars remain over the 50% line for progression-free survival. And we hope with even additional follow-up beyond this that that progression-free survival will remain not reached over time. We do need longer follow up. The overall survival, similar to the ABC brain mets trial, very, very favorable. This was a selected group of patients with brain mets that could meet the trial eligibility criteria, yet only a quarter of the patients had passed away during the first 18 months of this trial and that we see about three quarter of patients with melanoma with brain metastases are alive a year and a half into the treatment course. And so again, that’s very, very helpful for patients that have a new diagnosis of brain metastases, saying that if you met this trial eligibility and you get ipi/nivo, three quarters of a chance that you’ll be alive at least a year and a half into treatment. And it looks like the tail thereafter remains quite promising, as well. Now just a brief word to conclude about symptomatic patients in this trial. Unfortunately, the number of patients with symptoms in this trial, or requiring steroids, was generally lower in terms of number. There were only about 18 patients that were evaluable in this treatment group. And the patient characteristics are indicated on the slide, so I don’t want to go through that in detail other than to say there were a number of patients that had 3 or more lesions in this group of treatment, and that the median sum of intracranial target lesions, meaning how much burden of disease that these patients actually had, was 2.6 cm. A relatively higher burden of CNS disease. There were a little over half of patients that needed steroids at the baseline to treat symptoms related to neurologic disease. And when we look at the response efficacy in this group of patients, again small number, so it’s very big confidence intervals and really hard to be absolutely certain what the real response rate will be in this patient population. I think what we can say is that there are anecdotal responses that are achieved. There were 4 patients that responded in this cohort, with an overall response rate of 22%. Again, that big confidence interval from 6% to 48%, saying we really don’t know what the real likelihood of response is here. But the fact that ipi/nivo can work, at least in some patients with symptomatic disease, so is something that one might consider in lieu of whole-brain radiotherapy, for example, on the candidates that are not eligible for stereotactic radiosurgery, that don’t have BRAF mutations. But again, this is still a reminder that in patients with symptoms or requiring steroids, we need better treatment options in general, because 22% is certainly not high enough for an objective response rate. Among the 4 patients that responded, there were responders with BRAF-mutant disease and BRAF-wildtype disease, so like we’ve seen with extracranial melanoma responses, it’s not so clear that BRAF mutational status really matters in terms of likelihood of responding to ipi/nivo. And the indications for the kind of number of doses and treatment the patients got is shown on the slide. There was only 1 patient that had steroid use at baseline that responded in this trial. It still remains a question if you have a patient that’s on steroids to treat symptomatic CNS disease, how can you safely get that patient off of steroid before you start ipi/nivo to try to maximize response? That’s an open question, but I do encourage to try to taper off steroids if they can prior to start of ipi/nivo, unless they absolutely are needed for symptomatic management. And the progression-free survival in the symptomatic group, unfortunately not as high as we had seen with the asymptomatic group, only about a third of patients with longer-term progression-free survival. Whether that’s related to worse disease at the outset that required steroids or caused symptoms or something having to do with the steroid use in this cohort really remains unknown, but hopefully something that we can sort out as the numbers of patients are treated farther out. And again, overall survival less in symptomatic patients, as one would expect. Given this, with the summary that we have, we have data for all kinds of agents in the treatment of brain metastases from melanoma, but I think the most important ones that we focused this portion of the talk on, BRAF/MEK inhibitors and combinations of immune checkpoint blocking antibodies, are really the main way that we’re moving forward and treating these patients with systemic therapy. Immediate responses with BRAF/MEK, but unfortunately short durations of response. If you’re going with immune therapy, if a patient’s a good candidate for ipi/nivo, I would suggest that. Follow-up for asymptomatic patients looks quite good in that combination group. If the patients are on steroids and they can’t get off treatment, and they have a bad tumor that you really need to shrink right away, I would still say that BRAF/MEK inhibition does have a role in immediate palliation of those patients, hopefully to get them off of steroids and then maybe switch to ipi/nivo. And we’re still trying to figure out the appropriate sequencing of giving stereotactic radiosurgery and the use of surgery in the context of the improved systemic therapies that we’ve seen in this patient population. And hopefully future trials will try to tease out some of those questions. With that, I’d like to thank everyone, and I appreciate your discussion. |