Dissecting Adjuvant Therapy Options in Melanoma

January 4, 2019

Cancer Network speaks with Dr. Ryan J. Sullivan about adjuvant therapies in development for melanoma patients.

Today we are discussing adjuvant therapies in development for patients with melanoma with Ryan J. Sullivan, MD, a medical oncologist at the Massachusetts General Hospital, Boston, who specializes in the care of patients with advanced melanoma and leads clinical trials.


-Interviewed by Anna Azvolinsky


Cancer Network: Just recently, there have been approvals of both an immune checkpoint inhibitor as well as BRAF-targeted oral agents for melanoma patients. What are the options for adjuvant therapy for patients diagnosed with melanoma?

Dr. Sullivan: Thank you for inviting and allowing me to discuss treatment options for patients with melanoma. You’re absolutely correct, there have been two approvals in the past year in the adjuvant setting for patients with stage III melanoma. The first is for nivolumab, an anti-PD-1 antibody which was approved based on data in the Checkmate 238 study which compared patients who received nivolumab vs patients who received ipilimumab.

All of the patients had Stage IIIB, IIIC, or stage IV melanoma that had been resected and were randomized on a one to one basis to receive ipilimumab, which was the control arm, or nivolumab, which was the experimental arm. The data showed dramatic improvement in relapse free-survival in patients who were treated with nivolumab.

In addition, patients who were treated with nivolumab had less severe toxicity than those who were treated with ipilimumab. On the basis of that data, nivolumab received approval from the FDA for patients with stage III melanoma that had been resected. The other approval was for the combination of two agents, dabrafenib, a BRAF inhibitor, and trametinib, a MEK inhibitor, and this combination had previously been approved for treatment of metastatic V600K or V600E mutant melanoma.

On the basis of the COMBI-AD trial, which was a study of these two drugs, dabrafenib plus trametinib vs a double placebo, the combination demonstrated significant improvement in relapse-free survival, distant metastasis-free survival and overall survival and was FDA approved in early 2018 for patients with stage III melanoma.

A few interesting points about adjuvant therapy for melanoma is that we have options. I think one option is nivolumab, another is dabrafenib plus trametinib and I think the prior options, either ipilimumab or adjuvant interferon or observation are less likely going to be implemented for patients. I think it is pretty safe to say that ipilimumab and interfernon for patients with stage III melanoma is no longer an appropriate therapy.

The other interesting thing to note about adjuvant therapy is that all of the trials that have been completed took place in an era where there was a different staging system. In 2018, the AJCC [American Joint Committee on Cancer] 8th edition was implemented. There are a few changes in this staging system, probably for every cancer, but certainly for melanoma and in particular, in stage III melanoma there are now four sub-stages as opposed to three. Previously there were stages IIIA, B and C and now there are A, B, C and D.

Patients with stage IIIA in the older staging system had a higher risk of recurrence than in the new staging system which incorporates the thickness of the primary tumor amongst other features, including nodal involvement, which helps us to concentrate most of the risk to patients with stage IIIC or stage IIID melanoma.

It's a bit difficult to know how to take the data from these trials and apply it to those patients who have the lower risk Stage III disease, because the indication for nivolumab and trametinib and dabrafenib are in patients with stage III resected melanoma but that may not be appropriate for all stage III melanoma patients.

I think it’s important to recognize that there is a new staging system and because of that, we need to take that staging system into context when we are making recommendations for adjuvant therapy.

The other thing that changed was the surgical management of these patients. Patients who have a sentinel lymph node biopsy, if they have a high enough risk primary melanoma, if there is pathological involvement of the sentinel lymph node, they should have a complete lymph node dissection which is essential the removal of all of the lymph nodes in that lymph node basin.

There was a second MSLT study, MSLT-II, which compared patients who received a sentinel lymph node biopsy and randomized patients who had a positive sentinel lymph node biopsy to either lymph node basin monitoring with ultrasound or completion node dissection and the data, which was published in June 2017. This demonstrated that patients who had completion of node dissection didn’t appear to have any better overall survival or distant metastasis-free survival than those patients who did not have a complete lymph node dissection.

How that impacts our understanding of the data from the COMBI-AD and the Checkmate 238 is that those trials were conducted in patients with Stage III melanoma who had complete lymph node dissection. So, again, this is a slightly different population than those we are seeing right now. With that said, it is clear that myself and my colleagues are offering patients with stage III melanoma either nivolumab or dabrafenib and trametinib as a result of the findings of the two trials I mentioned.

Cancer Network: Switching to trials in progress, there are now immunotherapy combination approaches that are being tested in the adjuvant setting. Can you talk about the larger and/or more interesting trials that are ongoing and their rationales?

Dr. Sullivan: There are a number of combined checkpoint inhibitor regimens that are being tested. They have shown particularly with respect to ipilimumab and nivolumab, better response rates, possibly better progression free survival and at least numeric advantages in overall survival.

That is not statistically significant nor is it statistically sound to compare the treatment arms of Checkmate 067 which compared ipilimumab and nivolumab vs nivolumab vs ipilimumab. There is a trial in stage III ipilimumab and nivolumab called Checkmate 915 that has completed accrual and compared patients, and was initially supposed to compare patients in one of three arms, very similar in design as the Checkmate 067 trial except that was for metastatic melanoma and this trial is for stage III and stage IV resected melanoma.

The Checkmate 915 trial is comparing patients who are receiving ipilimumab and nivolumab vs ipilimumab or nivolumab in the adjuvant setting. With the results of the Checkmate 238, when the trial was launched, the Checkmate 238 data had not yet been reported but it was soon after so the ipilimumab was removed from the design so it became a two arm study of ipilimumab and nivolumab vs nivolumab.

Now, in the metastatic setting, the data from the combination of immune checkpoint inhibitors is interesting. You can do a lot with dosing and it seems like when you give two drugs, an anti-CTLA4 antibody like ipilimumab and an anti-PD1 antibody like nivolumab together, there is much higher toxicity than when you give either by themselves. However, there are modifications that can be made to the dosing and because the toxicity seems to be dose-dependent on ipilimumab, you can give a lower dose of ipilimumab with a normal dose of nivolumab and you can give less frequent dosing of ipilimumab. 

The Checkmate 915 study is actually looking at a lower dose of ipilimumab with a normal dose of nivolumab and is being given every 6 weeks as opposed to every 3 weeks, which is the standard use of the FDA-approved dose for metastatic melanoma. The standard is higher dose ipilimumab with lower dose nivolumab every 3 weeks for 4 doses and then you switch to nivolumab and stop giving ipilimumab.

In this 915 trial, ipilimumab was continued every 6 weeks at the lower dose and the nivolumab was given every 2 weeks. Because there seems to be activity with the combination in stage IV melanoma, the idea was could we maybe take advantage of that activity in stage III disease and so that is the regimen. We look forward to results of that study in the next couple of years.

Cancer Network: What about targeted therapies or trials that are combining different modalities? Are there any in the adjuvant setting that you could highlight?

Dr. Sullivan: There are three FDA-approved BRAF-MEK inhibitor combinations and technically, any of those could be used in the adjuvant setting as part of a clinical trial because only the combination of dabrafenib plus trametinib has been approved for adjuvant use. The other issue clinicians are interested in is whether or not there will be any benefit to combining targeted therapy with immunotherapy and there is talk of launching a cooperative group trial of combined immune checkpoint inhibitor therapy plus targeted therapy vs BRAF targeted therapy vs anti-PD1 immunotherapy.

That would be a fairly definitive look at whether or not the addition of immunotherapy plus targeted therapy will end up being a more effective therapy than giving either of the two components alone. There is some rationale for combing targeted with immunotherapy. There is a lot of preclinical data as well as  analysis of tumor samples that suggest that BRAF inhibitors in part work by activating the immune system and that the way that they activate the immune microenvironment is one that would predict benefit from anti-PD1 therapy.

There are a number of phase I and recently phase III trials demonstrating some efficacy of triplet combinations. In the case of the randomized study, it was dabrafenib plus trametinib vs dabrafenib plus trametinib with pembrolizumab. That study showed that the triplet therapy improved progression free survival compared to the doublet dabrafenib plus trametinib therapy and possibly improved duration of response in patients who got the triple therapy. I think there is clear rationale for looking at combining modalities but it’s not clear whether or not that will translate into better overall survival. The data from that randomized study that was presented at ESMO [European Society For Medical Oncology] this year did not have enough follow-up to demonstrate an overall survival advantage and ultimately, that is what we are looking for.

Cancer Network: In general, how do you approach choosing an appropriate adjuvant therapy for a patient?

Dr. Sullivan: This is the right question to ask and it is nuanced. Most of us try to simplify it as much as we can. Some of my colleagues flat out believe that immunotherapy is better than targeted therapy based more on philosophical grounds than necessarily strong data to support that. They will offer anyone, whether they have a BRAF mutation or not, single agent anti-PD therapy outside of potential enrollment on a clinical trial as their standard adjuvant therapy for patients with resected stage III or resected stage IV melanoma. Some of my colleagues, if the patient’s melanoma has a BRAF mutation, will offer BRAF targeted therapy for those patients and if there is not a BRAF mutation present, will recommend single agent anti-PD1 therapy and use the BRAF typing to stratify treatment decision.

I don’t think either of those approaches are right or wrong. What many of my colleagues do is to have long conversations, and these used to be short conversation with patients but these have gotten longer as there have been more treatment options available, and so we discuss the pros and cons of each of the therapies. In the absence of head-to-head comparisons therapy for BRAF-mutated patients who have stage III melanoma, as far as we know, it’s just as good to give a BRAF-MEK inhibitor combination therapy as it is to give single agent nivolumab, so a lot of it comes down to the toxicity profile.

Patients are less likely to have severe toxicity with nivolumab, however, you are much more likely to have permanent toxicity with nivolumab and so a lot of our patients will chose the therapy that may have less permanent toxicity or the therapy that has less severe toxicity in general. It is definitely one of the scenarios where we need to get the input from the patient to help make the right recommendation. My general philosophy is that if a patient is BRAF-mutated and have high-risk melanoma that has been removed, the likelihood is that they will need both targeted therapy and immunotherapy over the course of their lifetime to treat their cancer.

I often think about offering BRAF-targeted therapy in the adjuvant setting because it appears to work a little bit differently in the adjuvant setting than it does in the metastatic setting. For example, the data we have from COMBI-AD suggests that patients can stop therapy after 1 year of treatment and have sustained improvement in relapse-free survival and overall survival compared to those who got placebo. In the metastatic setting, it is very rare for someone to stop BRAF-targeted therapy and have ongoing control of their disease, most patients will relapse even if they have been on therapy for 5 years and had a complete response. They will stop and the disease will grow back.

In the adjuvant setting that does not seem to be the case, there seems to be a separation between the patients who receive the combination vs those who received placebo, suggesting that at least some percentage of patients are cured with dabrafenib and trametinib in the adjuvant setting vs, I think, there are very few patients cured with the BRAF-MEK combinations in the metastatic setting.

So because the combination is potentially curative in the adjuvant setting and likely not curative in the metastatic setting, it seems to be working differently. Nivolumab and ipilimumab, have definite improvement in relapse-free survival and ipiliumab has improvement in overall survival and we expect that nivolumab will have improvements in overall survival compared to patients receiving ipilimumab in the Checkmate 238 study.

We think it is curative for some stage III patients and we think it is likely curative for some patients with stage IV disease. It is not clear whether the percentage of patients cured in the adjuvant setting is different from the percentage of patients cured in the metastatic setting, and so nivolumab may be operating very similarly in the adjuvant setting. So, when you have the opportunity to offer one of two treatment regimens, thinking that you will probably have to give both of those types of therapies over the course of their treatment, giving them something that works differently in the adjuvant setting compared to the metastatic setting is at least, to me, and some of my colleagues, seems to make a lot more sense than giving a drug that likely works very similarly in the adjuvant and metastatic settings.

That is the rationale I often use to justify the use of dabrafenib plus trametinib in patients with stage III melanoma who are BRAF-mutated, over nivolumab, in the absence of any further data that supports that nivolumab therapy is better than dabrafenib plus trametinib. Obviously, in patients who are BRAF wildtype, the decision is much simpler, nivolumab is better than ipilimumab, and pembrolizumab, like nivolumab, is probably going to be shown to be better than ipilimumab or interferon in a cooperative group study that closed enrollment a few years back.

I am pretty confident that anti-PD1 therapy is the right therapy for BRAF wildtype patients at high-risk of recurrence after curative intent surgery and I think it is pretty straightforward to recommend an anti-PD1 therapy for that patient.