Managing Recurrence After Adjuvant Therapy in Melanoma



Jeffrey S. Weber, MD, PhD: What do you do when someone’s on adjuvant therapy, whether it be pembrolizumab, or nivolumab, or a BRAF/MEK inhibitor, dabrafenib/trametinib, and then they recur? Let’s say they recur with unresectable or borderline resectable disease. How do you manage those patients? Omid, what do you do? What do you tell those patients?

Omid Hamid, MD: If it’s unresectable, it’s clearly similar to recurring in a metastatic setting. We then discuss the clinical protocol, clinical trials, and steps for moving forward, or we may discuss the possibility of some form of neoadjuvant therapy. If they’re on a PD-1 inhibitor, then we will talk to them about trying to render them free of disease, going on to a BRAF and MEK if possible. And then, resecting—doing somewhat of a neoadjuvant/adjuvant approach. Whether I would switch depends on how early they’ve recurred and what I have available for them. It’s very hard to drop a PD-1 therapy in a BRAF wild-type patient without other options. We’ll probably think about adding something in there.

Jeffrey S. Weber, MD, PhD: Does time matter? Jason, what if it’s someone who relapsed off adjuvant nivolumab—standard adjuvant therapy? It has been 7 months since they finished their 1 year of therapy. They got through it and they relapsed. What are you going to do?

Jason J. Luke, MD, FACP: This is a really tough question. I think our entire field is going to have to grapple with this. Patients who are getting anti—PD-1, who recur, clearly failed the drug, right? If they have completed adjuvant therapy, how long before you also think that they have failed the drug? This is a debate that many of us have been having.

Some people are starting to say around 6 months, right? If the patient completed adjuvant therapy and recurred within 6 months, the case may be deemed as primary failure to a PD-1. Out beyond that, for 7 months, it could be either way. I kind of use that as a general principle. As you get closer to the 6 months, it obviously starts to get more gray. The very obvious ones are the ones on a drug. Once you get out there, especially in BRAF wild-type patients, PD-1—based therapy is so powerful that to not go back to it, if you have any sense that it might work, is pretty tough. This is a wide-open question that we just don’t know the answer to.

Jeffrey S. Weber, MD, PhD: Some of the data will come from an extensive in-depth analysis of what happens to patients after they’re on these trials. The problem is, by definition, with respect to toxicity, once you’re past the 100 days from your last dose, it’s no longer followed, which would be expensive for the companies. There’s no specific follow up for whether you respond to the next treatment. Just recurrence-free survival and overall survival are assessed. In any future adjuvant trials, I think we should build in the post relapse, natural history. This costs money, but somebody has to get paid to dig up those records. We’re in the ozone a little bit. Hussein, in your institution, at 6 months, if there is a relapse, do you go back to the original regimen? At less than 6 months, do you change?

Hussein Tawbi, MD, PhD: I will step back and tell you that with the new approvals of adjuvant therapy, I think we’re defining new phenotypes of metastatic disease. We’re always going to have the metastatic patient whose first presentation is metastases. For those patients, all of our current data apply in the first-line setting. But for patients who recur after adjuvant therapy, it’s a whole new phenotype. We will need to really take that into consideration.

Obviously, our job is to try to make sense of this. And so, in our clinical trials, we’ve been using 6 months as the cutoff. It’s more or less an arbitrary cutoff. We’ve used it for our patients who recur after adjuvant ipilimumab for the last several years. I think it’s a reasonable place to start, to start collecting data in the new first-line setting in patients who are no longer completely treatment naïve.

Jason J. Luke, MD, FACP: I have a question to ask the panel. This patient progresses outright on adjuvant PD-1 and is BRAF wild-type. Is your go-to regimen ipilimumab, or do you give them ipilimumab/nivolumab?

Jeffrey S. Weber, MD, PhD: If they’re re-resected?

Jason J. Luke, MD, FACP: No. If they have unresectable disease.

Jeffrey S. Weber, MD, PhD: OK.

Jason J. Luke, MD, FACP: I’m not sure everyone knows the answer to this. In my experience, I tend to go to ipilimumab/nivolumab in those patients. Again, it’s anecdotal because we don’t have an aggregated series. They tend to be more aggressive, in terms of their presentation. Therefore, I’m more concerned about…

Jeffrey S. Weber, MD, PhD: Well, let’s go around the table. Hussein?

Hussein Tawbi, MD, PhD: Exactly as I was just saying, this is a new phenotype. Most of the time, to start with, it’s going to be some PD-1—based clinical trial. But I would try to bring in ipilimumab as soon as I can in that space. We’ve recently seen ipilimumab plus intratumoral injection in a clinical trial. I choose that as treatment for patients who have recurred outright on a PD-1 therapy. I agree with you that ipilimumab still has a role to play. It still cures patients in the metastatic setting. We need to try to reintroduce it. Whether I use ipilimumab/nivolumab versus other combinations on a clinical trial depends on the exact case.

Jeffrey S. Weber, MD, PhD: Robert, what do you vote for?

Robert Andtbacka, MD, CM: Looking at the data, we know that these patients who went on these adjuvant studies were at about 3 months. They all had to be restaged. When they went on the study, they had no evidence of recurrence of disease. If they then recur within the first 3 months, which happens in up to 15% of patients who are treated with nivolumab or pembrolizumab, they are very difficult to treat. I agree with Hussein. At our institutions, we have a number of trials that allow these patients to go on with an agent. Many times, the recurrence tends to be locoregional rather than distant. I think that we would use that. If they have BRAF-mutant disease, we would definitely consider a BRAF/MEK combination regimen.

To Jason’s point about ipilimumab/nivolumab, this is really challenging. Again, we have minimal data on this. The data that we do have on this demonstrate that the response rates are not as impressive as we’d like them to be. They are about 15%. There are also limited data on injectables with ipilimumab. The response rates seem to be a little bit greater, but we’ve never compared them. I’d be inclined to say that we need to do something different. Bring this earlier on in that immune cascade, and try to change the microenvironment, if they have an injectable lesion. That would be my preference.

Jeffrey S. Weber, MD, PhD: Omid, they relapse off pembrolizumab or nivolumab. Are you going to give them ipilimumab/nivolumab?

Omid Hamid, MD: This is the greatest call for our clinical trials. What I’ve learned from sitting next to Jason Luke is that this is biomarker, biomarker, biomarker. I would look for some evidence of the right combination to use. I may go to PD-1, anti-CTLA4. Or, if there is any indication that there is a combination available that’s better…I do feel uncomfortable just giving a single agent.

Jeffrey S. Weber, MD, PhD: OK. And lastly, Michael?

Michael A. Postow, MD: This calls into question whether there is a PD-1 backbone in melanoma. If you start with a PD-1 therapy and progress, do you continue that PD-1 and add drug X, in trial, or an injectable or whatever? Maybe you would even consider ipilimumab with that backbone? Or, do you just stop the PD-1 and switch to ipilimumab as a new baseline treatment or combine ipilimumab with some other new experimental drug? We really don’t know. In melanoma, in the BRAF wild-type setting, when do you pull the ipilimumab trigger after PD-1 of some duration? Do you pull the trigger right up front? Do you pull it after someone has progressed after some period of time on adjuvant therapy? We can debate what that is. No one really knows. We have no randomized data yet to evaluate this. At the end of the day, using ipilimumab alone or ipilimumab and nivolumab after PD-1 progression is a gut check. It’s an emotional decision. It’s about what to do for the patient. I think we can sit here and debate about the science that we don’t really have yet for hours. It’s an emotional call, one way or the other, and I’m happy to admit to that.

The other thing that I would say is, even if you stop PD-1 therapy and switch right to ipilimumab monotherapy, if you think you’re giving ipilimumab alone—if that last dose of PD-1 was just a few weeks before—you’re still giving ipilimumab/PD-1 because the pharmacodynamics of the PD-1 alone are likely still in the system when you start the ipilimumab monotherapy.

Jeffrey S. Weber, MD, PhD: Which will cause a fair amount of toxicity.

Michael A. Postow, MD: As you’ve shown, Jeff, in your CheckMate-064 study, there’s more toxicity with ipilimumab monotherapy after PD-1, but it’s not really ipilimumab monotherapy because you still have the PD-1 support.

Jeffrey S. Weber, MD, PhD: Absolutely.

Michael A. Postow, MD: Don’t fool yourself that you’re really switching to ipilimumab monotherapy. We’re probably giving ipilimumab/PD-1 to all of these patients anyway, especially if the PD-1 was stopped recently.

Transcript Edited for Clarity

Related Videos
Michael R. Migden, MD
Daniel Olson, MD
Neil D. Gross, MD, FACS
Neil D. Gross, MD, FACS
Harriet Kluger, MD, Harvey and Kate Cushing Professor of Medicine (Oncology) and of Dermatology; director, Yale SPORE in Skin Cancer; vice chair, Translational Research, Internal Medicine; chief, Division of Skin and Kidney Cancer; associate cancer center director, Education, Training and Faculty Development; deputy section chief, Medical Oncology, Yale Cancer Center
Paul D. Nathan, MBBS, PhD, FRCP
Jeffrey S. Weber, MD, PhD
Patricia A. Possik, PhD