Transcript:Robert A. Figlin, MD: This is probably the single most important part of our conversation today. So, Elizabeth, I want to put you on the spot and ask you to help us all understand how to manage these patients. Let me see if I can summarize. We decided and had a conversation elegantly around what constitutes progression, and I think that has to be made on an individual basis, patient-to-patient, which is hard to do in a general way. We now have FDA-approved agents that are new. We have old agents like axitinib in this space. We have new agents such as cabozantinib, nivolumab, lenvatinib, and everolimus in combination. So, the practicing doctor is faced with multiple options, never having been compared with each other. They have axitinib compared to sorafenib, the others compared to everolimus—all approved in the so-called second-line setting with each of the companies knocking on their door saying, use my agent first. How do you navigate this?

Elizabeth R. Plimack, MD, MS: First of all, I think we are all grateful to have so many options—3 agents, or 3 straight approaches, FDA approved for renal cell in the last 6 months. So, it is a very exciting time. Even though we don’t know exactly how to sequence them, we’re really glad that we have them all. I think the conversation, now more than ever, with our patients is not which one do we use, but which one should we try next for you, right? And understanding and expecting that patients will hopefully get a chance to go through many of them. At least in my practice, we are seeing patients going through multiple lines of therapy as each one emerges and is available to them.

So, how do we select what’s next for that patient? I think some of the issues you brought up before apply, in terms of how they get on the thing that they got before, right? If they had immunotherapy as part of a clinical trial, it probably makes sense to move to something else. If they had a VEGF TKI and did really well on that, those are the people for whom we could try axitinib, to extend that before we move to something else. Cabozantinib has come along, and in that study they were allowed to have prior PD-1 inhibitors, and they were allowed to have more than one prior therapy. So, we have a little bit of evidence in that setting for cabozantinib. I think the one that’s most challenging, at least for me, is figuring out how to place lenvatinib with everolimus.

It’s different in a lot of ways. It was a smaller study, it’s a combination therapy, and it’s the first time we’ve really had FDA-approved combination therapy in renal cell carcinoma. And in my mind, the bar is a little bit higher for that combination. It’s more expensive, in terms of toxicity to the patient and in terms of societal costs. And, again, figuring out where that goes? I’m not sure, but I can say everolimus probably won’t be given without lenvatinib so much in the future, given that we do have comparative data in that setting. So, there are no clear answers, but I think a lot of room.

Robert A. Figlin, MD: Let’s just clarify one area, and I want to get some opinions of other panel members. Elizabeth, you made the comment that response to prior therapy may dictate how someone will do in the next line.

Elizabeth R. Plimack, MD, MS: Maybe.

Robert A. Figlin, MD: Tom, your thoughts about that?

Thomas Hutson, DO, PharmD: You know that has been the modus operandi that all of us have used in practice without data to support it. In fact, the data actually say the opposite, and the data are really retrospective. We have European data from the Italian groups who have been prolific in some publications, and then the largest data I’m aware of is our international consortium with the KCA (Kidney Cancer Association) sponsor data set from Danny Heng and Toni Choueiri. And in that data, it’s clear that response and even toxicity to frontline therapy doesn’t predict response and toxicity at second-line. But I hear what you are saying, Elizabeth. A lot of us all apply that as how we choose, but it’s very difficult.

Robert A. Figlin, MD: Nizar?

Nizar M. Tannir, MD: I think it’s important to mention about the pseudoprogression. I think this is really important. In my opinion, the patients who really have pseudoprogression are in the single digits; it’s 5%. And this is true for melanoma, as well as renal cell. So, the take-home message that we have to give to community oncologists is, if a patient is having symptoms on therapy with a PD-1 or a PD-L1 inhibitor, they continue to have these symptoms, and they develop a new, not-so-small lung nodule, or if they have already lung metastasis, but now they have a bone metastasis that’s symptomatic—those patients should not continue with their therapy because there are re-progressors.

Remember that only 25% of the patients on the CheckMate-025 study have had a major response. So, most patients are not going to respond. And I think if a patient’s tumor is already, I’m going to use the term “on fire,” time is of the essence. We cannot continue with therapy saying that this is pseudoprogression because time is of the essence. You really need to switch therapy before they end up with hospice. I think this is an important point.

Now, regarding the question that you posed to us about how we now sequence these three therapies, I will say there are five therapies. You have nivolumab, you have cabozantinib, you have the lenvatinib/everolimus combination, axitinib is there, and everolimus was there up to just a year ago or less. In my opinion, the ranking has to be immunotherapy/cabozantinib, or cabozantinib/immunotherapy, and then the rest are dropped. Everolimus is a single agent, and 2 phase III trials showed a survival advantage with nivolumab over everolimus and with cabozantinib over everolimus. So, it’s best to use everolimus alone, except for that patient whose tumor harbors the TSC1 (tuberous sclerosis 1), TSC2 (tuberous sclerosis 2), the mTOR gene pathways, mutations where they may have a response.

Now, I have to say I cannot embrace lenvatinib plus everolimus right now based on a 153-patient randomized phase II trial. So, if you took two patients from one arm to the other, that will change the whole statistics. If it’s toxic, 75% of the patients who received lenvatinib/everolimus had dose reduction. And you had 25% who discontinued therapy because of toxicity, compared to 8% on nivolumab alone and compared to 10% with cabozantinib alone in the phase III trial. So, I can’t put this on the same par with nivolumab and cabozantinib.

I think an important consideration, other than toxicity and cost, is the patient. There are patients who like to take an oral agent, even if you try to convince them, “Look, you need a new checkpoint inhibitor. This is going to provide you with a better quality of remission.” The other issue is the toxicity. So, if a patient, say, has history of Guillain-Barré post viral… I had that patient who had post viral Guillain-Barré, an orthopedic surgeon who had two prior TKIs. I wanted to give him nivolumab, but then I couldn’t when he had the Guillain-Barré. And the patient who has nephrotic-range proteinuria from a prior TKI, you’re not going to go to cabozantinib in that patient. You prefer once their proteinuria resolves, to treat them with nivolumab.

I think the important thing to mention about the immunotherapies here is that in my opinion, toxicity is less when you just use the PD-1/PD-L1 inhibitor. But what you look at that’s important here that may not be provided by any target agent is that tail at the end of the curve, that quality of response. And I think David has published on this from the phase I trials and the phase II trials, you may have 10%, 15% of the patients who are going to be in remission 3, 4, 5 years and beyond. I think we’re all waiting on these long-term survivals in remission, similar to the story with IL-2. If that holds up for 5 years and longer, then I think you have to give them two immunotherapies as the first one of the sequence.

Transcript Edited for Clarity