Neeraj Agarwal, MD: Why am I thinking about all this? Because you’re asking me whether I would pick up a TKI or this ipilimumab/nivolumab combination. I would tend to disagree with Nick when he said that all of these TKIs are equal. I think cabozantinib is clearly superior to sunitinib, established by a randomized trial with independent radiological assessment.

So, my TKI of choice in the first line is cabozantinib. Obviously, based on randomized phase III data, nivolumab/ipilimumab should be the choice for most of the patients. But if I look at patients with bone metastasis, like Brad said, cabozantinib was very superior, obviously superior to everolimus in the METEOR trial. I have no reason to believe why it wouldn’t be superior to any other combination in pre—bone-predominant disease, where I’m really worried about progression-free survival benefit.

And then, of course, for the PD-L1—negative patients where there was no PFS benefit with the ipilimumab/nivolumab combination, if I see a high-volume rapidly progressive disease, where I’m really worried about updating a PFS response, or progression-free survival, I think I would lean towards cabozantinib or a TKI in those patients. The patients you mentioned, like the prisoner who came with a big abdominal mass, need immediate palliation, and I doubt you would have chosen immunotherapy for that patient.

Nicholas J. Vogelzang, MD, FASCO, FACP: Oh no, I’d never do that. You need to know what your disease parameters are. I do respect the cabozantinib data, but again, I keep coming back to it being a randomized phase II trial. It’s fairly small, and with all due respect to Toni and everyone else, it was a label expansion, right? I’m good with it. I love it. I haven’t used it as first-line therapy yet, but I think that we all realize our choices there are good. The problem with ipilimumab/nivolumab, until it gets on NCCN guidelines, is that I’m not going to get any insurance company to even look at me. I can’t give nivolumab frontline. Maybe some people can, but I can’t. So, once we get the nivolumab/ipilimumab combination, I’m going to reluctantly use it. Now, I’m actually eager to hear about the presentation of atezolizumab and Avastin (bevacizumab).

Daniel George, MD: We’re going to talk about that in the next section, but I want to follow up on something that Neeraj said that I think is important. I don’t routinely check PD-L1 status in kidney tumors. And yet listening to you talk about this, it really sounds like knowing that PD-L1 status would weigh into your decision in how you’d manage these patients. I’m not saying it’s the only factor, but it would be one of the factors in deciding which way to go: TKI or immunotherapy. Do you routinely check for PD-L1 status in your kidney cancer patients?

Neeraj Agarwal, MD: I won’t say I check for all patients, but we do check PD-L1 expression in many patients. The most important problem with PD-L1 expression testing is that it varies among different vendors. PD-L1 testing may vary from one vendor to another, and that has been published recently in JAMA Oncology from the lung cancer perspective. So, that is challenging. However, if I’m looking for any indication, any suggestions, or any guidance on how to treat my patient when I have 2 or 3 different options, I think I have to go with something that may not be very reliable. But if I have to compare 2 drugs in the absence of a randomized comparison, I think I’d like to use PD-L1 testing.

Robert Alter, MD: I think we learned from CheckMate-025 that we cannot depend upon PD-L1 to give us the knowledge that we need to make the right decision. In the CheckMate-025 trial, patients received Opdivo (nivolumab) compared with everolimus. Whether they were PD-L1 positive or negative, patients still who received Opdivo did better, and nivolumab did better than everolimus.

With what we talked about for patients who present and have favorable risk, I would just consider them to receive a TKI automatically. In our practice, we did not test PD-L1, nor did we act upon it. I just defer that to my thoracic oncology partners. But for what I take care of in GU as well as for head and neck cancer, we do not need to have PD-L1 status to actually utilize our immunotherapies at all. To date, and I really agree with Dr. Vogelzang, I would not use the therapy of ipilimumab/nivolumab until it’s FDA approved. I would not use nivolumab as a first-line therapy until it’s FDA approved. But I still would not utilize that as a guide on how to take care of my patients yet. We may have to do that when we have newer immunotherapies or other options that come down the line.

But to date, I would still treat our patients with stable risk based upon how we categorize with the IMDC. I’d possibly treat them with a TKI therapy. But short of that, I think for each medium- and poor-risk patient, I would definitely consider immunotherapy upon the FDA approval. By the way, I don’t believe that all immunotherapies work slowly. I think you may be quite surprised that if you actually have a patient and you give immunotherapy, they may have significant symptomatic improvements within their first cycle of therapy. So, I don’t always think that a TKI adds symptomatic, bad toxicity. To me, is it’s more that the mechanism may be quite beneficial.

Nicholas J. Vogelzang, MD, FASCO, FACP: I think you’re right. I’m a little bit worried about the immunotherapy because in the bladder world, we saw chemotherapy being better at the beginning and immunotherapy as catching up. So, I think every tumor is different. Certainly, I’ve seen dramatic responses with nivolumab within a week or two. There’s no question in some patients, but it’s not the routine.

Robert Alter, MD: Correct.

Daniel George, MD: I think that’s fair. Finally, there are 4 TKIs, since those are still our standard for many of these patients. Where are we? What do we choose? It sounds like Neeraj has switched over to cabozantinib as his standard. It sounds like Brad is more in the cabozantinib camp. Nick?

Nicholas J. Vogelzang, MD, FASCO, FACP: No, I’m still a C10 Votrient (pazopanib) user.

Daniel George, MD: Still C10 Votrient. Bob, where are you?

Robert Alter, MD: We were part of the COMPARZ clinical trial as well. I had 7 patients all randomized to sunitinib. Upon FDA approval, we jumped on the pazopanib therapy, and our patients were doing better tolerability wise. But then I was seeing not as robust a response. So, I’ve adapted utilizing sunitinib again with a schedule-changing difference, using 2/1 rather than 4/2 in my patients. I find the tolerability is better. I find having the half-time happen earlier is important. I think the ability to have our patients maintain a full dose on an adjusted schedule has given me the sense of accomplishing a better response than what I was seeing with pazopanib. So, I’m still more of a sunitinib gentleman as well.

Daniel George, MD: There you have it; it’s a real split group.

Nicholas J. Vogelzang, MD, FASCO, FACP: I would add that the frontline mTOR inhibitor story should be relegated to ancient history. There’s no role for frontline mTOR inhibition in renal cell carcinoma right now.

Neeraj Agarwal, MD: In poor risk.

Nicholas J. Vogelzang, MD, FASCO, FACP: Yes, poor risk, that’s right.

Daniel George, MD: Right, that’s where its label is, where we look at it in terms of risk factors. I agree. I think that when that study was done, it was really a different historical time. It wasn’t even compared against TKIs. More recently, Nick, there have been some studies comparing mTOR strategies to TKI in this frontline setting. What are those results?

Nicholas J. Vogelzang, MD, FASCO, FACP: It was from MD Anderson’s Dr. Tannir. It was a small trial of pazopanib versus temsirolimus. As in the sunitinib versus everolimus trial, pazopanib was clearly superior to temsirolimus IV. So, I think that should not be used. I noticed that sorafenib came off the frontline NCCN trials. Those are drugs that are historically of interest, but not particularly relevant in today’s world.

Daniel George, MD: I would agree.

Robert Alter, MD: I still have 1 patient per year who always has poor insurance coverage, and they have to receive intravenous therapy and until nivolumab is an FDA-approved first-line therapy, I’m still giving them temsirolimus as first-line therapy intravenously because financial toxicity is significant.

Neeraj Agarwal, MD: Or with significant cardiovascular disease, where VEGF TKIs may be a problem.

Daniel George, MD: To your point, Nick, it’s nice to have options. But at the end of the day, I think you hear how our practices are changing, how immunotherapy is coming, how TKIs are evolving, and how mTORs are fading into the background. I think that’s going to continue and hopefully that’s helpful for you in your practices out there in looking at how to manage these patients with frontline metastatic RCC.

Transcript Edited for Clarity