In the final segment of a debate series, Brian Rini, MD, and Monty Pal, MD, discuss RCC adjuvant therapeutic options and how patient characteristics impact treatment choice.
Dr. Pal: I thought we’d open up the conversation discussing the adjuvant space in kidney cancer. This is pretty exciting. I remember when I was getting into the field, a lot of the trials that we discussed in the tyrosine kinase inhibitor (TKI) era were brewing, S-TRAC and the ASSURE study and so forth. Those I don’t think have really made their way into clinical implementation.
But KEYNOTE-564 I think really changed things. This was a study of adjuvant pembrolizumab. This study was positive for disease-free survival and showed a tinge of overall survival (OS) advantage. I’ll say again, very early on, very immature data, but at least the data looks to be heading in the right direction. This study took patients that were T2 grade 4, T3 grade whatever and up essentially, and I started to incorporate this into my clinical practice across the board for patients. I don’t know, Brian, are you a little bit more selective in terms of who you administer pembrolizumab to in the adjuvant setting?
Dr. Rini: Yeah, I struggle with this a little bit. You mentioned the positive KEYNOTE-564 data, and I agree, it’s a good trial and stands on its own. Even in those high-risk patients, the majority of patients still won’t recur, and our ability to predict on an individual patient basis is basically zero. We just don’t have those tools yet. We also have, as you know at ESMO, 3 other negative trials reported: adjuvant atezolizumab, adjuvant ipilimumab/nivolumab, and then a neo/adjuvant nivolumab study from the US Cooperative Groups. They were not just negative but flatly negative. The hazard ratios for disease-free survival were 0.9-something, and quite obviously no OS advantage. I think the field sort of took a step back and said, wait a minute. It’s not just, yes, any adjuvant immuno-oncology (IO) is going to work and let’s give it to everybody.
We can explain away why each of those trials were negative, be it PD-L1 or toxicity or some of the issues around treatment in the neoadjuvant trial. But at the end of the day, if you look at the totality of the data, for me, I took a little bit of a step back, and when I’m talking with patients, I tend to give adjuvant pembrolizumab only to those really higher-risk patients. So T3 high grade and greater risk. Because even though well tolerated in most, there’s 1 or 2 out of 10 patients who will need high-dose steroids and have lifelong endocrinopathy-type side effects, so it’s not without cost. I think it’s probably about 50/50 in my practice of patients that I offer it to that get it. Some just say, I’m just going to take my chance and get scans.
Dr. Pal: That’s so interesting. This debate keeps coming up all the time in oncology in general. I always listen to The Uromigos podcast to sort this out by the way. Not just a cheesy plug there, it’s the truth. But some of the convincing things for me were, for instance, seeing Toni [Choueiri, MD] and Tom [Powles, MD] kind of parse out the data from KEYNOTE-564 in that Lancet Oncology paper. They segmented out the high-risk patients, T4 and beyond. They looked at the node positives, the sarcomatoid subsets. Certainly there, especially if you look at M1 for instance, and a hazard ratio is hard to argue against, 0.29 or something mild like that.
But if you look at the T3s, I still think that there’s benefit there. Then that question arises of, can you really parse out the T2 grade 4s? Can you really parse out the T3a grade 2s? Is there a value in doing that? I kind of liken it to this PARP debate that we had on prostate cancer at ASCO GU this year. Do you treat everybody across the board with metastatic castration-resistant prostate cancer, with a PARP inhibitor, or do you limit it to those that are BRCA1 or BRCA2 altered? These sorts of things, like dividing up subsets, just become so complicated. With all the differences in the other studies that were negative, including my own, very sad about that, and the positive KEYNOTE-564 data, I tend to sort of look at that in isolation.
Dr. Rini: I think your analogy to the PARP inhibitors is a good one, and there’s a setting where we actually have a biomarker, but we’re still debating how to use it. That’s a wild area of debate. In RCC, we don’t have such a biomarker, except for perhaps being on the higher end of risk for sarcomatoid or something. But that’s unfortunately not most patients. Most patients have T1 disease, but even among the high risk, 85% were in the lowest risk category of their 3 risk categories, I think there’s a lot of work to do in adjuvant kidney cancer.
There’s other adjuvant trials going on, or at least 1 I know of, pembrolizumab plus belzutifan versus pembrolizumab. My personal opinion is that I think the biology of relapsing kidney cancer, of high-risk kidney cancer, is more driven by immune pathways than HIF/VEGF pathways. We saw basically all the VEGF trials were negative, and 1 marginally positive. My personal take, and there’s some recent data out of Ari Hakimi’s lab that’s come out to say that the biology of relapse in kidney cancers is immune-driven. My personal bias is that stacking immune drugs in that setting is helpful. Now I’ll contradict myself to say, well then why wouldn’t ipilimumab/nivolumab be positive? Why wouldn’t that trial be positive? I don’t have a good answer, except for toxicity limiting drug delivery.
Dr. Pal: I think that that toxicity question is key. The way that I look at it is, trials like S-TRAC, ASSURE, etc., they really were so limited by factors around dose intensity. From my standpoint, it was such a challenge to get patients out to a year of a VEGF TKI. I had challenges participating in EVEREST, getting patients to a year of an mTOR inhibitor. Whereas I’m thinking that with belzutifan, just based on tolerability profile, that may be a little bit different. It may be a great opportunity for us to really test the ability of something that strikes along the VEGF pathway to assist in the adjuvant setting. There might be hopes that most patients make it through a year of treatment in that context.
Dr. Rini: Yeah, I guess you have to postulate that it wasn’t the pathway but lack of drug delivery that hampered the VEGF studies. I think there’s some data to support that, but we’ll find out. It’s certainly a debated and exciting area.