Thomas Powles, MD: I’ll stay with you, Eric: Do you want to give a summary of where we are in the adjuvant setting? This is our last piece that we’re going to talk about before we wrap up, so let’s try to see where we are. There is a lot of information kicking around.

Eric Jonasch, MD: There were multiple studies that were presented at this ESMO [European Society for Medical Oncology] on adjuvant therapy, and so what we now have are 4 studies. One, which is KEYNOTE-564 [NCT03142334], which has resulted in an approval of pembrolizumab in the adjuvant setting. We now also have IMmotion010 [NCT03024996], which was presented at ESMO. We also had the PROSPER study [NCT03055013], which is neoadjuvant/adjuvant nivolumab. We also had the CheckMate-914 study [NCT03138512], which is essentially ipi-nivo [ipilimumab-nivolumab] in the adjuvant setting. The first 3 adjuvant studies have similar eligibility criteria, high-grade T2, and then there are some nuances of whether or not you allow M1 receptor and when the M1 reception occurred. CheckMate-914 did not have that group. The PROSPER study is a little bit different…we ended up having people who had clinical T2 or higher, and they didn’t specify clear vs nonclear because you couldn’t until you actually did the operation. We only have 1 study that’s positive at this point in time: the approved pembrolizumab. All 3 other studies were negative. IMmotion-010 was a large, randomized study with very similar eligibility criteria; the Kaplan-Meier curves, basically there’s no error between them, and if you look at a subset analysis, you don’t see anything. Maybe in the highest grade or highest T there’s a tiny directionality, but nothing you could pretend to call positive. In the CheckMate-914 study…[ipi-nivo], you see that it’s a negative study. You really don’t see anything, and again, subset analysis may be a trend, but it’s nothing that you could really say anything [about]. The PROSPER study, the most challenging, was a completely negative study from a DFS [disease-free survival] perspective. The question then is, why are these 3 studies negative and the 1 study positive? There are a couple of possibilities. One is that there’s something about the eligibility criteria or the study design that resulted in this. For example, in the PROSPER study, were these people clinically too low risk? But when you looked at the pathology specimen, two-thirds of these individuals had T3 or higher disease, so it wasn’t purely too low. Second is, was there some problem with trial conduct? Was there, for example, a high dropout rate or not getting drug in? In CheckMate-914, only 57% or so of patients ended up getting to the end of the treatment, so there was a number of people who did not get full drug. Maybe that matters. Also in the PROSPER study, you see that of the individuals who were randomized to neoadjuvant followed by adjuvant therapy, of the slightly over 400 patients that were in that arm, only slightly more than 300 got onto adjuvant nivolumab. Maybe there were some problems there in terms of trial conduct. Nothing about them being conducted wrong, but execution getting drug in. The last thing is, maybe there are some agents that aren’t as good as others. Maybe atezolizumab, which I think is demonstrated in the metastatic setting, isn’t as strong of a PD-1 blocking agent like pembrolizumab. That could be an explanation there. I think it leaves us with the nivolumab-based studies. There are big question marks; what is happening there? Fortunately, CheckMate-914 has another arm that is getting adjuvant nivolumab, and that’s going to help us understand the nivolumab quandary. It is very confusing right now. I don’t think we’re in the same place; I don’t think this invalidates the pembrolizumab data, but we are definitely in a slightly more confused state than we were a year ago.

Thomas Powles, MD: Ben, we’re confused. How are we going to express that confusion to our patients?

Benjamin Maughan, MD: I will add one other potential confounder to that mix, and that’s biological variability. The inclusion criteria were similar between them, but if we remember, for as dissimilar as biologically metastatic disease is compared to localized disease, except for CheckMate with ipi-nivo, we’re talking about an adjuvant treatment or neoadjuvant-adjuvant monotherapy with a checkpoint inhibitor, and in the metastatic disease setting the response rates are only around 20%.

Thomas Powles, MD: What are we telling our patients?

Benjamin Maughan, MD: It’s a difficult conversation, because I agree with Eric; the KEYNOTE study was well designed. I don’t think we should discount those results. At the end of the day, what are those results? It’s improved disease-free survival, but we still haven’t proven that it’s curing more patients or prolonging overall survival. Disease recurrence is a real issue. I talk to the patients about what the data is; we have pembrolizumab that can improve disease-free survival. We need more time to clearly prove if it’s curing more patients, improving overall survival or not, and the toxicities are what they are.

Thomas Powles, MD: How much pressure does it put on pembrolizumab’s overall survival signal?

Rana McKay, MD: A lot of pressure.

Thomas Powles, MD: What does that mean for your patients?

Rana McKay, MD: We want to show that we’re making our patients live longer. The issue with adjuvant therapy is that, if you look at the 2 extremes, we’re probably overtreating a lot of people, but for those people who are destined to recur and potentially recur in a short time, we could potentially be under-treating them because we’ve proven doublets are better when they have metastatic disease. We really have to show that we, in fact, are not just delaying progression, but rather preventing progression. Longer-term follow-up of the OS [overall survival] signal is going to be important. Right now, it looks promising; we’ve had the first interim analysis and then the updated data at 30 months. It’s nice to see the hazard ratio holding steady for overall survival, but only in 30% of events, so we have to watch that data out.

Thomas Powles, MD: My take on this is very similar to yours, Eric. The pembrolizumab study is a positive study if things are robust. It is delaying progression-free survival. The other trials are struggling for a lot of different reasons. One can look at that and say, “I can come up with a rationale, but it doesn’t seem to be a single reason.” I think the data needs to be presented to patients in the knowledge that there’s life-changing toxicity, but relapsed renal cancer is a real problem and survival on the end is going to be extremely important. When, heaven forbid, those patients do relapse, what’s the role of further therapy? What treatments do you offer your patients?

Ulka Nitin Vaishampayan, MD: It depends on the timing of the relapse. If they’re right through the adjuvant treatment and getting it actively and they have a relapse, adding a single-agent TKI [tyrosine kinase inhibitor] or going to a combination of, for instance, cabozantinib and nivolumab or len-pembro [lenvatinib-pembrolizumab] would make sense. It’s debatable, whether to continue the IO [immuno-oncology] at that point; if it’s 6 months to a year after they’re done with their adjuvant therapy, then I would clear-cut go with the immune combination regimen.

Transcript edited for clarity.