Transcript:Daniel Heng, MD, MPH, FRCPC: Let’s talk about another group of patients, the non-clear cell carcinomas. Nizar, you did some studies in non-clear cell carcinoma. Do you want to talk about that?

Nizar M. Tannir, MD, FACP: Yes. I think this is a very challenging group of patients to treat. Unfortunately, the studies have all been conducted in patients with clear cell RCC, except for the Global RCC trial with temsirolimus versus interferon versus the combination; 20% of those patients had non-clear cell histologies. But there have been data all over the place with targeted agents in non-clear cell kidney cancer. Initial retrospective studies showed activity with sorafenib and sunitinib.

Then, of course, in that global RCC trial I just mentioned temsirolimus was superior to interferon in that 20% of the patients who had non-clear cell histology. So there was interest to test an mTOR inhibitor in non-clear cell tumors. And there have been single-arm studies with sunitinib. I’m not aware of anyone who has published with pazopanib, although there is one study that’s been conducted by the Mayo Clinic, I believe. But with sunitinib there have been at least several single-arm studies in non-clear cell.

One Korean study was positive in the sense it had progression-free survival (PFS) in the six-month range with sunitinib. The response rate was in the 30% range, similar to what we see with clear cell tumors. But in our single-arm study with sunitinib that we published three years ago in European Urology, conducted in 57 patients, reported a 5% response rate and a median PFS of 2.7 months.

So we were not impressed with sunitinib in metastatic non-clear cell tumors, but that study had a mixture of patients with good-risk, intermediate-risk, and some poor-risk and there were 8 patients who were previously treated. But we decided to conduct a randomized trial looking at mTOR inhibition. We used everolimus in the ESPN trial which looked at everolimus versus sunitinib. It was a prospective evaluation of non-clear cell RCC. And the Duke team led by Andy Armstrong also designed a very, very similar trial, the ASPEN trial.

Those two trials basically compared everolimus and sunitinib, and the primary endpoint for both was PFS in the first-line setting. The difference between ESPN and ASPEN is that in ESPN allowed patients after progression with a first-line agent to cross over. ASPEN did not allow crossover. ASPEN recruited patients with papillary, chromophobe, and unclassified subtypes. We recruited the same and we added another cohort of patients with clear cell RCC who had more than 20% sarcomatoid features in the primary tumor. The results of both of these trials were very similar, and I’ll point some differences in the results but very similar.

Our trial was presented at ASCO 2014 and Andy presented the ASPEN results this past June at ASCO 2015. Sunitinib was found to be superior to everolimus, or at least everolimus was not found to be superior as the premise was in both trials. So the data were better with sunitinib, but both agents showed very modest activity, therefore reinforcing the idea that we really need to have more effective therapies for these subgroups of patients.

It’s a diverse group of patients with different biology than clear cell RCC which is driven by aberrant VHL, and therefore a VEGF disease. The non-clear cell tumors are not necessarily so, and they have their own different genetic and genomic alterations. So we really need to continue the research to study them as well as to develop therapies that are more effective for the non-clear cell kidney cancers.

But I’ll come to the differences between the two trials. With ASPEN, they found that chromophobe patients did better with everolimus than they did with sunitinib. And patients who had poor-risk disease in the ASPEN trial benefited better with everolimus. Now, we did not find everolimus better than sunitinib in the chromophobe patients. In fact, our two trials, the first one, the single-arm study we published and the ESPN trial both showed efficacy of sunitinib in chromophobe tumors. The ESPN trial had only 2 patients who had poor risk so I can’t comment on that. So the way I treat my patients with non-clear cell histologies is that I enroll them on a clinical trial. We need to enroll them on clinical trials.

Paul Nathan, MBBS, PhD, FRCP: The problem with this area and he eloquently described it is that we’re not dealing with activity in one cancer type. We’re dealing with activity in a whole range of different cancer types. Because of the relative rarity, the studies are never large enough to be able to do a powered subset analysis and really get an answer about whether or not you can identify a group of cancers that has benefit from one drug as opposed to the other. So the situation is that we’re dealing with effectively a whole series of niche cancers, rare cancers.

Now, there are ways to do rare cancer studies, but they’re not generally best done in the way that we’ve done them so far, which is take a whole series of them and you have 50 patients, either randomized or not randomized. You don’t learn enough from that. And so there are international rare cancer programs in a whole variety of different trials where people work more collaboratively where you generate databases and registries, and you pool data and get some sort of higher powered insight into what’s going on.

Daniel Heng, MD, MPH, FRCPC: I think we’re getting better at that because now there’s a single arm savolitinib study that slowly looked at papillary renal cell carcinoma. And now there’s an Intergroup study comparing different MET inhibitors for just papillary tumors. So I think they’re starting to get the momentum going to study a single, more homogenous sample size.

Paul Nathan, MBBS, PhD, FRCP: It’s certainly possible with papillary tumors because you’ve probably just got about enough patients to do that, but with the other ones it becomes even more difficult.

Nizar M. Tannir, MD, FACP: But even there though you have papillary type 1 and papillary type 2; even papillary type 2 is different subtypes.

Daniel Heng, MD, MPH, FRCPC: Correct.

Nizar M. Tannir, MD, FACP: As we know, molecularly….

Susanne Osanto, MD, PhD: Papillary type 1 is c-Met driven predominantly and the c-Met inhibitors are doing very well in papillary type 1 tumors.

Daniel Heng, MD, MPH, FRCPC: But maybe not type 2, who knows.

Susanne Osanto, MD, PhD: Type 2 is different. Type 2 is genetically different.

Carlos H. Barrios, MD: That again is a challenge in all oncology. We’re starting to genetically study patients and tumors and finding a small number of patients with specific abnormalities and that’s going to be, like you said, a very big challenge for all of us in how we’re going to be able to get enough patients in order to actually conduct the appropriate trials.

Daniel Heng, MD, MPH, FRCPC: Let’s move on to second-line therapy now.

Susanne Osanto, MD, PhD: I want to quickly add something which I think is extremely important and you pointed it out. I think that the more we have, I think it’s really critically important that we as doctors learn from our patients and that we’re really aware of how they feel and how their lives are. And I think that COMPARZ was a very important trial and PISCES was a unique trial. It’s not the only trial in cancer to investigate patient preference on top of the doctors’ preferences.

But I really think that we should do more investigations in that area because it’s so important— how the life of the patient is. I was a bit surprised because years before, in 2012, there was the JCO publication of pazopanib as a phase III trial. But we already knew from phase I data that pazopanib is a very easy drug for patients. So I tended to use it. I started to use it for the elderly patients, and I’m still very happy, even before we had the COMPARZ data and the PISCES data. It’s something which is important.

Transcript Edited for Clarity