Tian Zhang, MD: I’d love to delve a little bit more into this topic and ask Dr Ornstein, which patients are you choosing lenvatinib [Lenvima]/pembrolizumab [Keytruda] for versus cabozantinib [Cabometyx] with nivolumab [Opdivo], versus even a pure immunotherapy combination, ipilimumab [Yervoy] and nivolumab?

Moshe C. Ornstein, MD, MA: I completely agree with Mehmet, what was just said, that obviously, the first question is to decide IO/IO [immuno-oncology/immuno-oncology], ipi/nivo [ipilimumab/nivolumab] versus 1 of the combinations. I certainly take all those things that were just mentioned of comorbidities, the chance for having a surgery or underlying autoimmune disease, sarcomatoid, histology, etc. I’ll just try to highlight a couple of different things maybe. One of them is that it is important to mention, as Dr Braun said earlier, that at the end of the day, the IO/IO, the ipi/nivo combination was first to the market, has the longest follow-up, the longest durability responses. Because it was first, it does have those 5-year data and you can see percentage of patients alive and PFS [progression-free survival] at that mark. I think if you’re looking for what has the longest duration of data, the longest durability of response, where we sit right now, it’s going to be belimumab [Benlysta] and nivolumab. It also has the benefit of not requiring any pills, so there is no daily TKIs [tyrosine kinase inhibitors]. Once you’re done with the 4 doses of ipilimumab and nivolumab, it’s single-agent nivolumab once a month, so there’s a lot of benefits there with that long follow-up and basically single-agent therapy once a month once you’re done with the induction. That said, that has to be balanced with the real benefits that come with any of the IO/TKI regimens. And that is a clinical benefit across risk groups, of course, cross-trial comparisons limiting, it is fair to say that there’s better response rates and better PFS when we’re talking about IO/TKIs versus ipilimumab and nivolumab. The follow-up notwithstanding, the durability of responses notwithstanding, for a patient who walks into clinic who’s symptomatic either with fevers, chills, night sweats, or they have back pain or abdominal pain or flank pain from their cancer, there are higher response rates for the patient that needs a response now, and the response rates really are fairly drastically higher when you compare ipilimumab and nivolumab versus the IO/TKI regimens. In my practice that’s why I tend to lean that way. I have a lot of patients who I keep them on surveillance. Generally, I’m treating patients who are a little bit sicker, and therefore, I tend to lean towards the IO/TKI combination because I want to be able to provide that response to that patient and I want to be able to tell them that there is a longer PFS. What’s going to happen in 5, 6 years? We’ll have to wait until the data play out, but at the end of the day it really is immunotherapy even if there’s no anti-CTLA4. There should be at least some durability of response.

Now, designing within IO/TKIs becomes really challenging because we always get yelled at when we try to do these cross-trial comparisons. There’s a benefit to the axitinib [Inlyta] and pembrolizumab combination because the axitinib has the shortest half-life so it’s easiest to stop it and to tease out the toxicity. It’s very easy to titrate it, but really on the other end of the spectrum, you do have cabozantinib with a very long half-life, but when you look at the response rate and the PFS, it is a little bit longer. I think when you take the data in totality and you look for the trial that did demonstrate the highest response rate and the longest PFS and the highest CR [complete response] rate, it still is lenvatinib and pembrolizumab. Now, there are those that will argue that is there really a demonstrable difference between any of the IO/TKIs? We can’t say that for sure because we’ll never compare them head to head, so I do tend in some ways to lump them together, trying to highlight the benefit of each 1. I’ll give you an example. If I have an elderly patient with multiple comorbidities where I know that I need to stop the toxicity right away if it happens, I’m more likely to use axitinib and pembrolizumab. But for the patient who’s highly symptomatic, where I really need that response rate regardless of toxicity, I’m more inclined to use something likely lenvatinib and pembrolizumab because the response rate is 71% and the PFS is close to 2 years. I think, again, we’re first looking at IO/IO versus IO/TKI, and then within IO/TKIs, each 1 has merit. The best efficacy if you’re looking clearly at data, again, across trials, you’re looking at lenvatinib and pembrolizumab, but I think any 1 of them is reasonable. Again, it goes back to being able to manage patients on any 1 of them because the toxicities can be similar and the 1 that we’re most comfortable with keeping the patients on for the longest duration of time is probably the 1 that’ll provide the most benefit to an individual patient.

Pedro C. Barata, MD, MSc: Tian, if I may add to what my colleague said, I’d just like to highlight 2 points. One has to do with histology, how that helps us to make a decision, just briefly to mention. First, for the non-clear cell histologies because all these phase trials, for the most part, actually, included patients with a RCC [renal cell carcinoma] clear cell competent. Really, we also know that immunotherapy, immune checkpoint inhibitors, tend to respond last in the non-clear cell histology compared to clear cell. I have to say in that scenario, especially if I think of popularity, I try to bring 2 different mechanisms of action, so I think of a TKI and IO instead of an IO/IO. To be quite frank, we know for non-clear cells, especially popularity data came out from pop Metro and is a very elegant study that basically shows superiority of cabozantinib versus sunitinib [Sutent]. Perhaps, it could be used as a backbone TKI and then if we will, we could potentially think of something like cabo/nivo [cabozantinib/nivolumab].

If we’re thinking of cabozantinib for those patients, then bring nivolumab in addition to cabozantinib. The thought for that is, as my colleagues had mentioned, we never know exactly how the patient is going to respond to immunotherapy, but it is possible that we can achieve those durable responders with immune checkpoint inhibition. That would be my 1 comment. My second comment has to do with sarcomatoid features. When we talk to our colleagues in the community, in some places I found out that sarcomatoid features are not always are being reported in the path report, and actually that’s something that needs to be emphasized because I think Dr Bilen, as you mentioned, there is a predictive value for higher response rate or higher response rate with checkpoint inhibition in the presence of sarcomatoid features. I guess all we’re discussing in these last several minutes is the value of immune checkpoint inhibitors in thinking of a combination therapy for patients. Think of those factors that we can find a path report that can also provide us some guidance about what age or regimen in this case to use.

Tian Zhang, MD: This has been a really comprehensive mix and wonderful clinical input from all 3 of you.

This transcript has been edited for clarity.