Read the Full Video Transcript

Pedro Barata: Hello, welcome. My name is Pedro Barata. I'm a GU oncologist at the University Hospital Seidman Cancer Center in Cleveland, Ohio. And I'm really, really happy to be joined this morning by the one and only Dr. Eric Jonasch. Dr. Jonasch is a professor in the Department of Genitourinary Medical Oncology, Division of the Cancer Center at MD Anderson in Houston, Texas. He also serves as the Director for the VHL Clinical Center and the Co-Director for the Kidney Cancer Research Program. Eric, so nice to have you here and learn from you again.

Eric Jonasch: Great to be here, Pedro. Looking forward to our chat.

Pedro Barata: Absolutely. And the reason why we're chatting today is a fantastic one. By the way, congratulations on your fantastic talk this last ESMO. Very, very interesting data. And for those who don't know, you are, Eric, one of the persons that I know, at least with the most experience in the world regarding HIF-2 alpha inhibitors around VHL HIF. And so it's really, I think, we all stop when you bring novel data and novel concepts from that perspective and point of view. And that's kind of what happened at ESMO, right? You did present phase one/two data on this new compound NKT-2152, which I think is a small molecule, a new HIF-2 alpha inhibitor. But I'm wondering if you may kind of walk us through how the study was born and can you summarize for us the data—very promising data, in my opinion—for patients treated with clear cell RCC with this compound?

Eric Jonasch: Sure, absolutely, Pedro. This is a small molecule inhibitor of HIF-2 alpha, as you mentioned. It works very similarly to the way Belzutifan works, in that to get transcription from HIF-2 alpha, you need to get heterodimerization of HIF-2 alpha or HIF-1 alpha, for that matter, plus HIF-1 beta. And to prevent that heterodimerization, you can get a small molecule to get into the PAS-B domain of HIF-2 alpha, which basically creates steric hindrance and prevents that heterodimerization. This drug, NKT-2152, does that in the same way as Belzutifan does. This is a study that, very much in the way that the phase one/two study was designed for Belzutifan, is in a heavily pre-treated group of individuals with advanced clear cell renal cell carcinoma. They had a median of three prior lines of therapy. What we presented at ESMO this year, really, were the data mainly on... There's two parts to this: there's the dose escalation part of the study and then there's a dose expansion part of the study, and we're really presenting here on the dose escalation.

This drug has a fairly long half-life, about 38 days. And that has resulted in some different scheduling of this agent, with weekly dosing actually being one of the most... something that's possible here. But what I'm going to show you here is just a couple of slides of some of the efficacy data. What we're first looking at here are the objective response rate data. We have 100 patients overall that are evaluable. But just in the phase one, the data for that entire group are relatively immature. But in the more mature data, we do see that objective response rates are in the mid-twenty percent range in the dose escalation part, in 57 patients. And we can see here, very similar again to Belzutifan, we're seeing that the PR rates overall are about 20%, and PD rates in the 28% range. We have a forest plot figure here on the left, which shows that there appears to be a function of risk subgrouping, that the responses are better in the better risk subgroupings in the higher ECOG performance status.

But prior lines of therapy don't seem to be dependent on that, and in the mTOR-naive subgroup, of which there are 45 patients, there was the best response in that group. All hypothesis-generating, very wide confidence intervals, but interesting data. In terms of the depth of response and durability of response, we can see that once again, on the plot on the left, in this waterfall, we're seeing responses in the poor, intermediate, and favorable risk subgroups, with the majority of individuals having some degree of disease shrinkage. On the right, the spider plot, we're seeing that there is a subgroup of individuals who clearly progressed up front. Then there is a group of individuals that have a durability of response, and the median duration of response, I think this is kind of important, hasn't really been met yet. From a progression-free survival perspective, we're seeing that the median progression-free survival across all groups that were evaluable at this point was 7.4 months, and the 12-month PFS was in the low to mid-forties.

Once again, we're seeing that as a function of better features, like favorable versus intermediate versus poor PS of zero versus less than zero, the PFS is better in the better risk group. These, I think, are the highlights that I wanted to show you here. You can see that this is an active agent. This is a heavily pre-treated group of individuals. We are seeing this efficacy signal.

Pedro Barata: Thank you for sharing this with us. Very, very interesting data. I think one of the things too that folks are now asking regarding Belzutifan is whether or not it's possible to identify a biomarker. You played a big role in the VHL disease in that context. And some folks ask about somatic mutations of VHL. Does that matter at all? Kind of interesting to hear your thoughts on that. And then I'm going to also ask you a little bit about the safety perspective. If I recall correctly, this is also very well tolerated. The safety profile of this compound was very good. Can you expand on that, Eric?

Eric Jonasch: Yeah, I'll talk about the safety profile first. This agent had the expected toxicities. We saw anemia as the most common toxicity. We also did see hypoxia as a not-so-common but still present toxicity. The toxicity profile is one that one would expect from this class of agents. For the large part, the agent was tolerable. With regards to biomarkers, you're absolutely right, Pedro. We don't yet have data on this study, but we're looking for it, nor are there data with other agents whether or not we should be selecting individuals that have some either somatic mutation in VHL or a functional inactivation in VHL. It stands to reason that those would be the ones most likely to respond. But having said that, we do see these trends with better outcome or better risk subgrouping performance status. We typically know that patients who lose VHL mutations, their disease ends up becoming a little more aggressive. So we anticipate that we're going to see that.

Having said that, we also have a subgroup of individuals in this study where they had pretty aggressive disease, but they had dramatic responses. I think we have to be careful when we do look for these biomarkers because I think there's a subgroup that has perhaps VHL-loss-dependent disease, but then others that have other types of addictions that may be favorable from a response perspective.

Pedro Barata: And what's your take on the mTOR group? Quite interesting because it is true that we've pushed the mTOR inhibitors a little bit towards heavily pre-treated folks, right? But I'm thinking probably a lot of us are actually doing it, when possible, combined with lenvatinib, for example. I'm thinking of lenvatinib and everolimus, but perhaps some folks are using everolimus out there. So what's your take on that? I found it interesting that it actually seems enriched in that group of patients. Do you have any explanation for that? Any thoughts?

Eric Jonasch: No, and it could be a function of lines of therapy. It could be a function of some sort of interaction between the presence or absence of the use of this agent and the microenvironment. We do need to study it further.

Pedro Barata: Got it. I guess as I kind of think of the data you just presented and put that into context, of course, early trials, but still, I mean, 100 patients provide a very, very good data set. It appears that, exactly like you presented, about one out of five patients will respond, about one out of four or so will probably not get benefit, with PD as the best response. But then the duration of those who respond seems to be quite long, which I think is something we've seen already, and it seems concordant and consistent. And also, it sounds to me that with a favorable safety profile, we can think of combination strategies with other perhaps mechanisms of action. Is that your take on this? And when you analyze these different compounds with a similar mechanism of action, what do you see that can be differentiated? What can make the compound different when you're thinking and brainstorming about where we're going from here? What are your thoughts on that?

Eric Jonasch: Yeah, I think it's difficult at this point in time to say. There are multiple agents, as you know, that are in development. Do things like half-life and toxicity profiles differ sufficiently to differentiate one agent from the other? We don't know. And I think we're going to have to have maturation of data, both with this compound and with other compounds, to say, well, this is different in a favorable or unfavorable manner.

With regards to combinability, I agree with you that this agent—its toxicity profile—would make it amenable for combination. The big question really is: what should we combine it with? The effect of HIF-2 alpha inhibition on the immune microenvironment doesn't seem to be negative, but we don't know whether it's facilitatory either. We don't know whether combination with a TKI, sort of essentially a vertical augmentation of a vertical pathway inhibition strategy, is something we should be doing here. Whether or not there are other novel compounds or novel mechanisms of action we should be combining with this agent, these are all questions that we don't have answers to yet, but will likely get over the next couple of years.

Pedro Barata: Got it. Well, I trust you're going to play a major role in those. As I said before, every time you come up and present data, we all stop and learn a lot. I certainly do every time. So thank you. Well, this was wonderful. Thank you so much for taking the time to meet with us and educate us on this. Very, very interesting. Again, congrats on your great talk and looking forward to getting updates on this data.

Eric Jonasch: Thanks a lot.