Read the Full Video Transcript

Pedro Barata: Hi there. I'm Pedro Barata, a genitourinary oncologist at University Hospitals Seidman Cancer Center. I'm very, very happy to be joined today by Dr. Brian Rini. He doesn't really need an introduction at this point, but Dr. Rini is the Chief of Clinical Trials and Ingram Professor of Cancer Research at Vanderbilt Ingram Cancer Center in Nashville. Brian, thank you so much for joining us.

Brian Rini: Thanks for having me.

Pedro Barata: Right. So congrats, great presentation as usual at ESMO in Barcelona, and a wonderful overview and update on the very important—to me at least—Phase 3 LITESPARK-005 study basically comparing belzutifan against everolimus for patients with refractory clear cell RCC. So we've seen that data before and I think it's always important to go over follow-up data and kind of tease out some of the important details that I think you presented. So maybe it would be a good chance. Do you mind walking us through the highlights of the data you presented?

Brian Rini: Yeah, sure. Thanks, Pedro. Yeah, so I'll give you just sort of a brief update. As you mentioned, this was LITESPARK-005. This was the Phase 3 trial that led to belzutifan's approval December of '23 in the U.S. and now sort of increasingly around the world. The initial data was presented at last year's ESMO, the publication actually just came out I think a few weeks or months ago. And then these data, sort of the final analysis and update of all the clinical endpoints and toxicity and such. I think most people know belzutifan's an oral HIF inhibitor. It had some single-armed Phase 2 data in refractory kidney cancer across multiple trials, response rates in the 20%, reasonable PFS, although single-armed studies and good tolerability, and all that data collectively led to this registration study.

So this is just a reminder of the study schema. Again, it was refractory kidney cancer that had received prior PD-1, PD-L1, and VEGF-TKI—at least one of each alone or in combination. So most patients were third or fourth line, as you would imagine in other sort of standard eligibility, patients stratified by IMDC and number of prior VEGF-targeted therapies. And then about 700 patients randomized to either belzutifan or everolimus at the sort of standard dose and schedule. This was a dual primary endpoint, PFS, which had been reported, and then OS, which was updated this year at ESMO. And then some key secondaries about response and duration of response.

So one of the primary endpoints was progression-free survival, shown here. These data are updated but almost identical to the original presentation and publication. A hazard ratio of 0.75 favoring belzutifan. There's somewhat interesting curves in that the medians are identical. You see those curves are largely overlapping. Through about six months or so, the medians were 5.6 months in both arms. But then you see this interesting separation of the curves. At 12 months, there's twice as many patients progression-free on belzutifan, and at 24 months, although the absolute numbers are reduced, of course there's over four times as many progression-free.

So it speaks to a few things. I think that there's likely a subset that's highly sensitive to this molecule. We don't yet have biomarkers, we didn't look in this effort, but those analyses are ongoing and hopefully we'll have some data soon, and that it's tolerable long term. We knew this from the VHL syndrome population where it's been a transformative drug in my opinion, and I think we're starting to see some of that long-term tolerability in this population.

These are the primary OS data. So these data were immature at the time of the first analysis and publication. They're now mature. You can see about two-thirds of patients with an event. And although the median favored belzutifan and the hazard ratio is below one, the p-value is not significant at 0.18. It's been hard to show a survival advantage in a refractory population like this. This was not a crossover study, but many patients, as you can imagine, received available subsequent therapies. So no OS advantage, but a PFS advantage, and I don't have slides, but a response rate advantage about 23% versus 3%, including some complete responses and a duration of response advantage as well.

This is just looking at the AE profile. So in the first column lists the common adverse drug reactions/toxicities, you see anemia being the most common. Hypoxia, some dizziness, fatigue, et cetera. In general, really well-tolerated drug, although there is a subset who won't tolerate it, anemia being the most common. This was a slide on first onset, and you can see the vast majority of these are pretty much right away. So when I'm starting somebody on belzutifan in practice, I tend to see them monthly. I might see them at two weeks if they're a little more frail or I'm a little more worried or at least have my nurse or nurse practitioner check in with them.

But the side effects tend to be right away. There was another slide not shown here about duration. And the duration tends to be relatively short as well with the exception of anemia, which is pretty easily managed with growth factors, although that can take a while to sort of kick in, if you will. So again, I think these data reinforce the long-term tolerability of this drug. I know I use a lot of this largely third line in my practice after an IO-TKI and then a second-line TKI. So I think an important update, and I think really just reinforced the primary data as presented and published.

Pedro Barata: Fantastic overview. And of course, I was kind of thinking two or three questions that came to mind when you presented this at ESMO, and same now, and you kind of already touched on those, but let me kind of structure that maybe two or three questions for you, Brian. So one, since you end up, let's get with the last one, which is the safety piece and the early onset AEs. So one of the things we've heard in our interactions with folks in the community, et cetera, is how much of a learning curve this drug requires in terms of some people like, hey, do I need to start checking sats, right? Which I've not used to do that for other agents or even EPO thinking about that there's some real-world data suggesting it's safe to do it and doesn't appear to impact outcome. So what is your comment regarding, does it exist a learning curve with belzutifan or not really? What are your practical tips for our listeners out there in the community about using this therapy?

Brian Rini: Yeah, I mean, I think any new drug has a learning curve, even a well-tolerated one, right? Because there are some unique side effects. Hypoxia is unique. I think it's fairly idiosyncratic, although you can imagine patients with chronic lung disease and COPD, et cetera, are going to have more of that. And we've certainly seen that in practice. I do have patients buy a pulse ox. You can get them on Amazon pretty cheap. And so a lot of them have them from the COVID days, but I say, yeah, go ahead and use that. Because when they call into my nurse, we want to know is their O2 sat 96 or is it 86, right? Those are big differences and we'd manage it differently, of course.

I don't really check routine EPO levels if patients get anemic. It's so predictable. They start at a pretty normal level and then they kind of drift down after they start the drug. And so it's pretty obvious what it's from. It's certainly reasonable to rule out other causes of anemia as you would in your normal practice. And obviously kidney cancer itself can cause anemia. And then I tend to institute growth factor pretty early and do it a little bit as needed. Meaning somebody drifts down sort of into the nine, below nine range. I'll start some form of red cell growth factor, which is often dependent on how often can I give it? How far away from my center are they? Sort of practical considerations. And then I'll give them a few doses, they drift back up and then I'll often stop. And then they'll drift back down or I'll start again. So that tends to be how I do it.

I haven't given a blood transfusion for belzutifan-induced anemia in a long time. I think that should be relatively rare if you are comfortable using growth factors. And there's data around sort of misuse of growth factors from years ago and risks and stuff. But this is a pretty refractory solid tumor population, right, with a relatively short survival. So I'm a pretty liberal user of growth factors. But to answer your question, there's definitely a growth curve. There's always a growth curve and nurses get nervous when somebody's pulse oxygen is 86%, and rightly so. So I think it's educating not just the docs, but the staff as well.

Pedro Barata: Right. Yeah, no, that's a fantastic point. I agree fully. So you also touched base on that point around survival. So it's quite interesting, right? The debate about endpoints, and of course we want to see overall survival. We've seen this across GU and probably if you go beyond GU. But on RCC, it appears that we've been accepting for a long time that we see clinical benefit, even in case we don't have a survival benefit. That's not exactly the case in other tumor types like for prostate. But in kidney cancer, it appears that we do accept that.

But at the same time, it's interesting to look at data not just from belzutifan with other agents that got approved without an OS signal, even when we're considering late refractory patients, if you will, and where we look at real-world data where most patients are not around to get five and six lines of therapies, right? Four or five, six lines of therapy. I guess you did mention potentially crossover. International trial, I wouldn't imagine a lot of people got access to HIF inhibitors upon progression in a lot of ways.

Brian Rini: No.

Pedro Barata: So what is your explanation? And it's not just belzutifan, it's a broader conversation around. Because people in the community would expect to see a signal. We're helping people, we're delaying progression. Why aren't they around longer? How do you take that overall?

Brian Rini: Yeah, I mean, I think if you sort of take a sober look at the data, right? There's a PFS advantage, but the hazard ratio is not 0.4 or something, right? It's 0.75, the medians are identical. You're helping probably dramatically a subset, but that's not enough to move the survival needle for the ITT population. It is refractory again, third and fourth line, although they didn't get crossover HIF inhibition, there's plenty of TKIs available which can have activity fifth, sixth line, et cetera. So of course we want survival benefits, right? That's not the question.

I guess if you look at it the other way and say, would I rather have this agent available in my practice or not given the data and its limitations, the answer is absolutely right. I think if it was TKI number eight, that was really toxic, we'd say, this doesn't really add much, right?

Pedro Barata: Right.

Brian Rini: We'd say marginal PFS and response rate, but it's really toxic and no OS, and I don't think anybody would be too excited. So I think it's novel mechanism and tolerability that give it a little bit, I don't want to say have a pass on survival, but maybe put the totality of the data into context.

Pedro Barata: And not just focus on OS. I agree. Final question for you. You did mention that the curves don't look like we normally see them or at least suggest there's something there about that subgroup of patients, right? Exact medians that you risk reduced by like 25%, which I think a lot of the comment when the curves were presented initially, and then when your data, when you showed the updated data was like, oh my God, we need a biomarker. We need to find out who the tail of the curve patients are. And we think back on the Phase 1 where a lot of the patients were without progressing for over a year, right?

So are we close to get that? Some folks say, well, maybe we'll get a somatic VHL, but then there's the comment, well, we see a lot more than 30% VHL mutations. Is that about a little something else? What is your general comment about if you had the power to how kind of biomarkers I can come up with? What would you focus? Would you focus on genomic alterations DNA level for your VHL or RNA-seq, something else? What are your thoughts?

Brian Rini: I mean, it's a big question. I thought we were close to a biomarker in kidney cancer 20 years ago. So that shows you my ability to prognosticate is poor. It's clearly not DNA alterations because as you say, the majority of clear cell—it should be 100% of clear cell—has VHL mutation, but we're seeing a subset of whatever, 20, 30% depending on where you look at the landmark. So I'm not convinced it's going to be DNA-based. I'm more of an RNA guy when it comes to biomarkers.

There are some HIF signatures out there, but they've been really tested in cell lines, not yet in patients. I know Merck is doing biomarker work on this trial, not yet reported, but hopefully in the next year or so. And then I think as you well know, biomarker work is really hard. And so it's going to take additional prospective studies. We're thinking about such a study using an RNA-based HIF signature to look at enriching, but it's frustrating, because you know it's there. You just can't find it. And so you give it to everybody knowing full well that there was a fairly high rate of primary progression in the study.

Pedro Barata: A third.

Brian Rini: And so another thing we did is try to look at the patients who were on that tail of the curve. There were like 54 patients, and it was just a clinical look. It was post hoc and retrospective. And not surprisingly, they had more favorable risk, lower tumor burden, fewer sites of disease, more nephrectomy. So the kind of patient you might expect would do better with a drug that maybe acts a little more slowly and isn't as cytotoxic, if you will, as TKIs.

Pedro Barata: Right. So Brian, I cannot resist to ask you. There's a question comes up, and I promise you now is the last question I have for you. Folks ask, okay, if we are going to do this exactly like the peer population that you just described, what do I do later on? I know there's not a lot of data out there what to do beyond belzutifan. Do you have any practical tips about, do you have any recommendation? What are your thoughts? What are you doing for patients progressing on belzutifan?

Brian Rini: For me, in the third-line setting, it's kind of belzutifan or tivozanib, right? Those are the biggest data sets. So if I use one third, then I use the other fourth. So I guess thinking about it—I hadn't thought about it—but thinking about it, tivozanib would probably be my go-to. Is there a ton of post-Belzu experience? No, it's just, it's available, there's Phase 3 data. It's really well tolerated and I'm familiar with it, and that's sort of good enough. But if you said you wanted to use pazopanib, okay, I don't know. You know what I mean? Nobody knows. A lot of drug development and drug use is a bit arbitrary about how the trial was done and things like that, as you well know. So there are other HIF inhibitors being developed. Whether they can be used sequentially like VEGF inhibitors is a big question. Not today, but I think that'll be a question. Mechanisms of resistance we have very little insight into. So this is the first chapter of this HIF inhibitor story, not the last.

Pedro Barata: Wow. Wonderful way to wrap this up. Thank you so much. This was a fantastic conversation. As usual, I appreciate you taking the time. I'll see you soon.

Brian Rini: Sounds good.