Zachary Klaassen: Hi, my name is Zach Klaassen, urologic oncologist at the Georgia Cancer Center. On UroToday, I'm delighted to be joined by Dr. Pedro Barata, who is a medical oncologist at University Hospitals Seidman Cancer Center in Cleveland, Ohio. We're going to be discussing his ESMO presentation and his trial involvement with OPTIC RCC, specifically looking at a novel RNA-seq biomarker. So Pedro, as always, great to chat with you on UroToday.

Pedro Barata: Thank you so much for the opportunity to highlight this work.

Zachary Klaassen: Absolutely. So we just had ESMO. Lots of great data. OPTIC RCC is very unique. So before we get into the trial, just tell us about all the work developed for this biomarker. Maybe just clarify what Cluster 1/2 is, because that's important for the trial.

Pedro Barata: Yeah, it's a fantastic question. So as you alluded to, there has been an emergence of RNA-seq-based signatures, actually across solid tumors, but in GU tumors, particularly kidney cancer. The best work out there has probably been around the combination of atezolizumab with bevacizumab, first the phase II and then the phase III, the IMmotion150 and IMmotion151 trials. And so several clusters, which is basically a statistical analysis of how you see the gene expression grouped versus the others. So in that particular trial, and this is all published, we've seen a number of clusters, seven to be precise, where Cluster 1 and Cluster 2 are basically angiogenic. Then you have a few other clusters that tend to be inflammatory—Cluster 4, Cluster 5, if you will—and then a few others that are neither of the above.

So that data came up, and at least in that trial, the folks with the angiogenic signatures seemed to benefit from the TKI and folks with inflammatory signatures seemed to benefit more from the combination with immunotherapy, which would make sense. The challenge has been to extrapolate the different signatures from one particular trial and apply them to different other studies. So that's the story behind the biomarker-based approach. So currently we have a number of different IO-based combinations for metastatic clear cell renal cell carcinoma. We always have this ongoing conversation: should we think about a dual immunotherapy with ipilimumab-nivolumab, or a combination of immunotherapy with TKI—cabozantinib-nivolumab, lenvatinib-pembrolizumab, axitinib-pembrolizumab? And the debate is an ongoing debate because in fact we don't know who's going to benefit from what the most, and we want to get the most out of it, right?

So with that in mind, the PI for this trial, Dr. Brian Rini, out of a DOD grant, put together an investigator-initiated study or a signal-seeking study that basically asks the question, can I use gene expression signatures to select my treatment? So it's basically two cohorts. You do gene expression signature analysis, which comes out of NGS testing based on tissue, and I can walk you through the specifics afterwards. And if you have Cluster 1 or 2, you are assigned to IO-TKI cabozantinib-nivolumab. If your tumor harbors inflammatory signature, you are assigned to ipilimumab-nivolumab. And at ESMO, Scott Haake did a fantastic job presenting the data for the cabozantinib-nivolumab cohort, 27 patients. Out of a number of sites in the US, we had a meaningful contribution there. Obviously Vanderbilt also had a meaningful contribution. Other sites helped out. And Katy Beckermann, Brian Rini, Tian Zhang, Scott of course, we at University Hospitals/Case Western Reserve, and then we have a few other investigators who were fantastic playing a role. And so that's basically the data we've seen, and very provocative data I would argue for this proof-of-concept study.

Zachary Klaassen: It's awesome. I think this is just a beautiful example of taking a signature and using it to inform trial design going forward, right? And you laid out perfectly how that was designed. Tell us about some of the high-level results presented at ESMO.

Pedro Barata: Yeah, sure. So first of all, just to say we were expecting to see more patients assigned to Cluster 1 and 2 than we have for inflammatory signature, for example. So that's a reason why the cabozantinib-nivolumab cohort grew faster. You expect to see them in about or close to half of the patient population. The results were quite striking. So basically a response rate over 75%, which is really impressive. Most patients do have solid tumor shrinkage, zero progressive disease, zero patients progressing as best response. So at some point we had a good number of patients treated, and at some point this gave us such a level of comfort that when I had a patient interested in OPTIC, and I saw the results of the gene expression signature, I said, "Listen..." They're asking, "Doc, how likely is it going to work?" I was like, "Listen, so far I have no one where this treatment did not work, so it's going to work." The question is for how long? We have not seen that data in terms of progression-free survival. We don't know that yet, but I can tell you a lot of my patients are still ongoing on treatment.

Zachary Klaassen: Wow. That's great.

Pedro Barata: So I do anticipate the median progression-free survival will beat what we've seen in CheckMate 9ER with cabozantinib-nivolumab. Those are my words. We don't have that data. So I'm just saying patients are doing well. So that's one. The other thing that I think is relevant from this study I would like to highlight is the learning process from getting tissue sampling to actually getting patients on study. And Scott did this amazing job showing us the breakdown. Initially it was taking about a month or so, and actually that number came down to almost three weeks, right? About three weeks. And that basically means you need tissue from a metastatic site because they don't necessarily match with the primary tumor. So we started biopsying metastatic sites, you send it out to a genomics company—Tempus was collaborating with the study—and then you get the results back, and then it's standard of care. Cabozantinib-nivolumab is standard of care or ipilimumab-nivolumab would be standard of care, and then patients start on treatment.

So it's proof of concept that actually doing this is feasible. And as a treating provider, I actually felt very comfortable offering treatment based on the gene expression information that basically allowed me to walk the patient through what I'm expecting to see. And so from that perspective, it's very provocative, very contemporaneous with clinical applicability. So we'll see where we go with this. And of course we keep going with the study ongoing. And I think longer-term results will be critical in terms of progression-free survival, is it possible to achieve longer remissions, maybe even cures? Those are important things that we have to tease out as time goes by with more mature follow-up data. And then of course also the overall survival. So again, being not a large study, but I really think it fits the purpose, and kudos to Dr. Rini and of course all the folks involved in the study who were able to leverage it.

Zachary Klaassen: Yeah, great. And you're absolutely right. I think when we saw that data, 27 patients or so, but the signal was impressive. And I think it kind of leads me to my next question, and you already alluded to it a little bit. It's going to continue enrolling, Cluster 5/6, Cluster 1/2. Let's move forward and say this informs further trials. How will this potentially shape the conversations with the patient when they show up with metastatic RCC?

Pedro Barata: Right. Yeah, I do think that right now tumor sequencing is not standard of care. In other words, many patients are not being offered that because we have been lacking, to be honest, treatment implications, right?

Zachary Klaassen: And we don't even have data until now to say, "Hey, we should be doing this on everybody."

Pedro Barata: That's exactly right. And so we do it for research purposes. Many others do it for probably similar reasons. So I think, and by the way, doctors tend to behave based on therapeutic implications of their management, right? So if we see the first therapy or approach being available because of something that we found at a genomic level, at the molecular level, I think that's going to be a big incentive, motivation to actually change practice. So I do think we have to validate the concept that, number one, is this going to be the signature or are there going to be others?

The signature needs to be validated beyond a study where it has been developed, because it's almost like correlative data out of a trial. So the validation piece is a complicated piece because there are many signatures out there. So that's the first step. And the second is of course running a study where doing that is the right approach. So I think this is a proof of concept. It makes a lot of sense, but as you can imagine, Zach, we have to prospectively validate the approach. And I probably predict we'll need a larger effort to validate that approach. And the setting, we'll have to think about how that looks. Is it going to be within a cooperative group or not? I don't think we're there yet because I think it's also important to understand with ipilimumab-nivolumab, do we see the same kind of signal, the same strong signal?

And as I said, we need a little bit more longer follow-up to understand endpoints that matter, like the amount or duration of tumor control, longer remissions, overall survival—I think those will be very important. And the last point I'm going to make, we are talking about a selective group of patients, as you can imagine. Because if it takes you three to four weeks to get the genomic information to make a decision, unfortunately, not all patients have the luxury of starting treatment within three to four weeks after coming to see us. So it's probably not for everybody. But at the same time, genetic testing is getting faster as time goes by. So I actually think the turnaround time to get information is going to get shorter. So I think this is definitely the future, and I think we're going to walk toward that goal of making smarter decisions based on molecular characterization of the tumors.

Zachary Klaassen: Great discussion, as always, Pedro. We're looking at perhaps precision medicine in mRCC, which is exciting. Again, lots of work to do. We'll certainly look forward to seeing the nivolumab-ipilimumab results coming down the pipeline, further follow-up from Cluster 1/2. OPTIC RCC, we'll be looking and talking more about it, for sure.

Pedro Barata: Wonderful. Thank you for the opportunity. Kudos to Dr. Rini, Dr. Haake, Katy Beckermann, and all the teams of course. And we're happy to be part of it. So thank you for the opportunity to highlight the great work.

Zachary Klaassen: Of course. Thanks for joining us on UroToday, Pedro.

Pedro Barata: Thank you.