Pedro Barata: Hello, everyone. Welcome to another conversation here during AUA 2025. I'm Pedro Barata. I'm a GU oncologist out of University Hospitals, Cleveland, Ohio.

I'm being joined by the great, one and only Dr. Scott Tagawa, director of the GU research program at Weill Cornell in New York. Scott, amazing to have you here again. Thanks for coming.

Scott Tagawa: Thanks. Always glad to be here.

Pedro Barata: Absolutely. And we're here to chat a little bit about some of the work that you actually presented at AUA in our oral presentation, again, focusing on radioligand therapy with your actinium PSMA, in this case, really looking at patients who have received prior lutetium and then were exposed to your therapy. You have a huge amount of experience handling these and different RLTs over the years.

So perhaps I'll ask, I don't know if you want to quickly recap what's unique about your compound that you presented here at AUA. And tell us a little bit about the highlights of the patients who were treated with these sequencing, the different radioligand therapies.

Scott Tagawa: Sure. So this was an analysis-- so three separate prospective trials, a post-hoc analysis. The specific drug is actinium J591. J591 is an antibody called rosopatamab now against PSMA. It was actually the first one that targeted the external domain of PSMA and is internalized and kept inside the cell.

This is radiolabeled with actinium, which is a potent alpha emitter, emits four alphas, a high linear energy transfer, short range compared to betas overall. The hypothesis is based upon the fact that, generally speaking, any treatment, including PSMA treatment, at progression, generally the PSMA remains.

So loss of the target is not the main mechanism of resistance. So PSMA generally remains at least at progression. And because of the potency of alphas over betas, the hypothesis is that an alpha is going to work after beta.

There were three separate trials, two of which were agnostic to prior therapy. So prior lutetium, radium, et cetera was allowed, and we analyzed those with prior lutetium. And then one of them very specifically was a target radionuclide after any prior target radionuclide, most of which was lutetium.

So we analyzed those. And for AUA2025, about 40 patients, 37 patients that we looked at overall, none of whom were screened or selected based on PSMA expression. They got some form of actinium J591, and 41% had a PSA50, at least as best response, PFS of about 3.5 months, overall survival of about 10.5 months.

Multivariate analysis looked like PSMA expression by PET scan wasn't associated with response, but was associated with PFS. And then clinical factors that we expect, like CALGB, Halabi nomogram factors, are associated with OS in a multivariate analysis.

So basically, the summary is that this is in line with the hypothesis that PSMA targeting with an alpha is going to have some efficacy after a beta. Overall, these are mostly phase I dose-escalation studies. Overall, it looked like it was safe across all grades, mostly grade 1.

Xerostomia was the highest in terms of frequency, but these are all patients that started with some-- mostly it was grade 1 at baseline and stayed grade 1, but just got a little bit worse. So we called it grade 1.

Pedro Barata: OK.

Scott Tagawa: And then, in terms of high grade, not unexpectedly with an antibody, high-grade myelosuppression. But generally, a small percentage is for a short period of time, so a temporary myelosuppression that we would expect.

Pedro Barata: Got you. So as you probably predict, my next question, it's such a hot topic, right? We're just coming from hearing that there's regulatory approval for moving lutetium PSMA, also known as Pluvicto, early on pre-chemotherapy. The expectation is more patients are going to be exposed to lutetium PSMA.

And so questions like, should you bring in an alpha particle afterwards would be very important to understand, right? I think when we look at the drug development of different sponsors, they're trying to understand what the signal is. Is it fair to say that keeping the same target, to your point, looking at PSA responses, which are actually in the 40% range, is quite attractive.

But it appears that it's almost not matching really the durability of those responses measured by PFS perhaps. Is it fair to say-- does that suggest that maybe some-- and those patients were not selected, I should say, by PSMA scan like we normally do for lutetium PSMA. Is it fair to say that perhaps that speaks to the heterogeneity of the tumors?

And even if we see PSMA expression, things like SUV mean or intratumoral or intrametastatic lesion heterogeneity of the target would explain why you might see some initial responses not translate into durable tumor control. Is that the sense that you're getting from the patients you're treating or no?

Scott Tagawa: So, yeah, I'd say overall, broadly speaking, I would say heterogeneity is the bane of cell-surface targeting. And that's why I kind of like combinations. This retrospective study included mostly prospective studies where patients got a single dose, not over time. So I think that maybe--

Pedro Barata: How much they got, Scott, by the way? Those--

Scott Tagawa: Most got just the one time done.

Pedro Barata: OK. How much--

Scott Tagawa: Oh, it was a range.

Pedro Barata: A range of different doses. OK.

Scott Tagawa: Off the top of my head, the lowest was, I think, 20 kilobecquerel per kilo.

Pedro Barata: Right.

Scott Tagawa: And we didn't hit an MTD in our single one. So 93.3 kilobecquerel per kilo was what we called recommended phase II dose for the single one. And that was the most common regimen in this retrospective analysis. So it is what it is.

Pedro Barata: Right.

Scott Tagawa: I suspect, and it looks like from AUA last year, that the fractionated dose, the PFS is longer. I think we spread out the dosing for longer. So that makes sense. And that's the regimen that's going forward with several different companies with PSMA-targeted radionuclide therapy. Particularly two antibody companies, Telix and Convergent, are looking at that regimen.

So anyway, but this in a more refractory patient population, patients that already had lutetium, yes, I would expect, number one, to have some inherent rate of resistance and maybe more heterogeneity. That hasn't been absolutely proven yet, but that does make sense that it's going to happen. The more heavily pretreated a tumor is, the more resistance and heterogeneity is likely to have.

Pedro Barata: Got you. And then the other question is, of course, just pulling up data. And PEACE-3 was presented, radium with enzalutamide, maybe for some survival, quite interesting survival, median overall survival in those patients over 40 months. I think actually it's one of the first times we saw that.

If you take the PARP inhibitor story, you put it aside with the combination talazoparib. But 42 months or so, something like that, is actually, you know, it's like nine months more than we would expect, up to three years for median overall survival in the CRPC setting. I'm curious to know.

We have data about safety of offering things like radium-223 prior to lutetium. I'm wondering, have we looked into that, of patients who have been exposed to radium-223 if it gets available in the US, combining with different things like an ARPI? Do you expect it to be safe to do another alpha emitter like actinium? Do you have any experience with that?

Scott Tagawa: So, yeah, off the top of my head, I'm going to say that-- actually, I would say our trials with actinium J591 are very similar to trials with lutetium, where about 20% were previously exposed to radium.

Pedro Barata: Got you.

Scott Tagawa: Now, to get into the trials, you had to have good organ function.

Pedro Barata: Yes.

Scott Tagawa: So we're excluding patients that had kind of crash and burn. I think that's more cancer than it is actually radium toxicity.

Pedro Barata: OK.

Scott Tagawa: But, of course, to get into a trial. But there's all the trials pretty much across the board or the retrospective analyses. It's something like 20%, sometimes a little bit less, sometimes a little more. And now, we're seeing the other direction.

So I think the PEACE-3 data are great. I think there's two problems. One is that they were generally, I mean, some people had abi, but it was generally ARPI-naive. And then across every radium trial, the phase III, whether it's PEACE-3, whether it's DORA, no prior lutetium allowed.

Pedro Barata: Right.

Scott Tagawa: And that's not today's world. And we'll see what happens with PSMAddition. But we've seen for sure that PSMA PET has totally disrupted the landscape. Moving lutetium early is going to do the same thing potentially.

Pedro Barata: Right, fantastic point. One final question around the target. You made a very good case about the target not changing. I'm personally not aware of a lot of data looking at things like SUV mean over time with exposures to different therapies.

We're working a little bit on that. I'm sure we're going to have more data. But I guess looking at the different targets out there, we have HK2, FAP, et cetera, knowing what you know today, how would you design it if you were to accept that, yes, most patients are going to be exposed to lutetium hypothetically, right?

Assuming they're radium or other RLT-naive and they got lutetium. They're coming out of that. Studies like yours are super important, because we need to understand how patients do afterwards. How do you see?

Would you still keep the same target? Would you think of different target to try to salvage some of the patients where the expression might not be the greatest, might measure by SUV mean or by intratumor heterogeneity? Would you try and get to different targets and keep the same actinium, for example, or what are your thoughts on that?

Scott Tagawa: So my thought, my practical thought is that PSMA is the first. It's the head. And I think it's likely that the next wave of approval of agents, of radionuclide agents is probably going to be PSMA, because PSMA is just ahead of the other ones.

Pedro Barata: Everybody else.

Scott Tagawa: I don't necessarily think that's the best. But one thing that PSMA does have is very available imaging. I think it is not to say crazy, because I've enrolled many patients and just treated them, regardless of the imaging results.

Pedro Barata: OK.

Scott Tagawa: But that was for phase I trials.

Pedro Barata: Yeah.

Scott Tagawa: I think all cell-surface targeting drugs, ADCs, et cetera, should have some imaging biomarkers that are embedded into the early development and to go from there. So we know HK2 is possible because the phase I, I'm sorry, the phase 0 was published.

Pedro Barata: Correct.

Scott Tagawa: But that was with MSK and it's not in the trials. But that's something that is missing. By the way, no prior lutetium allowed for that either.

But that, I think, is missing from there. STEAP1 has an imaging agent, or did, but is not part of these trials. So I think that's a problem. And so I can look at it for PSMA anytime I want because I have a commercial scan.

Pedro Barata: Because you have it available.

Scott Tagawa: The other ones, even if they're part of research, my bias is that these are important biomarkers to add in to be able to answer these questions.

Pedro Barata: Right, right. No, fantastic insights. Again, I consider your presentation is really, really relevant. Because a lot of people are going to be looking at it, I'm sure, with a very-- paying a lot of attention to it because that question is going to come up.

So we're really looking forward to the next steps of your research at Weill Cornell. Again, congratulations. Really well done. Beautiful presentation at AUA. And thanks for taking the time to chat with us today.

Scott Tagawa: Thanks for the invitation.