Read the Full Video Transcript

Emmanuel Antonarakis: So my talk is unaltered between two hours ago, when the FDA made their announcement. So I'm also not going to be disclosing any unpublished information about ECLIPSE. So if you were expecting that, you can go for a break now because we're not allowed to do that until it's public. So we'll just speculate about ECLIPSE.

Same disclosures. The particular ones are Curium and Novartis, which are relevant here. So if you look at the NCCN guidelines, and you create a two-by-two table of the metastatic CRPC disease spectrum, you've got post, pre-ARPI, post, prechemo. The bottom right-hand corner is the indication, as of yesterday, for PSMA radioligand therapy—Pluvicto.

And the goal that was, apparently, achieved today, without me knowing about it, was that, at least, we have one, at the moment, radioligand therapy, that has, as of today, become FDA approved in the post-ARPI, pretaxane setting. So these are the three trials that have been conducted and completed. The first one and the second one are presented or published to some degree.

The third one, ECLIPSE, the information from that is through publicly available sources, such as clinicaltrials.gov. We did have a press release from ECLIPSE in November, about four months ago, that the trial was positive, meeting a statistically significant and clinically meaningful RPFS endpoint. But the hazard ratio, and confidence intervals, and p values have not been disclosed, so please do not ask me what they are.

Now, there are some key differences here. These are all prechemotherapy, post-ARPI trials. They're all randomizing the PSMA radioligand against an ARPI switch. The PSMAfore trial had a 1-to-1 randomization. The other two trials had a 2-to-1 randomization favoring the radioligand arm.

There were critical differences in dosing regimen. I think that's the most important thing about this slide. So PSMAfore used six doses of 7.4 gigabecquerels each. SPLASH used a much lower cumulative dose, at 6.8 gigabecquerels times 8—so it was stretched out—plus maximum of four doses.

And then ECLIPSE—this is public information—was initially designed with four doses of 7.4 gigabecquerels. And then halfway through the trial, it was modified due to dosimetry results from the kidney that looked safe. And about half of the patients in ECLIPSE are available to get up to six doses. We don't know how many of those did get six, but we can expect that about half will have gotten four, and half will have gotten six.

Now, we don't have data on the demographics on ECLIPSE, but we do for PSMAfore and SPLASH. Of course, SPLASH is the PNT-2002, also known as INT. It's the same compound. PSMAfore is 617. So they are different drugs. You can see that the US versus European distribution was the opposite.

So PSMAfore was 80% EU, 20% US, whereas SPLASH, the PNT-2002 trial was 80% US, 20% Europe. ECOG status for both trials was about 50% to 60% ECOG 0 and 40% ECOG 1. The SPLASH has not disclosed the number of patients that had previously received enza versus abi. In PSMAfore, it was about half and half.

These are the results shown in a table. Medical oncologists get accused of showing lots of Kaplan-Meier curves. And I didn't put any Kaplan-Meier curves in my talk, but I put tables instead because you can get more information into one slide. So if you look at the PFS and OS, first let's focus on PFS so that the hazard ratio in PSMAfore for PFS, this is the final PFS analysis.

Hazard ratio is 0.49. The first time the RPFS was presented, it was a hazard ratio of 0.41. Second time, it was 0.43. And the third time, it landed at 0.49. So 0.49 is the most accurate one. It kind of got, quote, unquote, "worse" over time with more events, but that's the final hazard ratio.

With SPLASH, we saw a hazard ratio that was statistically significant, but it was 0.71. So numerically, it looks to be more modest or worse. And then confidence interval there does not include 1, so it is statistically significant. Now, looking at OS, which is the next row there, the final OS analysis of PSMAfore—this is public—the hazard ratio was 0.91.

Many of you might remember when this was first presented at ESMO 2023 by Oliver. The hazard ratio for OS at that time was 1.16, and everybody was freaking out. Is it making survival worse? Over time, with the second and third survival sweeps, now, the hazard ratio for OS in PSMAfore, which is the data the FDA used to make their decision, was 0.91—not statistically significant, crosses 1, but is numerically lower than 1.

With SPLASH, we just have a single analysis so far of overall survival, which is immature. There will be a final one coming later, which we have not seen. But the first hazard ratio for OS was 1.11. Objective response rates were a bit higher with PSMAfore, 50% versus 38% for SPLASH. And that might be a dose-intensity effect because of less activity delivered in SPLASH relative to PSMAfore.

And these are the dosimetry results in the two trials. If you look at the second column and the fourth column, so gray per gigabecquerel per dose—that's a mouthful. If you focus on kidneys, you can see that gray per gigabecquerels per dose for kidney dosimetry was 3.1 for 617 and 4.9 for PNT-2002. So there appears to be more dosimetry dose going to the kidney in PNT-2002 relative to 617.

However, the lacrimal and salivary dosimetry results look numerically worse. There's higher dose being delivered to lacrimal and salivary glands with 617 as opposed to PNT, and the rest of these are not significantly different. So more dosimetry to the kidneys with PNT, more to the lacrimal and salivary glands with 617.

Treatment-related adverse events. Grade 3 was about 30% for both studies. Serious treatment-related AEs was 20% and 17%, so these are very comparable. The difference is, when it comes to the dry mouth, dry mouth was 58% in PSMAfore with 617 and 37% in SPLASH with PNT. So more salivary toxicity.

The rest of these look relatively comparable. I'm sorry that the table is so busy, but you can look at it carefully, later on, if you want. Fatigue, interestingly enough, was 54% in SPLASH, 23% in PSMAfore. I'm not sure if that was just being measured more frequently or often, but maybe there's a signal with more fatigue with SPLASH, although it's hard to understand, biologically, why that would be happening.

We do have the Telix trial which is in the same space. It's a post-ARPI, pretaxane CRPC study. The dosimetry part, on the left, part one, is completed. The part two is the randomized part. The difference about this study is it's a radio antibody, not a radioligand.

Secondly, the control arm is best standard of care, but it includes a taxane. So you can choose ARPI in the control arm, or you can choose docetaxel. And then the interventional arm is not the radio antibody by itself versus standard of care. It's in addition to standard of care.

So both groups have the standard of care. Physician has to decide ahead of time what they're going to choose, and then they have to stick with it. And then in the radio antibody arm, they get the addition of the rosopatamab tetraxetan. There. That was hard to say.

Now, one other difference is the dosage and administration and cumulative radiation exposure. So with the radio antibody, you get two doses, and then you're done. The doses are delivered two weeks apart. So within a 14-day period, you are done.

The cumulative radiation exposure with two doses of the radio antibody is 152 millicuries, comparing to four doses of PNT-2002, which is 736 millicuries, compared to six doses of 617, which is 1,200 millicuries. So, again, no results on this one. Time will tell.

Thank you very much.