Read the Full Video Transcript

Tanya Dorff: All right. Thank you to the organizers for having me here. I will talk a little bit about sequencing. The good thing is, there's not a lot of data, so it's really mostly just some thoughts. So as a member of the NCCN Prostate Guidelines Panel, I have to say, we hate our product, generally speaking.

Maybe not everyone on the panel, Dan, but we recognize this is not super helpful, especially when it comes to what's an optimal sequence. And yet, it's really important that we be able to individualize, because each prostate cancer patient really does behave so differently across their journey.

And so our goal here is really just to empower clinicians to know there are lots of choices and be able to try to select the right one for their patient in that moment. And we look at things like, what will be most effective. So ideally, that would be biomarker-selected, but we have very little of that. What side effects are there? What is the patient willing to tolerate in this moment in their cancer journey? And we're starting to think about cross-resistance, although I would argue it's still most important to choose what will be most effective in this moment when we can know that and when that's possible.

So I would argue, and many of my colleagues, I think, agree, that the one thing we know is PARP inhibitors work really well if you have a BRCA2 alteration, and you want to use it earlier rather than later. So TRITON-3 really showed this, because there was head to head against docetaxel, which is rarely done. And the patients did better with the rucaparib.

So in a patient with a BRCA2 mutation, I'm going to try to choose that PARP inhibitor as early as possible. Now I know there's some concern about cross-resistance, but that may not be due to the PARP inhibitor pretreatment. It might be due to the biology of the actual HR alteration.

And these patients just don't do as well with our standard treatments. You'll see that the docetaxel RPFS here is shorter than what we saw, for instance, in an unselected population. So we know that, as we saw some of the genomic data presented earlier, we're in our infancy of understanding which genomic changes predict for benefit versus resistance to lutetium PSMA-617, but BRCA alterations do seem to be associated with less benefit.

Now, that doesn't mean I'm not going to use this radioactive treatment in patients with BRCA alterations. I just might sequence them in later. And in this experience, AVPC did not predict for lower likelihood of benefit, which is interesting, but I think we need obviously a lot more data.

So we've been talking a lot about PSMAfore, and I think the panel earlier did just an outstanding job of thinking about some considerations of who should really get chemo first. But is there a rationale? Like, does this drug work better before chemo versus after?
You could argue the RPFS in this trial was longer than it was in VISION. The objective response rate, though, was fairly similar. But I do think the disease tends to get more and more heterogeneous as it progresses—maybe less PSMA expression, less androgen receptor-driven. And that would be another reason to think about using the PSMA-targeted therapy earlier rather than later.

On the other hand, in terms of lutetium versus radium—and we do have some early sequencing data—we know that it can be safely given after radium. We don't have a lot of data yet for radium after lutetium PSMA-617, but anecdotally in my clinic, I've done it, and there have been some abstracts showing that it's safe. But judging efficacy of radium will be very hard, because that's hard to do as it is. But we do know, at least from this small experience that was published, that we can give the lutetium after. And I don't know, philosophically, what makes more sense.

Sometimes I like to think about using radium when a patient is bone-only, and we know that maybe they'll develop visceral mets later and then radium comes off the table. But it also could be seen as maybe consolidating after you get rid of some of the more visceral or aggressive clones. So I think we really don't know, is the bottom line, what's the optimal sequence.

Because of TheraP, I think a lot of us would choose lutetium PSMA-617 before cabazitaxel. And that's because it had greater PSA responses. Also greater pain responses. So when the panel was saying someone's very symptomatic, you might want chemo, I've actually seen really gratifying pain responses in my patients, sometimes after the first dose—even like a week in—with lutetium PSMA-617.

So I think we still have to know it when we see it. That patient who really needs chemo—I think visceral could be a very good example of that. But pain is not necessarily a reason that I would choose cabazitaxel over lutetium PSMA-617. And the real nice advantage here is that there's a lower rate of serious adverse events, like febrile neutropenia. So most patients are going to prefer getting the radioligand therapy.

Even though the RPFS was similar, it was definitely shorter in this trial. This was a different population. And I think when we're talking to patients about it, we have to remember that if we're going to say that this was more effective than cabazitaxel, these were patients who were selected using different PSMA PET criteria, not your VISION criteria. This group also had a high rate of pain progression, so they were just a poor-prognosis group, and that probably is reflected by that shorter RPFS that was seen in TheraP compared to VISION.

So what does optimal sequencing look like in mCRPC? We really don't know. We need a lot more data. But I would argue that we should biopsy when there's progression when at all possible, especially visceral. I think there's a lot of questions about if you see neuroendocrine differentiation or AR indifference emerging, even if their PSMA PET is positive, would that be something that would be associated with a lower likelihood of benefit from PSMA-617? So I think getting that tissue is really critical.

We also have to remember to repeat our NGS. Now that can be done from this new biopsy. It can also be done by ctDNA. But we know that actionable mutations do arise as patients are exposed to treatments, and we don't want to miss those opportunities to include things like PARP inhibitors or pembrolizumab if we find MSI or high tumor mutational burden.

I think for symptomatic rapid progression, I will reach for either chemotherapy or PSMA-617. I think the PET scan—really, what we need is to use our PET scans as a better biomarker in optimally selecting patients for whom this is the right treatment now, and it should be prioritized over the other options in that big boxy list from NCCN.

And there are lots of other things we do. We don't always have to jump for those really slow progressors, asymptomatic. We still do have things like immunotherapy with sipuleucel-T. We have clinical trials, and we want to try to line up as many options for our patients as possible. It's not a race to the finish. It's not a race to get to our radioligand therapy. We really want to make sure we're individualizing and using as many options as possible along the way.

Thank you.