ASCO GU 2023
Latest in CRPC
Dr Neal Shore – Carolina Urologic Research Center, Myrtle Beach, USA
Dr Elena Castro – Spanish National Cancer Research Center, Madrid, Spain
Prof Eleni Efstathiou – Houston Methodist Cancer Center, Houston, USA
Prof Gerhardt Attard – University College London: Cancer Institute, London, UK
NS: Welcome everyone to another great programme by ecancer. We’re here today in San Francisco at ASCO GU 2023; we’ve had great presentations. I’m Neal Shore, I have the privilege of just moderating a discussion today with my wonderful friends and colleagues – Gerhardt Attard, renowned international medical oncologist from London; Eleni Efstathiou, a renowned medical oncologist from Houston, Texas; Elena Castro, a renowned medical oncologist from Madrid, Spain. It’s a great honour and privilege to be with you today. We’re going to get right into it. We had great presentations at ASCO GU 2023 this year; we’re going to review some of the takeaways and the controversies. We’re fortunate, we’ll begin with Elena Castro, she was the discussant where we had Alan Bryce present the top-line data from TRITON-3. We had Neeraj Agarwal present the top-line data of the phase III TALAPRO-2 trial and we also, later on in the programme, had Noel Clarke present the pre-specified final analysis from the PROpel trial. So we have three very important combination trials with PARP inhibition and additionally we have follow-up survival data on the MAGNITUDE trial that has been presented by Eleni Efstathiou. We also have additional poster presentations that Elena presented on what happens when we sequence as opposed to concomitantly starting patients with niraparib and abiraterone. So a lot to discuss, a lot to unpack. We want to really focus on the key takeaways, also talk about maybe side effect profiles – are they different among the PARP inhibitors. Then lastly I want to have Gerhardt weigh in on this because he has done some really tremendous research throughout his career on what’s the ultimate long-term effect on the bone microenvironment, particularly when we start patients earlier on with PARP inhibition. So let me turn it over to you Elena.
EC: Yes, as you mentioned, we have learned the results of two phase III trials that really help to fill in and complete the data we have from other trials on PARP inhibitors in prostate cancer. Very briefly, the TRITON-3 trial compared rucaparib with other standard treatments such as a second hormonal agent or docetaxel in patients with alterations in BRCA1, BRCA2 or ATM. Rucaparib resulted in improved radiographic progression free survival compared to either docetaxel or a second hormonal agent for BRCA patients but not for patients with ATM alterations. One of the most relevant things of this study is that it’s the first direct comparison of a PARP inhibitor with docetaxel.
NS: Which was great, wasn’t it? So what were some of the findings, just looking at the Kaplan-Meier curves of how rucaparib stacked up against docetaxel versus physician’s choice of abiraterone/enzalutamide in terms of rPFS?
EC: For patients with BRCA alterations rucaparib definitely improved the rPFS compared to any of them, either docetaxel or a second hormonal agent. For patients with ATM alterations we didn’t see any benefit so for these patients docetaxel could be as good an option as rucaparib after progression to the hormonal agent.
NS: So here’s a quick question: we’re all testers, we test, I know we test routinely, we test germline, we follow the guidelines, it’s almost universal testing, it’s not quite there yet. But we get germline testing in almost everyone with a prostate cancer diagnosis, certainly with metastatic disease and in many of our metastatic patients we’re getting somatic testing too. So if you have ATM, basically if you have that alteration is it essentially dead, not a consideration for PARP inhibition?
GA: We have that easy because in Europe the licence for olaparib did not include ATM, it was only BRCA. I think the data is now incontrovertible from PROfound, from MAGNITUDE, from TRITON-3, really elegantly shown yesterday, that there is no evidence of benefit using a PARP inhibitor in the ATM mutants. So although there was a signal in the TOPARP trial initially, really the justification for using a PARP inhibitor in that group is no longer there.
NS: So let’s switch, a lot of great findings in TRITON-3 – a unique trial design comparing directly with docetaxel which a lot of people have asked for.
GA: The first trial, the first trial to show superiority over docetaxel in twenty years.
NS: The first one, yes. So hats off to them, that’s a really important notation. But let’s go and talk, for purposes of time, what about TALAPRO-2? Neeraj Agarwal presented that data. So this is a little bit of what some would call a tiebreaker because in MAGNITUDE we saw no benefit in an all-comers population, we did see benefit in combination with the HRR mutation. But now we see an all-comers benefit on the rPFS primary endpoints comparable, some distinctions regarding PROpel. Eleni what are your thoughts regarding the rPFS data of TALAPRO-2 and the presentation?
EE: I’m going to resort to a compadre of mine, Greek, who is a statistician. You’ve probably heard of him, Dr [?], very controversial. But what we all agree is that reproducibility is key. So these three trials are non-reproducible, obviously differences in design, we were all contributing to one or the other trial. We know that when it came to MAGNITUDE there was a pre-specified futility analysis early on that just severed that HRR negative cohort. So it failed to launch, essentially, we said around 200, ‘We’re stopping this,’ right? So that means there may be a signal there but that signal in the HRR negatives is where we need to thread the needle. I think we would all agree that we cannot dismiss it, there is something, there is a subset that may benefit and you may have that synthetic lethality induction. How do we identify it? Do we need to go beyond a nine gene panel or a BRCA panel and identify through other methodology – radio tracers, other technology that would give us more of that information? I would actually start focussing on those patients in the HRR negative population that, at least in the control arm, seemed to have a more grim prognosis – they may be enriched – a way to go. We can’t answer that in daily practice so we remain with that question – should we consider the combinatorial strategy, if approved by all agencies, for HRR negative patients? I would argue I have no answer yet to that.
NS: So it’s almost on a case-by-case basis. Now, you had a really nice slide, I thought, in your discussion section where you said, ‘Okay, in terms of who is going to do better from a left-to-right viewpoint.’ Do you want to go over that, you know what I’m talking about? You said, ‘Clearly the most benefit would be in HRR positive…’
EC: BRCA better than HRR. We don’t know the benefit on non-BRCA/HRR, that data has not been presented. Those HRR do better than non-HRR and the all-comers population is in between the HRR and the non-HRR. But somehow we all expected that, right?
NS: Yes. I think we expected that but it’s important for the audience to see that hierarchical way of thinking about it because that really is the big elephant in the room. Do you just say, ‘Well, I can’t test, I don’t have access to testing, I’m going to do an all-comers approach’? Or what about the patients who I do test and who are not HRR and I want to be aggressive and do a combination approach, two novel mechanisms of action? But, Gerhardt, let me ask you, we know that there are, as a class, toxicity associated with PARP inhibition. I want to ask everybody to weigh in at some point if we think there are differences amongst the PARPs, but particularly regarding the issue around myelosuppression and anaemia and neutropenia, thrombocytopenia, how do you think about it in terms of the long-term impact on the bone milieu?
GA: I guess you’re specifically thinking, as we move PARP inhibitors to the mHSPC setting, and I presented data yesterday, or summarised data we presented at ESMO, that especially the low-volume mHSPC population have excellent outcomes. 40% are still in remission after eight years, so we need to think really carefully about treating with them PARP inhibition for this long period of time. The experience in the late-stage mCRPC is that these are manageable toxicities; we’re used to using chemotherapy so these are levels of myelosuppression we’re used to handling. But the long-term implications of longer use, which is I think, clearly, what you are highlighting we don’t fully understand. In fact, even in ovarian and breast cancer the long-term use of the treatments is not that long. So I think we need to be very careful and that really underscores a selective strategy for mHSPC. The all-comer approach was being explored in line one mCRPC and there’s going to be a debate on this; I would be really concerned about that approach in mHSPC unselected where there is going to be a large subgroup who have really excellent outcomes and, increasingly, will not die from prostate cancer.
NS: Yes, those are great points. The good news, at least for a lot of our colleagues who are uninitiated or lacking in experience in using PARP inhibitors, is that, at least in the first line mCRPC, the anaemia is manageable. You monitor CBC, the GI toxicity is manageable as well, and clearly there is a benefit for patients who are mutation positive. The rPFS data which now was rather robust, even a little bit more robust on the TP2 compared to PROpel, how do we think about this debate that we heard regarding rPFS as a surrogacy for approval whereas we had questions that came from the floor, I think Johann de Bono really brought it up, that was a controversy regarding other cancer types where rPFS is a surrogacy? But we don’t necessarily have that, clearly not, and that would be somewhat paradigm changing in prostate cancer approvals and acceptance, although EMA has now approved the PROpel data based upon the rPFS. How do you think about that, Elena?
EC: I agree that a benefit in rPFS doesn’t need to be translated into a benefit in overall survival. rPFS may be a regulatory endpoint but I’m not sure this is a clinically meaningful endpoint if that results in a similar overall survival or it could even be a shorter overall survival. So for patients who are HRR negative we need to see… well, for all patients but for HRR positive and BRCA it’s more likely to be positive but for the non-mutated patients I really would like to see more data about survival.
NS: So, Eleni, let me ask you. You had a poster and a presentation on some additional survival data from MAGNITUDE, can you talk about that and even talk about Noel Clarke’s presentation on the third data cut-off point on OS that he presented in PROpel?
EE: Yes, thank you for that. I think we need to take into account something that we all take for granted right here – the fact that the subgroups that we’re all focussing on, especially the BRCAs, have the grimmest of prognoses of the whole cohort. What is interesting is we went and looked specifically at the wildtype population from the legacy study COUGAR-302 and we saw for an unselected population on abiraterone alone it was rPFS 16.5 months. For the BRCAs on the niraparib trial, the abiraterone alone arm is 10 point something months, like you’re cutting it almost in half, just the rPFS. I expect the same to come when survival matures because we don’t have maturity. So if you take specifically into account the grim prognosis plus the fact that we know that in the BRCAs this pathway is the driver pathway of progression, not just carcinogenesis, we proved that even with the PROfound trial at a clinical level, then the associations are clean. It becomes more of an issue in the HRR negative, going back to my earlier comment. Having said that, we’re essentially looking at all the trials doubling the rPFS for the BRCAs and that’s what we showed yesterday with the update of MAGNITUDE. We went from 10 months to 19 point something months, that’s doubling. The same thing with MAGNITUDE and we’re going to see what evolves with TALAPRO. We’re not mature yet with events in MAGNITUDE by the IPCW analysis that corrects for baseline characteristics and subsequent treatment, there is a clear trend – 45% improvement in risk of death – when you look at that. PROpel, that is more matured, put it to bed when it comes to BRCA – they get the OS benefit and they meet very closely their overall threshold for secondary endpoints. So we’re very close to the point that PROfound showed before that, yes, in the BRCA subpopulation there is that surrogacy association; it’s the rest that remains to be determined.
NS: Yes, agreed. So let’s come up with a hypothetical, just brainstorming here. You have now, by EMA approval, you have combination now in first line; the combination for olaparib/abiraterone is before the FDA as we speak. Soon we’re going to have, presumably based on TALAPRO-2, we’ll have enzalutamide/talazoparib as another alternative; we may also have niraparib/abiraterone for HRR positive patients. So in a world where cost is not an issue and accessibility is not an issue how would you, Gerhardt, describe this to a patient who is either incapable of getting testing, so if it’s the all-comers population, or you get tested and does not have an HRR mutation and they’re first line mCRPC, they progressed from monotherapy, mCSPC, although we recognise that more and more patients will be getting couplet and triplet therapy, that’s a whole other discussion which we can have with ecancer, someone else will have it. But how do you have that conversation with a patient in my hypothetical?
GA: Yes, my view is that the data doesn’t support PARP inhibitor plus enzalutamide or abiraterone for line one mCRPC in the absence of an HRR alteration. I say that both based on the underlying biology but also based on the absence of a survival benefit in PROpel for that group yesterday. We don’t have that data for TALAPRO-2; we have that data for BRCA mutants in PROfound and in PROpel as well, that subgroup had a clear survival benefit. So there are risks – this toxicity we mentioned earlier – the toxicity is clearly outweighed by that benefit in the BRCA mutants, for the rest it’s not outweighed with the evidence available. So I’d stick with NHA, any NHA of choice.
NS: I appreciate that, I think you have a similar opinion but now I’m not sure, Elena, where you are in this. But let me ask the two of you, having heard what Gerhardt just said, I’m the patient sitting across from you and I’ve read all three trials, I’ve read the data, I’ve seen the approval and I’m fit, I could take any therapy I want, cost is not an issue. I want to be as aggressive as possible. So I guess where I’m going is this notion behind the patient-physician shared decision-making discussion. We do recognise it’s about patient preference, the informed patient. So what do you say to that patient that says, ‘I want my best shot, my first shot on goal. I really want to go for it.’ What’s your conversation, especially given what Gerhardt just said?
EC: In that situation, say, ‘Okay, it is approved but we don’t have evidence of benefit unless we test and we see that you have one of those alterations. But we know that the toxicity is this, that this treatment, the combination, will require you to come more often. We need to monitor your blood counts, you may need a blood transfusion at some point,’ and then it’s up to the patient, of course, to decide. But we need to explain them the balance and the potential benefit and the potential side effects. That’s how we need to make the decision.
NS: Yes, preference value, patient preference value, how risk-seeking/risk-intolerant. Eleni?
EE: Well, we go to the whiteboard, we have one in the room, and we go and we put all of that down. You know, I’m thinking while you’re talking and I’m like, ‘There is a probability that the patient will push.’ I will say, ‘Okay, let’s go. We’re not married to it.’ Not that if you’re married it doesn’t change. So we start and then based on monitoring, which has to be intense based on what you also discussed yesterday, we decide if we continue or not. It’s an option if we’ve got the approval.
NS: So, purposes of time, interesting, right? I like the notion of giving options to patients, I like the notion of starting even dose reductions in the therapies. It’s a good point because enzalutamide, that combination would be four pills daily in a fed/non-fed state, abiraterone non-fed state, the use of prednisone, some additional pill counts. These are all the things that I think are practical real-world issues, concerns. But let me get back to one other thing in the purposes of time. What about sequencing? So, Gerhardt, a patient comes in, and this is really apropos of the data that came through in MAGNITUDE regarding what is a reasonable amount of time to sequence versus to layer, we call it layering, versus concomitant use of both of the medications? What’s a reasonable strategy there? If you want to start with enzalutamide or abiraterone for is it a few weeks, a few months, versus all the trial data everyone pretty much started concomitantly.
GA: I think the first thing is these data are only relevant to NHA naïve patients because the whole hypothesis of blocking AR and PARP at the same time only counts for that population when we use NHA. Now MAGNITUDE, because there was a pre-screening period, the trial allowed patients to take four months, up to four months, abiraterone while their screening was being conducted. So I think about half of patients or 30%. There’s no subgroup analysis looking at the impact or the difference in those subgroups with prior abiraterone. I think we need to be careful about over-interpreting that because we may reach a conclusion that two months is fine but four months isn’t which doesn’t entirely make sense in many ways. So pragmatically if a patient has started abiraterone or enzalutamide and he’s found to be BRCA mutant and we want to combine with a PARP inhibitor we can add that in. I think that’s what the MAGNITUDE trial allowed and that seems like a pragmatic approach. The bottom line is test as early as possible so when the patient develops mCRPC you’re ready to make that treatment decision.
NS: I totally agree. With that let’s thank you very much, Gerhardt Attard, Eleni Efstathiou, Elena Castro. What a great pleasure to be here, a great meeting here at GU ASCO 2023. As always we really appreciate the folks at ecancer for allowing us to get together and have a really candid, full throated conversation with you. Thanks very much.