Christopher Sweeney: Thank you.
Matthew Cooperberg: Really, a pleasure to have you here. Tell us a little bit about where we've been and where we're going in this space, in this complex space of advanced disease.
Christopher Sweeney: For sure. I think the first thing is to address the first thing you said, "Where have we been?"
Matthew Cooperberg: Yeah.
Christopher Sweeney: And so in the 1990s, we basically had testosterone suppression, simple ADT, we call it, with or without a weak antiandrogen such as bicalutamide, and maybe mitoxantrone for CRPC, and then later docetaxel. In that paradigm with access to those agents, patients were living a shorter time. We talked about a median survival of about 18, 24 months in every paper we ever wrote. There was a tail on that, about 25% of patients alive up to 8 years. And this was great work out of the SWOG portfolio of clinical trials. And so there is a tail, it was documented, 25% of patients live at 8 years with just the weaker therapies than what we have now. What we do know, fast-forward, when we added docetaxel to those who needed it, synchronous high-volume, for example, this overall survival of the general population, metachronous to synchronous to high-volume to low-volume, we now get up to about a 35% 8-year survival. Now with the emergence of the ARPIs, where we've sequentially gone from noting drugs that are active in CRPC into the hormone-sensitive, we saw the ARPIs were active and then we brought those back into hormone-sensitive. And with some patients who seem to need it, possibly docetaxel added to the ARPI or ADT ARPI alone, we now have a 50% 8-year survival. And recognizing this is a population of men who are on the studies who are between 65 and 70, going all the way up to 80 sometimes on the accrual, and some of these men are diagnosed in their 90s even. But anyway, in the trials, 15% of these patients are now dying of something else other than prostate cancer. So many patients, this is becoming a chronic disease with long-term androgen suppression and ARPIs, and so we clearly have our backbone being, without any equivocation, ADT and an ARPI. Which ARPI?
Number one issue is which one can you get? What we've also learned is that the ARPI probably not most suitable for men over the age of 75 is abiraterone. And there's a lot of reasons why that would be the case, such as the chronic adrenal suppression, the age-related frailty, and needing adrenal glands, presumably, the impacts on the mineralocorticoids, the blood pressure, the blood sugars, the liver, all these host of collateral effects from abiraterone. So best thing is to not treat frail men over the age of 75 with abiraterone. But otherwise, carte blanche, whichever one you can get, get them on ADT and ARPI. If they're not a candidate for an ARPI for whatever reason like short life expectancy, that's a rare patient population and you can think about that, but your default is ADT plus ARPI. And then the next thing is, what do you add to that? So work is ongoing to try and work out which patients benefit from adding docetaxel to an ADT and an ARPI. We know that adding docetaxel has the most benefit when added to the testosterone ADT alone in men who have poorer prognostic disease. And there's a gradient of effect from synchronous high-volume, most effect, metachronous high-volume, similar effect, but smaller patient population, synchronous low-volume, a smaller benefit, but we know those patients also benefit from the less-toxic radiation to the prostate, and then no benefit and metachronous low-volume. We showed that in a collaborative effort by analyzing and reanalyzing the cooperative group data CHAARTED, and then as a global collaboration, we got together and confirmed that by looking at all the cooperative group studies with adding docetaxel to ADT, GETUG 15 from France, STAMPEDE led by the UK, but European accrual as well, and CHAARTED.
And we showed, in a global collaboration, sharing individual-patient data, we could get this refined information of who benefited. We now need to do that for docetaxel added to an ADT and ARPI. And in that like-minded collaborative approach, we've been exploring different gene expression profiles, PAM50, Decipher Gene Classifier. We saw some signals there. The one that really got supported with the Decipher Gene Classifier in the STAMPEDE study was adding docetaxel to ADT in men with Decipher high, cut 0.8, 0.85. And then what we've done most recently is written a pre-specified SAP, statistical analysis plan, and analyzed it in the ADT enzalutamide plus or minus docetaxel, not randomized, but patient selection for it. And we've got that data under review to try and get a biomarker selection as to who benefits from adding to a baseline of ADT/ARPI. So baseline right now, consensus by the APCCC, is high-volume ADT/ARPI plus docetaxel if you think they can benefit and tolerate it. For those who are synchronous low-volume, radiation to the prostate, not, probably, need the docetaxel, but work to be done, and there's some data supporting that from the PEACE-1 study. And even in patients with metachronous low-volume, the best prognosis, there is still a clear overall survival benefit, very similar to the treatment effect of adding enzalutamide to ADT in the biochemical relapse EMBARK, but you go one step back to metachronous low-volume, a few spots on a conventional scan, there's still a major treatment effect with a hazard ratio for overall survival in the 0.5 range, increasing the 5-year survival from about 65% to 80%. It's a big number.
Matthew Cooperberg: Yeah. Yeah.
Christopher Sweeney: ADT plus ARPI across the whole spectrum, docetaxel, probably, for the higher-risk, poorer prognostic group, and radiation for the less-poor prognostic group. That's where we're at, but we need to refine that with the biomarkers. Now, let's also realize that we need to think about a number of other things, bone health, cardiovascular management, counseling about alcohol and smoking, which increase blood pressure and cardiovascular disease, and we worry about the ADT. We need to think about that. And did I mention that 50% of these men are now living 8 years? So these issues really need to be thought about. And then the notion of psychosocial health and long-term management. And so 8 years of testosterone suppression, that's a problem. So what do we do next about that? And so there are a number of studies trying to answer that question of a treatment break. So a treatment break, ADT plus an ARPI after 6 months, the EORTC is leading a study called DE-ESCALATE, which is a great study and a cooperative group study. The Alliance is leading a phase II study, with the PI being Atish Choudhury, looking at ADT/ARPI and stopping after 18 to 24 months to see if you can have prolonged suppression and then maybe have patients off therapy for a long time. And we've all got some patients plus or minus SBRT for the metastasis who are 5, 8, 10 years testosterone recovery and that cancer never came back. So there's people trying to work out a treatment break based on our minimal residual disease, not a ctDNA, but just a PSA nadir. The other studies that are ongoing are some really important studies led by the CCTG with SWOG, TRIPLE-SWITCH, ADT/ARPI at 6 months, PSA less than 0.2, yes or no. If no, then you get randomized to docetaxel versus continue ADT/ARPI to see if you can select out the docetaxel in the patients who benefit from docetaxel just by that simple blood test. And then the Alliance is looking at it upfront, ADT/ARPI plus or minus docetaxel based on the presence of a tumor suppressor gene mutation, be it p53, RB1, PTEN.
So I'm emphasizing a lot of the cooperative group studies because they give us the basic information we need to know, long, long-term survival, and doing some treatment optimization, whereas we've been doing a lot of maximum dose for a maximum amount of time in the broadest patient population. But going forward from 2026, we need to be thinking more about treatment optimization and pulling back on therapy for some and escalating in others. Biomarker-directed. So we can use PSA, we can use some existing exome data that we know, gene expression profiling that we talked about for the Decipher assay, but it is a collaborative global effort to get to these biomarkers. Now, where do we go from here in 2026, Matt? So we now have three studies of biologically-targeted therapy. Lutetium PSMA is a radioligand going to patients with PSMA expression, PARP inhibition with patients who have sensitivity to PARP inhibition, and there's the whole panel of the DDR genes. I'm most confident in the data that I've seen from AMPLITUDE in the BRCA-mutated setting, patients with the mutation. I am not convinced about the non-BRCA setting. And with the AMPLITUDE, we see a clear rPFS benefit supported by an interim OS hazard ratio less than 1.0, fairly narrow confidence intervals up to less than 1.2. So it's trending in the right direction. But pause, what's the treatment burden? Significant amount of anemia and red cell transfusions, and there can be some life-threatening aspects with that from cardiac compromise, so we need to think about that. So modest treatment benefit, some treatment burden. We also now have lutetium PSMA in the PSMAddition study. Again, modest treatment benefit in terms of an rPFS, met its primary endpoint.
The impact on the clinical benefit, I would say, is modest, and there is a treatment burden as well. So the quality of life data is being released, and we just saw yesterday with Mike Morris reporting a decline in the quality of life on the lutetium PSMA, and then it appears to come back off of that. And then there seem to be some splitting of the quality of life late, which may speak to the issues of some long-term toxicities from radiation that we need to parse out. Some patients benefit, some patients have some futile therapy, so we need to optimize who benefits from that, and we'll come to that in a second. And then we also have AKT inhibitor by PTEN loss by immunohistochemistry. Again, modest rPFS benefit, supported by an OS benefit. PTEN probably isn't the best selection marker, at 90% loss. There are other studies like IPATential, which also suggested PTEN can identify patients who have an rPFS benefit, but not an OS benefit. Similar seen with capivasertib in CAPItello. So all consistent findings. And again, a treatment burden with rash, diarrhea on therapy, and then some life-threatening, at times, hyperglycemia. So we see something. To me, I see there's a treatment benefit for some, not for all. And I feel like we really need to optimize the biomarkers to work out who really benefits from this. So I think we've got it for BRCA, for the PARP inhibitor. Lutetium PSMA, can we start to work out how much lutetium PSMA is associated with benefit from lutetium PSMA? So we know SUV mean greater than 10 seems to identify men in CRPC post-ARPI, post-docetaxel. What's that number in combination with an ARPI? Should they be getting the lutetium PSMA when the PSMA is gone? So should we be doing dose adaptation?
As you know, in Australia, we're a big fan of that idea. But also in the AKT inhibitor, is the NGS assay possibly more reliable about the AKT activation pathway, as we saw in IPATential, as was seen with breast cancer with fulvestrant and capivasertib. So there's a lot of work to optimize our current biomarkers to optimize therapy for patients with metastatic hormone-sensitive prostate cancer. Because again, remember, these men with our backbone, our standard therapy alone, 50% are alive at 8 years, 15% of them are dying of something other than prostate cancer. In the real world, toxicity goes up with drugs with somewhat of a smaller therapeutic window, and efficacy goes down outside of clinical trial eligibility and constraints and guidance about toxicity management. So I'm not convinced for routine use of these yet. Others may feel the data's there, but I think one of the aphorisms I've started to adopt is, "The Wright brothers didn't get it right the first time. In Kitty Hawk, they've got a couple of meters, but I just flew in on a 15-hour flight from Adelaide to San Francisco with a bunch of other people." So it takes time to get where we want to be and I think we've got to do some further work to optimize those biomarkers. And I think that's where the cooperative groups step in again and optimize all the therapies we have with companion diagnostics to work out the maximum benefit. And I think we can do that with a debulking strategy and then partition patients out. Just like you've just discussed with Martin Gleave about the GUNS study. We can do this when we put our minds to it with the right companion diagnostics, right drug for the right patient at the right time and the right amount of drug, and hopefully optimize therapy.
Matthew Cooperberg: Let's talk about the diagnostics a little bit more. You talked about Decipher and you presented yesterday looking at Decipher as a prognostic, not too dissimilar from disease burden. And Decipher has been used in a number of contexts to intensify or de-intensify based on the score.
Christopher Sweeney: Yes. In the localized setting.
Matthew Cooperberg: Yes. Localized and the intensification with radiation with apalutamide.
Christopher Sweeney: Yes.
Matthew Cooperberg: There's a few contexts where it appears to work, but that's really the Decipher score as a prognostic. But there's all these other things that come off that platform or it can be read from sequencing data, as Marty's doing with GUNS. You talked a little bit about PAM50. It seems that there are clearly patients who, at baseline, do not have particularly AR-driven tumors and are really ramped up in cell cycle and other predictors of response to chemotherapy rather than ADT.
Christopher Sweeney: Yes.
Matthew Cooperberg: But it doesn't seem like there's ever a conversation about, "Well, let's start with ADT plus docetaxel," CHAARTED style, "without an ARPI early." When you talk about minimizing toxicity and minimizing cost as well, it has struck me that we should be making even earlier decisions based on not just biomarkers as an indication of risk and disease burden, but as an indicator of biology and choosing therapies early that are more biologically-driven. Do you think we're heading in that direction or is this backbone kind of inviolable of starting with the AR pathway no matter what?
Christopher Sweeney: Provocative question. In 2026, I do not have a reliable biomarker that I can say to myself or patients or anyone, hold the ARPI. Is there a patient population who would benefit from just, potentially, ARPI alone and not have to go through castration? Possibly.
Matthew Cooperberg: EMBARK style.
Christopher Sweeney: Exactly. Or ADT alone, but unselected? We don't know who they are. Matt, let me tell you about MSTIR. Movember SAiGENCI Transformational Integrated Repository, which is a database that I'm developing with support of Movember and Prostate Cancer UK and STAMPEDE and MRC and the Spanish group who's doing the CAPTURE and the PROCURE with Elena Castro and David Olmos, University of Tasmania, low-risk germline DNA, and ANZUP. So the goal is to get all the clinical genomic data from the ANZUP studies, the MRC STAMPEDE studies, the Spanish studies from at least germline DNA for exome and SNP arrays for patients with low-risk disease, all the way to synchronous high-volume with tumor gene expression, tumor exome, germline exome, and SNP, to actually have probably about 6,000 patients worth of data with clinical genomic RNA and exome profiling to actually ask those questions with the whole spectrum of disease treated with testosterone suppression alone ADT, to ADT/ARPI, to ADT docetaxel.
So we have the data in this repository to be able to answer those questions. So it's a future state, but a very important future state that we could start asking. And this is where machine learning can go in and find those cell signatures that you're talking about. We're starting to do some of that work already with the ENZAMET RNA expression profile data to answer those exact questions. So we have the technology, we have the assay data now, and we will have the repository to ask those questions. And hopefully we'll be rolling out data over the next 6, 12, 18, 24 months to answer those questions.
Matthew Cooperberg: See you back at GU '27 with that update, hopefully.
Christopher Sweeney: Yes. And should just hark shout out to Movember who supported it as part of the MSTIR, Movember, and SAiGENCI is the Adelaide University, my team, Cancer Institute where we've got the mathematicians who've built this so that this will be an accessible database for academics and commercial entities to be able to make the data, democratize it, if you would, so that a lot of people can go hog wild and come up with the best science.
Matthew Cooperberg: Very, very exciting. Last question, something you did not touch on yesterday, but I'm going to ask the question because you spent a lot of time in this country before returning to a country with a less insane healthcare system. How do we afford all this? As the armamentarium expands, and this is happening in bladder cancer concurrently as well, we don't talk much about cost, but the burden to the patients is quite substantial in some cases. And the burden to the system is incredibly substantial. So we talk about additive toxicity, but there is the additive financial toxicity. How are these conversations happening as the treatment paradigms evolve? Who's leading the conversations? Is anybody leading the conversations?
Christopher Sweeney: So the market forces have failed with all the drugs when they are on patent and there's exclusivity. Darolutamide, apalutamide, enzalutamide, they've all had the same pricing, the same copays. Interestingly enough, Matt, just looking at it, abiraterone is still widely used in the United States because the generics and the lack of a copay concern. Abiraterone, in areas where the drugs are paid for by the government and there's not much of an out-of-pocket expense, it's covered. And so abiraterone, with all its issues with age over 75 and et cetera. It's actually rarely used outside, in Australia, for example, because of the comorbidities and the comedications. So I think the market forces will drive it down. So enzalutamide, when it comes off patent, and it is in some regions already, it'll drive that market force down. The other way to drive the pricing down is actually defining the treatment breaks. And so I've failed to mention the LIBERTAS study where they're stopping the ADT and continuing the ARPI in a treatment optimization, but that's stopping the cheaper option, but that's the way the market forces are. But it's a scientific question. But these drugs will come off patent and they will. So it's the loss of exclusivity that'll drive it down, I think, in the next couple of years, but also the results of these cooperative group treatment optimization studies.
We've got a health economic analysis of ENZAMET using the Australian MBS and PBS data, and the papers are in circulation between journalists trying to get it reviewed and published, but it's basically defining the cost of the drug that actually is associated with cost-effectiveness in various mannerisms. Interestingly enough, we're able to see the cost of the falls and fractures from ADT and ADT-Enza, and that's an expensive line item. So some of the cost, actually, may be better supportive care to prevent the complications. So I think the market will be driven by the generics and bring cost down, and I think more rational, optimized treatment breaks, if we can show that, it's beneficial.
Matthew Cooperberg: Any last thoughts or thought for the future?
Christopher Sweeney: I am optimistic that we can optimize therapy with optimized biomarkers and patient selection. So I may be a little bit less enthusiastic about the data than some other people, but it's not saying that I'm not enthusiastic about it. I just think we need to take a pause, build the better plane, and get it right in the long run.
Matthew Cooperberg: Well, thanks so much for your time. Thanks for all you've done for the patients and for joining us at ASCO GU.
Christopher Sweeney: Thank you for the great opportunity to share some thoughts, and it's a team sport.