Welcome, Andy.
Andrew Armstrong: Oliver, it's a pleasure to be with you. Thanks for the invitation.
Oliver Sartor: A really important publication just came out, accompanied with a presentation here on Prostate Cancer Working Group 4. Now, you're going to give us, first of all, a little bit of the big picture. What is this prostate cancer working group? And why are we in version number four? Can't we just quit at number two or three? We keep going and going. So why are we here today?
Andrew Armstrong: Yeah, thanks for the big picture question. When I was in high school, the first PSA working group came out and prostate cancer is a unique disease. We've always used a biomarker to screen men, to track men, to detect minimal residual disease. Where did those metrics come from for defining PSA responses and progression? It came from that first working group because prostate cancer is that unique disease that has that biomarker. And we used to have radiography, bone scintigraphy, not a lot of cross-sectional imaging, not a lot of science behind any ctDNA or CTC assays. But having the PSA response really helped identify candidate agents to move forward, go no-go decisions in clinical trials and help move the field forward towards the development of androgen receptor pathway inhibitors.
So the first working group was really around the PSA as an endpoint. The second and third working groups were really around imaging as an endpoint, redefining disease states, identifying what we called the castration state where we learned that prostate cancer is not truly hormone refractory when a man's PSA is going up despite androgen deprivation therapy. The tumor rewires itself to make more androgen receptor to synthesize androgen and so that we have new agents that could extend the lives of patients that worked on the hormonal axis.
And so identifying what Mike has talked about today with the flare phenomenon to keep patients on drug while the bones are healing and not just get twitchy and remove patients unnecessarily because of transient changes. So I think ultimately the working groups established guidelines for both eligibility, endpoints for monitoring, for identifying patients most likely to benefit from our therapies and keeping patients on therapies that are benefiting them rather than getting rid of those therapies prematurely.
Oliver Sartor: So here we are with Working Group 4. And first of all, I'd like you to highlight, if you don't mind, some of the big changes that have occurred between three and four, and you were intimately involved with both three and four. And congratulations for your contributions, by the way.
Andrew Armstrong: Thank you. So in the past 10 years, as you know, because you've been part of these changes, there have been tremendous advances. So the AR pathway inhibitors have advanced from the late disease to now even the non-metastatic hormone sensitive setting. We're now using androgen receptor monotherapies in some patients. We have PSMA imaging without really metrics on how to define response or progression using that much more sensitive molecular imaging. We have genotyping of all patients now recommended by national guidelines for advanced prostate cancer, germline and somatic. And the integration of that into the use of PARP inhibitors into PD-1 inhibitors, the personalization of therapy and family planning, the domino effect of genetic risk counseling. The novel therapies that we're seeing emerge around AR pathways.
The need for new nomenclature, I think we've realized that there are patients that are progressing on these AR pathway inhibitors without castration. We know that a lot of patients don't like that word. We don't like that word when we're talking to our patients. And so there's the need for an evolution of thinking and language in the discourse of our writings of our protocols. Prostate Cancer Working Group 3 was one of the most cited papers in the whole field the past 10 years, because every protocol uses that as their template. And the FDA has used it for approvals, for radiographic progression-free survival, for metastasis-free survival, so that we can accelerate drug development for our patients.
So we've used this new term to focus on the new nomenclature of androgen pathway modulation. The androgen pathway modulator term is a broad term. It encompasses androgen deprivation therapy. It encompasses an AR pathway inhibitor, but also broadly other agents that intersect on that pathway, cofactor inhibitors, AR degraders, some emerging really cool new therapies. It reflects that some patients will progress with normal testosterone level, but still on one of these inhibitors. And so defining disease states in a more patient-centric language that reflects the actual therapies they're getting.
But layering on top of that, many more important phenotypes or characteristics of our patients, how they feel, what they're reporting to us, their genotype, their patterns of spread, their imaging that was used to define that pattern of spread. The patient reported outcomes in the end are really important part of Prostate Cancer Working Group 4. So a lot of new things in the past 10 years, we've also recognized there's frequently a disconnect between some of our biomarkers. For example, I told the story yesterday of a case of a patient of mine who had metastatic prostate cancer who responded beautifully with a PSA decline, metastatic high volume disease with triplet therapy, but a year later, despite a low PSA developed low PSA radiographic progression, that discordant progression is now being increasingly recognized as a common phenotype of lineage plasticity, de-differentiation, a poor prognosis. And illustrates the need to collect information, not just on PSA, but on imaging, on how patients feel, and distill that out into some evidence-based guidelines.
Oliver Sartor: Let's drill down a little bit in some of the neuro nomenclature, and I'd like to expand on that a little bit, because I think this is a difference that people may not be accustomed to. So help go from the older into the newer and a transition way.
Andrew Armstrong: Yes. So one of the motivations behind the change is to pick a term that doesn't presuppose the outcome. So we used to use terms like hormone sensitive, castrate sensitive before we would give the therapy. We're now employing the word naive for that state. So don't presuppose it, just call it naive to that treatment. And that would be applied to PSMA lutetium, a PARP inhibitor, docetaxel. So a naive state to that therapy. An androgen pathway modulator naive patient hasn't had ADT or an ARP, for example. A sensitive patient would be somebody who's responding to that therapy. A resistant patient is simply somebody who has some manifestation of their disease getting worse. Usually the PSA going up, but could be imaging, could be symptoms. So very simplistic, but those three disease states in terms of naive, sensitive, resistant, as applied to the APMs.
Oliver Sartor: And let's talk a little bit about biomarkers beyond PSA, and I wonder, and let's not the imaging per se. So let's talk about biomarkers beyond PSA, just for a point, because this is all part of our conversation today.
Andrew Armstrong: So we break down those biomarkers into credentialed response biomarkers and progression delay prevent biomarkers. So talking about response, these will be useful when you're screening a new therapy. And the more boxes you can check, I think the more likely you are to have success in later stages of development. Success with one single biomarker such as a PSA change may not be enough. And again, we want those successes to be durable. That means confirmation with subsequent values. A PSA decline of 30% or 50 or 90. We want to see that confirmed and the duration of that response is really critical. That's true of cell-free DNA. You want to see that normalization of plasma DNA content. Circulating tumor cells, same deal. You want to see normalization, alkaline phosphatase, LDH, some of the most important prognostic biomarkers.
Just like pain and symptoms, we want to see resolution to document complete remissions using any of these standards. This would include pathologic complete response as we move therapies into the neoadjuvant state. We'll have imaging definitions of that. Those are our most credentialed biomarkers. And we're going to integrate artificial intelligence. We're going to integrate more of a holistic look at each of these as we try to decide and decipher which of the therapies to move forward.
For delay, prevent, I think we use these differently. We now know that many patients with PSA progression using the older definitions will have radiographic progression without PSA progression, for example. When you're on an AR pathway inhibitor, APM, we often see the PSA as ultra suppressed and that when radiographic progression happens, the majority of those patients don't meet our old definitions of PSA progression. So we need to adjust for that. Monitoring any PSA confirmed rises above 0.2 could be important when the AR pathway is suppressed.
Other delay prevent endpoints are imaging based, and I think Mike covers that really well, the FDA has approved metastasis-free survival using conventional endpoints, for example. Will they approve based on just a new PSMA PET lesion? I think that is up to us to prove to the world that a new PET lesion is prognostically important, that it decreases survival when that occurs, and that intervening at that time may be helpful to that patient's life, and that new metastasis events on PET are associated with worse survival.
Oliver Sartor: And you have a feeling we could talk for a very long time. It's a beautiful manuscript, and thank you for your leadership...
Andrew Armstrong: Thanks for your contribution.
Oliver Sartor: ... to make it happen. And I'll simply say we're delighted to have you here on UroToday and I have a feeling this is a conversation that's going to extend over a very long time.
Andrew Armstrong: Thank you, Oliver.
Oliver Sartor: Thank you.
Andrew Armstrong: Look forward to that.