Read the full video transcript:

Alicia Morgans: Hi, and welcome to another podcast discussing recent trials and updates. My name is Alicia Morgans. I'm a medical oncologist from Northwestern University, and I have here with me today Dr. Charles Ryan, who is a professor of medicine and the Division Director of Oncology at the University of Minnesota. Thank you so much for being here with me today, Chuck.

Chuck Ryan: Hi, Alicia. Happy to be with you.

Alicia Morgans: Great. So, I think one of the most important things that has come out recently that I wanted to speak with you about is the IMAAGEN study, which is a really interesting study that you've been able to participate in, and I think really lead, and you had a recent publication on the study. I was hoping you could tell us a little bit about it.

Chuck Ryan: Sure, thank you. So yeah, IMAAGEN is interesting because while we just published it in The Journal of Urology very recently, last month in fact, it took a long time for this publication to come out because the patients were doing really well on this trial, and we were having relatively few events. And so, it took a while for the data to become mature enough where we felt comfortable publishing it. That's the sign of a successful clinical trial. The other interesting point about this trial is we started this I think around the same time, maybe even slightly before the apalutamide SPARTAN study and the enzalutamide PROSPER study were open. Those are of course phase three studies with the placebo arm.

With IMAAGEN, what we did is we took abiraterone. It's a phase two without a comparator arm. While we don't have that level one evidence, if you will, that the apalutamide study produced and the enzalutamide study produced, we have what I think are some fairly impressive data. One of the interesting aspects of apalutamide's development and the PROSPER study with enzalutamide is people were looking at these two drugs and saying, "This is really interesting data. It's important for us to be having this conversation about non-metastatic castrate-resistant prostate cancer."

But a number of people said, "What about abiraterone? When are we going to get data there?" These are that data and are probably going to be the only data that we will have where we look at the non-metastatic setting in terms of the treatment of abiraterone.

Alicia Morgans: Can you remind us who the patients are in this study? I know it's a non-metastatic or biochemical-recurrent, hormone-sensitive group, or it's the castration resistant group. Can you remind us who these patients are? This, again, is the non-metastatic CRPC patient population that we saw in SPARTAN and PROSPER, but how did you select these patients?

Chuck Ryan: Right. So this was before we had fully defined the non-metastatic CRPC group, I think, and we chose to look at two different patients. They had to be men with non-metastatic disease. But there were two camps. One was patients who had a PSA that was 10 nanograms per mls or greater, or those who had a PSA doubling time of 10 months or less. So you could get in with a high PSA that was rising slowly, or you could get in with a low PSA that was rising rapidly. One of the interesting things is the study was initially designed and completed the enrollment of 131 patients.

But we screened almost 300, and we had a very high screen failure rate, which no surprise was due to the fact that about a third of the patients who we screened just hadn't ... We thought they had non-metastatic disease, but they just hadn't had a scan in a while. When we went and we did scans, they actually had metastatic disease and thus were ineligible for the study. Those are the two patient populations who were enrolled. Now, I can get into the efficacy data now if you'd like. But we could talk a little bit more about those two groups of patients, and how they fit into the bigger picture of non-metastatic disease.

Alicia Morgans: Yeah, so just to follow up on your comment about screen failures and the way in clinical practice that we find ourselves sometimes watching these patients of biochemical recurrence, it can be challenging to think about when exactly we should be imaging these guys, especially with the slowly rising PSA over who knows how many years potentially. So, as you mentioned, it's not surprising that nearly a third of these patients ended up being identified as actually having metastatic disease. How do you think through that clinical problem? Before we move on to the rest of this study, how do you think through that imaging issue? And how does that fit into your practice in the context of this trial and study design?

Chuck Ryan: Sure. Well, there's been a lot of conversation about non-metastatic CRPC and how it's an iatrogenic disease. If we weren't starting hormone therapy on patients with very low burdens or no burden of disease, we wouldn't get patients with a CRPC non-metastatic phenotype. And so, that's actually an interesting conversation in and of itself. So a lot of these patients, I think most of them in fact were patients who had undergone radical prostatectomy or radiation therapy and had a rising PSA and then started their ADT in the context of a rising PSA.

And so, we think that that's probably a better prognosis than a person who for example had metastatic disease and then had a complete response and became radiographically and without evidence of disease during the course of ADT. So, right there, you tend to get patients who may have a slower rate of rise of their PSA and a better overall picture from their disease. But my personal practice is, first of all, that these are a minority of the patients we see. That's a pretty small minority. I try to do two things. One is that I don't scan on a regular basis, a regular time interval. I try to do risk-adapted evaluations and scan them when they have a rapidly rising PSA.

I'm concerned about them developing metastatic disease based on their PSA doubling time. In point of fact, in standard practice, I might be less inclined to do scans on a patient with a PSA of 11 that's rising very, very slowly than I might be with a patient who has a rapidly rising PSA that's 2.2. We still have a lot more to figure out on how to do all that and how to do it in an economically feasible way.

Alicia Morgans: Yeah, that makes sense. But it's those patients, the ones who have a PSA of 12, but it was 11 a year ago that we don't end up scanning necessarily. When that trigger point should be is something that we're still thinking about. Luckily-

Chuck Ryan: The other piece I was going to add is ... excuse me ... is that with novel imaging, which these patients did not have, it's possible that a greater number of them actually had metastatic disease if we were to integrate PSMA, PET scanning, and those types of things.

Alicia Morgans: Absolutely. But the good thing I think too is that a lot of these patients generally are fairly asymptomatic, so even with metastatic disease, because if they weren't, they would be getting scans. Back to your trial, the treatment was to give them abiraterone, and you gave them prednisone, only five milligrams a day. Is that based on the LATITUDE-STAMPEDE approach? Or can you talk a little bit about that?

Chuck Ryan: Yeah, well this was pre-LATITUDE. I mean, we didn't have the results from the LATITUDE when we designed this study. One of the major goals of doing this study was to look at the long-term safety of abiraterone with five milligrams of prednisone, and so, that was why that was done.

Alicia Morgans: Okay, great. So get into the efficacy data.

Chuck Ryan: Yeah, so we've found a very high rate of response. In fact, 87% of patients had a 50% decline in PSA or higher. So it's an 87% response rate. We only had a few, a handful who had no PSA decline, which is actually sort of interesting when you think about it. Only about 24% of the patients had confirmed radiographic evidence of disease progression while on the study, and we at the time of doing the analysis, still 76% were censored since they had discontinued treatment prior to disease progression or were continuing on the study. After four years of therapy, 62% of the patients were progression-free.

When you compare that to the median PFS in the metastatic setting, which was about 17 months, it suggests a very favorable outcome for this group of patients. Again, is that because of burden of disease? Is it because of the more inherently benign biology that drives non-metastatic disease? We don't quite know. But high rate of response, long duration of response. And actually, we still have some ongoing analyses to look at — what types of progression patients experienced and what that has in terms of clinical significance.

Alicia Morgans: Great. One of the things that I think about when I'm thinking about the efficacy data of course is balancing it with the side effects data. It was interesting to me that a significant number of these patients seem to have a grade three or higher AE. I think it was just over 60%, which is actually surprising to me in the mCRPC patients that I end up treating with abiraterone. It seems generally so well tolerated, and of course, I'm not doing a CTCAE assessment on them at every clinic visit. But I just wanted to know what your thoughts were on that and the side effects. Was it generally well tolerated? What did you see in the study?

Chuck Ryan: With regards to the safety, we've actually reported this at AUA and other meetings in abstract form. It is interesting that you point this out. The duration of exposure to abiraterone and five milligrams of prednisone is very long. Like we said, median time, you know, four years or so of patients on the drug. And so, we did have a high prevalence of AEs over time. Hypertension was present in 42% of the patients. Fatigue in about 40%. Actually, hypokalemia occurred in 44 patients, which is about 34% of the total. I think that's a function of two things. One is prolonged exposure, and two is lower prednisone dose.

I think one of my messages to clinicians around this issue of prednisone is that it's really kind of an individual thing, and we need to monitor patients and maybe titrate their prednisone. I've done this several times in the clinic. I've had patients who have had a lot of these mineralocorticoid side effects on five, and I put them on 7.5, and they go away, or they go on 10 milligrams and they go away. And so, this does call for a continual assessment of adverse events. Along that line, I would like to point out that we enrolled about 14% of the patients on this trial were African American.

This group of patients had a slightly higher rate of grade three hypertension. It was 47% versus 20%. This is an interesting phenomenon, and we've seen this at ASCO that people have reported differential outcomes of African American patients on abiraterone, and it seems to suggest that they may be slightly more amenable to the mineralocorticoid side effects. On that note regarding efficacy in the African American patients, we actually saw that they had a very similar time to PSA progression on this study, which is different from the analyses that were presented at ASCO. But anyway, an interesting sub-observation in this trial.

Alicia Morgans: Absolutely. I think there's definitely more work to be done in that area because although there were ... There was a reasonable percentage of patients in this study. The numbers overall were still pretty small. But definitely, I look forward to seeing where that information goes and to other prospective studies, really looking more deeply at differential response in terms of efficacy and side effect profiles in different patients. But I have a question for you really about the application of this study, because it was really interesting to me that the AUA guidelines that came out very recently, for CRPC actually, include abiraterone as a possible option.

I spoke to Mike Cookson about this. It's a possible option for the non-metastatic CRPC patient population. I talked to him about it because there isn't phase three data, and like you said, this really doesn't have a comparator arm. But I think his thought was that abiraterone at some point will be going off patent and will be potentially more accessible to patients. The committee thought that based on things like the IMAAGEN study data as well as prior use by the urology community of ketoconazole that this is a reasonable option, potentially, for patients who did not have the ability to get other AR directed therapies.

I was just curious as to your thoughts on that because it was a little bit surprising to me, though of course I understand the reasoning behind it after talking to Mike.

Chuck Ryan: Yeah, I agree with that assessment. I think that we have a lot of questions to answer about non-metastatic CRPC whether we should treat patients at all, when we should treat them, and with what agent we should treat them. I think that one of the reasons I'm happy to see this trial come to fruition and mature and be published is to address that exact point that you raised, Alicia, which is it looks very similar to enzalutamide and apalutamide. And so, maybe it should be considered within that group of patients. We don't need to have randomized phase three placebo-controlled trials for every possible clinical decision that we're going to make.

This is one where we have three very similar agents, two of which have had placebo-controlled, randomized phase three trials, one of which has not in this setting. And yet, if you were to line up the control arms from the SPARTAN study and the PROSPER study and the treatment arms of apalutamide, enzalutamide, and abiraterone in this study, I think you'd see very similar-looking curves. Then the other thing about abiraterone is ... You know, I continue to think that it's a fascinating agent because of the hormonal issues associated with it.

But also, it's got the most, I think, pharmacologic flexibility because we could continue to develop abiraterone at lower doses perhaps, maybe with the food effect. We can continue to work on titrating the prednisone dose. We see that here we have a therapeutic situation where the median time to the PSA progression was 29 months, whereas in the 302 setting, with metastatic disease, the median time to the PSA progression was about 11 months. So we're looking at almost an additional year and a half of therapeutic exposure.

There's a lot of room given that to think about how we could approach this from a pharmacoeconomic perspective, a pharmacokinetic perspective given all that we're finding out about abiraterone and its flexibility with respect to food and other things. Now, I don't know if any of those studies are underway. I'm not involved in any that are doing that at this point in time. But it's a big world, and there are a lot of patients out there who are getting therapy in this setting, and to have a third potential option I think is a good thing.

Alicia Morgans: You raise a really good point about pharmacoeconomics. Whether there are any groups interested in assessing this, I think ... Hopefully, there are because it may be that drugs that can act not necessarily an exact same mechanism of course, but both acting on the ... or all acting on the androgen receptor pathway. It may be that we have a lever here that can be used by multiple drugs and potentially really maximized. This, I think as you mentioned, would be more economically viable, particularly if we're able to give fewer pills with the food effect.

I look forward to future conversations with you about that and certainly talk with our colleagues who study that a little more intensely. So, what are your closing thoughts? How are you using the results of this study in your day-to-day practice? And where does this fit in for you?

Chuck Ryan: I think the main take-home lesson that I take from this trial is a couple things. One is that we now know, and we know this from this setting, and we also know from LATITUDE that starting on abiraterone and five milligrams of prednisone is a perfectly acceptable thing to do. However, you need to monitor your patients. Some people need more than five milligrams. So I think that that's one of my main messages to the clinician. The other thing is that patients with non-metastatic CRPC generally do very, very well whatever therapy they're getting.

This study does not answer the question of when and if we should treat patients with non-metastatic disease. It simply adds to the growing body of literature that says, "When you do, the outcomes are going to be fairly useful." Then I think the next phase is to know what we've done to the natural history of the disease by treating it this early. What type of progression do these patients have? What ultimately happens with them subsequently? The other thing is ... And I find this fascinating. We really haven't delve into this is, what are the features that are associated with these extremely good responses that we see?

We have patients on the study who had essentially a complete response and persisted on it for years. I've seen patients like this is my practice, and what we should really try to be pulling out are some of that extreme responder phenotype in the setting of androgen deprivation therapy, with abiraterone and ADT. We did a little bit of this. In fact, as you know, Alicia, a long-standing theme of mine is the study of the relationship of baseline testosterone levels and outcome, and what we found was that patients who had above-median testosterone levels actually had a much more significantly longer time to PSA progression.

So that's another take-home message, which is that we need to continue to monitor testosterone levels, and that is prognostic and may in fact color other factors related to the choice of this treatment.

Alicia Morgans: Absolutely. So this study, we're not necessarily looking for survival data in this study. But certainly for patients treated on the PROSPER and SPARTAN studies, we're looking for survival data to really understand what happens with this earlier integration of intensive androgen deprivation therapy. I also really want to just follow up on the thread of thought that you had regarding the differential responses and then presumably differential mechanisms of resistance that evolved, because it may be particularly as we're integrating these therapies earlier and earlier that we need to start thinking clearly and designing trials regarding sequencing, which we've already tried to do.

But what is the best drug after an intensive androgen deprivation in the M0CRPC setting? And are there potentially different, evolving pathways where we can pick out using molecular approaches, "Which patients should get this drug versus that drug"? I think we continue to strive in that direction. It's certainly not happening tomorrow. But I do look forward to a day when we can say for you, based on your profile, and your pathways of resistance, and maybe even the side effects that you're experiencing with these particular therapies, "The next treatment that you need down the line is this and then that and then this."

So, kudos on the excellent work. Thank you so much for taking the time to speak with me today.

Chuck Ryan: Always a pleasure. Thank you.