Daniel George: Thanks, Neeraj. Happy to be here.
Neeraj Agarwal: So let me start with Dan and Neal first. Could you please tell us about PTEN biology? You are the ones who did the trial. And what is the clinical meaning? Even though I briefly elaborated on that, but please give us your feel as the investigator in the trial.
Daniel George: Yeah. Maybe I can start, Neal, and I'll let you follow up. The truth is, Neeraj, that we've known about PTEN loss and prostate cancer for over 30 years now. It's one of the real common tumor suppressor genes that are lost. And one of the more common mutations that happen in prostate cancer, probably 25% of prostate cancers that are aggressive and all the way up to probably 40 or 50% when you get to late stage castration-resistant disease. So it's a mutation that tends to increase in its prevalence over time. And it plays a key role in growth factor signal transduction. It regulates this protein AKT, which is a driver for a lot of growth and survival signals in cancers broadly, but especially in prostate cancer. And it's kind of a parallel pathway, if you will, to the androgen/androgen receptor pathway that we all know so well.
And what's important about that is as we put more pressure on androgen deprivation and androgen receptor signaling, the biology of PTEN loss, if it's present, becomes more important. And that active unregulated AKT can be a driver of disease resistance. And we've never really studied that prospectively. It's been largely retrospective analysis. And one of the great things about the CAPITELLO study is it's really the first metastatic hormone-sensitive prostate cancer setting to focus on this PTEN loss population and how they do. But we know throughout the natural history of disease that loss of PTEN is associated with a worse outcome from localized disease all the way through to castration resistance.
Neeraj Agarwal: So Neal, what is your take on the real-world testing? Are we testing for PTEN loss?
Neal Shore: So we're not, and I really appreciate the way Dan framed it. I've been testing regularly for close to eight years, but I recognize I'm a bit of an outlier. I've been getting NGS testing on all my metastatic patients. Of course, I get all my metastatic sensitive and certainly resistant. I get germline testing. And I get germline testing on anybody who's got localized disease, especially if it's high risk, and if they're even not high risk, if they have a significant family history. At least in the US, germline testing is pretty inexpensive. And definitely for patients who have metastatic disease, I've been routinely doing it. And what I've seen to Dan George's point is that many of these patients in the resistant population upwards of 40% of PTEN loss, 25% in the sensitive biology, PTEN loss.
And what I've kind of empirically observed, to Dan's point, and the data bears it out, when you look at the control arm in CAPItello, and we'll talk about that, you mentioned that in your intro, is this is a really bad prognostic tumor suppressor gene. So I really enjoy the fact that CAPItello was designed to exploit specifically patients not only with NGS-PTEN loss, but specific immunohistochemistry protein loss. We'll talk a little bit more about that. I had the privilege of doing a resistant trial with Johan DeBono, and Chris Sweeney led it, and it was an all comer study, but we saw clearly a remarkable signal in the patients who had PTEN loss. And so I'm really happy that the folks at AstraZeneca decided to do this study, which we all participated in. And I think the notion of testing, as you say, so important. If you're not testing, you're just not going to know. And then learning the MOA, the PI3K/AKT pathway and the synergy potentially with an AR blocker, AR pathway inhibitors is what we want to leverage.
Neeraj Agarwal: Absolutely. David, I'll come to you now about your practice and then the practice you have observed in your setting. How is testing done and what is the rationale for combining this drug known as Capivasertib, the AKT inhibitor with abiraterone?
David Morris: So I think Neal's comment on synergy between AR axis and AKT inhibition is an important one from urology standpoint to think of the synergy of testing because we've been getting the message slowly out on BRCA testing, HRR testing, and really just starting to scrape the surface of even getting the testing done appropriately for those disease states, which have been around and with approvals for almost five years now. So PTEN, another reason to be doing testing upfront. And honestly, the more indications we have, we're hoping that prostate cancer becomes like lung cancer where we have very clear indications depending on which pathways are mutated, and we know we need testing upfront to help make a decision. Having just one actionable mutation, really a category up until now has limited some of the uptake, I think.
But I think NGS testing is probably the most common across urology groups in the US, and that's typically finding a commercial partner to send out for NGS testing, looking at the DNA, the RNA level. IHC looking for protein loss is a little more within the world of pathology. And so if you have pathology in house, it's certainly possible to do IHC testing in house. And I think many of those commercial partners are starting to recognize that they need to offer IHC testing in parallel to their NGS testing. And so my hope is that those commercial partners will pick up the slack and allow us to do it all one shop for not just HRR, but PTEN and also by IHC.
Neal Shore: If I could just piggyback for a second, Neeraj, on David's really important comment. So a lot of our academic colleagues, they do a lot of their NGS in house. I don't know if Duke is doing it, they may be. But when you're in the community, it's all done commercially. And I'm agnostic to really all of the really wonderful companies that are out there. But a few years ago, I just started telling them that, "I insist that you do IHC PTEN testing. It's really easy to do. It's very inexpensive. And also if you have in house pathology, you can do it in house. Your own in house pathologist can do this. This is not arduous, especially you have to have a certain amount of scale."
And I think it piggybacks on the importance of doing HER2 testing or the tumor agnostic indication of TDXD, different molecule, but getting also commercially known as HER2. But the thing is, there's really two reasons now, PTEN and HER2 to be doing IHC testing for prostate and bladder cancer patients. Here we're talking about the CAPItello-281 trial that's important to focus on, but it's for our colleagues to say, "Oh, it's not necessarily just a one-off."
Neeraj Agarwal: I think it's a matter of getting drugs in the clinic and people start testing. It is already being done in the clinics of our colleagues. In lung cancer patients, for example, breast cancer, we have seen these testings being a part of routine workup for a really long time. So I think it's a matter of time. And I'm so glad you mentioned IHC testing being not too cumbersome, in fact, very easy to do, which is great to know. Coming back to David before we go to Dan. Dan, so what is the rationale for combining this AKT inhibition with AR inhibition?
David Morris: I think Dan touched on it earlier that the AR pathway and that AKT pathway are run in parallel as different signaling pathways that can drive cellular development. And so PTEN loss actually leads to this AKT signaling the mTOR pathway for activation. And so if you're suppressing the AR axis, you're basically shunting patients to that pathway for their cellular turnover, and the cancer basically grows by that activation pathway. And so there's some synergy, basically, if someone's got PTEN loss to basically try to inhibit both. If you're just doing doublet therapy and inhibiting the AR axis, the cells are going to escape that suppression by having the AKT pathway for escape to grow. So that's really the reason for targeting these two together. It's a combination therapy that you need to actually basically attack both signaling pathways at once. So that's my scientific urology level understanding of why we're working on both pathways at once.
Neeraj Agarwal: That's fantastic summary, I must tell you that. Dan, being one of the chairs of the trial, could you tell us more about the design of the trial eligibility criteria before we ask you to talk to us about the results of the trial?
Daniel George: Yeah, absolutely, Neeraj. This was an ambitious undertaking. As I said, loss of PTEN, although it's one of the more common mutations, alterations that we find in prostate cancer, it's still a minority. So we knew we're going to have to screen a lot of patients to find this subset of patients that have this. And then it was really determined to focus in the metastatic hormone sensitive setting. As Neal mentioned, there was a prior study, another AKT inhibitor that had been studied in the castrate-resistant setting where we know there's more prevalence of this loss of PTEN, but where there are other genetic alterations that are acquired as well in that context. And so while the drug worked there, in particularly it worked, as Neal said, in the patients that had loss of PTEN, for the overall population, it didn't change the progression-free survival.
For the subpopulation of patients that had loss of PTEN, particularly loss of PTEN that was over 50% with the greatest benefits in patients with loss of PTEN in over 90% of the cells. So that was the population that the sponsor AstraZeneca focused on, and that we as a steering committee really identified as the highest yield the patients with loss of PTEN, and that was at least 90%. And that dwindled our numbers down to just 25% of screened patients that met that criteria. So this was going to be a big study. It was a randomized study, and at the time, a lot of studies had been, again, standard ADT as a control arm, all of our big registrational studies for AR inhibitors.
But this study was being compared to abiraterone and ADT, a standard of care at the time for metastatic hormone-sensitive prostate cancer. And to your point, this was a one-to-one randomization of patients getting abiraterone prednisone with or without either placebo or Capivasertib. And the medicines dosed 400 milligrams twice a day, four days on, three days off. And we'll talk about that a little later and why that built-in break is important in terms of patients being able to tolerate this long-term.
Neeraj Agarwal: That's great. So again, please, for our audience, what were the primary endpoints?
Daniel George: Yeah, so this study, a large study with primary endpoint was radiographic progression-free survival, but overall survival is a key secondary endpoint. They do a hierarchical methodology to the secondary endpoint. So overall survival has to be met to then look at a number of subsequent ones such as the time to subsequent therapy or pain progression or castrate resistance or what have you. So a number of secondary endpoints, but radiographic progression-free survival, the key primary endpoint.
Neeraj Agarwal: And really a large study of more than 1,000 patients.
Daniel George: It is a large study. And I think recognizing to get a thousand patients, they had to screen over 6,000 patients. So a Herculean effort. And to add to the challenge, Neeraj, this was launched in July of 2020, which is just a few months into the COVID pandemic. So you can imagine taking on this endeavor worldwide global study, thousands and thousands of patients screen, most of them not eligible. For the investigators to really keep the faith to answer this question is remarkable and I think a testament to our field and the need to really be able to develop these precision strategies in this early frontline setting.
Neeraj Agarwal: Just for our audience, I'd like to just show the design here of the trial schema. And we can see that, as Dan, you said, 6,000 patients were screened for 1,000 patients to be eligible, and they were randomized to these two arms with the progression-free survival as the primary endpoint. Before we get to the results, could you tell us about baseline characteristics and why it suggests that these patients are more aggressive disease?
Daniel George: Yeah. Neeraj, this was a global study. And so you can see we ended up with 500 patients or so per arm, average age 67, 68. But what we see in our typical frontline setting, and what you'll see here is that the characteristics of this patient population are pretty similar to the overall population that we're going to see in our other registrational studies like LATITUDE or ARCHES or TITAN. And so there's not a phenotype to loss of PTEN. You can't say, "Well, gee, if you don't have liver mets, then probably your PTEN is intact." No, only visceral mets here or lung and liver mets were only about 20%. So 90% of these patients had good old garden variety bone mets like we see as the most common site. So all this is to say maybe there's a few younger patients, but overall this is a pretty similar population.
I will point out as a global study, there were a number of Asian countries involved. We had 37% Asian countries, which is really interesting and remarkable, but this was not a US European focused study. This was really a global study. But as a consequence, we had very few Black patients. So that, again, we struggle with that population. But overall, very high degree of Gleason 8 or above, almost 80%, but not 100%. Some of these patients were less than Gleason 8. And again, high volume, low volume are shown here as well. So sort of a representation across that. So also say is that you can't predict who these people are.
This is the progression-free survival endpoint. And I'm going to point out first off, this is much lower. This is a much lower median time to progression in our control arm, 25 months for abiraterone, prednisone than what we typically see. And again, it's not a particularly higher risk or higher volume population than the other studies. It's the biology of loss of PTEN that's the real prognostic marker here. And it's really showing us the unmet need. If half our patients are progressing in two years, we're almost back to where we were with ADT alone and a need to improve this. We improved the median progression-free survival by 7.5 months. It was statistically significant, about a 20% improvement in hazard ratio of 0.81 with about a 33-month median progression-free survival for the arm with Capivasertib.
Now, as I mentioned, there was hierarchical secondary endpoints and overall survival just hasn't been read yet. We've only had about 26% of events. There is a trend towards improved survival, but these are the early deaths. These are the patients that probably progressed fastest on abiraterone ADT. And with Capivasertib, it may just not be enough in that population. As we follow this up, these curves may widen and we see that with a lot of studies. So I think we need to hold judgment still on the overall survival, and they will read this out again when there are 50% death. So that will, I think, give us a lot more information regarding the lasting effects of this therapy.
Neeraj Agarwal: And then during the... Sorry, just that reminds me to talk about what Dr. Karim Fizazi, when he was presenting the data in the ESMO meeting in Berlin, he discussed the OS by the degree of loss of PTEN.
Daniel George: Yes.
Neeraj Agarwal: Can you please tell us more about that?
Daniel George: Yeah. So what was really interesting is while this study was going on, some of the analysis is coming out of the castrate-resistance study were being published and we're able to see the greater effect of loss of PTEN even above 90% in the IPATential150 study. So we did this post-hoc analysis where we looked at cutoffs for overall survival and progression-free survival of 90, 95, 99, and 100% loss of PTEN. And I'm just going to show here on this click, the 100% loss of PTEN. And in the light blue and light pink, you can see all the other analyses. They kind of trend down. What you see is that the loss of PTEN at 100% didn't really affect the Capivasertib arm. Those blue curves are largely staying up together. But what you see is in the control arm and ABI ADT, we're seeing these progressively worse overall survivals all the way down to where it's almost at the median there for the 100% group, and it's around 33 months.
So this is a really poor survival for this group, and we're seeing a proportionally greater survival benefit associated with that loss of PTEN and 100% of patients when we add the Capivasertib. So it really does suggest that PTEN testing, it's not a plus or minus. The degree of loss of PTEN is going to be important. It's going to help us understand not just who should get Capivasertib, but who's likely to get the greatest benefits because they're likely to have a even worse outcome than the patients who have only partial loss of PTEN.
Neeraj Agarwal: Thank you, Dan, for summarizing the design, endpoints, and the results of the trial for us. Coming back to Neal, we saw that one-third of these patients experienced radiographic disease progression without meeting the criteria for PSA progression. What does it tell us for our own practices?
Neal Shore: Yeah, that's a really good point. I think we have been lulled oftentimes into a false sense of security regarding the predictive nature and the accuracy of PSA. It's a good marker, and in different populations, it can be excellent. For example, post-radical prostatectomy, you expect the PSA to go to undetectable. It's probably its most clear utilization there. There's variance in terms of the definitions that we've struggled with over the years in post-radiation therapy. We think about challenges with PSA regarding doubling time in the biochemical relapse. We can debate that. We have some challenges in understanding though in our patients with metastatic disease that a Nadir PSA is not always a sine qua non for disease stabilization or deep remission.
And we've all been burned by not adequately evaluating from an imaging standpoint, assuming the patient's not having any symptoms, just because their PSA nadired and stayed low, and then a scan's not ordered, and whether it should be at three months, six months, or annually, and suddenly the patient's blossomed from their baseline imaging, and they now have not only increased in intensity of lesions, but they have new lesions and worsening lesions, and they may not always be symptomatic to tell us, "Get the imaging sooner rather than later." And this is another study that clearly demonstrates that. So I know many of us out there are like, well, I don't want to order unnecessary amounts of imaging. Some patients have fret over their radiation exposure, and maybe there's some cost issues. But at the end of the day, if someone's progressing and we're missing it because we're not getting a cadence of the appropriate testing, then it's kind of shame on us as the healthcare provider.
Neeraj Agarwal: And I'll be honest with you, Neal, and Dan and David, I do not obtain CT scans and bone scans every three months for all of my patients, especially the PSAs going down if they feel better, especially they are traveling long distances to see me. Many a times I do scans once or twice a year and tell them that it's okay. But looks like, David, practice may need to be changed for at least some of these patients. So what is your take on this one?
David Morris: I think for imaging intensity, I think we're looking for markers that suggest that we need to be doing something more. And PTEN is certainly one of them based off this trial data. I would say the HR mutations are another one that suggests that we know things are probably going to go worse than the average RB1. There are things that we've seen mutation status suggesting it's not going to act like a classic prostate cancer, just like urologists are aware of the dreaded neuroendocrine conversion where PSAs are low and disease is blossoming. And so we're just starting to recognize that some of these are signals that we may catch early on if we do somatic testing upfront.
And so if you get that data and you think, "Oh, I don't have a drug to pair with it." Well, at least in some of those cases, maybe we think we need to pair imaging with it where we haven't been doing a very good job of that. And I'll freely admit that we often just get imaging like, let me look through the chart. Oh, it's been a while. It's almost a year. Let's just get new X-rays. But maybe we have a better defined cadence on that if we see certain mutations.
Neeraj Agarwal: That's great. A great discussion on imaging and how often we should be doing it. Now coming back to testing, regardless of whether we use Capivasertib in these patients or not, right now I don't have the drug available, but if I have PTEN status known to me, I think I'm going to change my practice in terms of monitoring, as we just suggested, prognostication and counseling and telling patients what to look for. Even slight pain in the back may be ominous, more ominous than they did if they did not have PTEN. So David, coming to you, Neal did mention this briefly, how we are going to make it more commonly testable alteration in the clinic, NGS or IHC or both. How are we going to do it in the clinic?
David Morris: For places that have IHC available, that is a fairly cheap, easy thing to start to request. I think the easiest foray into the urology offices is to have commercial partners and vendors that do a good job. It is very challenging. These don't integrate with our EMR software. Everything requires going to a different system to get it ordered. So simplifying that workflow in clinic, that's going to be the thing that actually gets the needle to move. One-stop shopping to be able to get both germline and somatic and potentially IHC. But I want groups that hear like, "Oh, I have to do IHC." That's not necessarily the message. And I think there was some retrospective analysis from IPATential that actually showed NGS loss is actually just really high IHC loss for PTEN.
And so if you've got NGS, that's actually the first step. So I mean, if you don't get IHC, don't feel like you've done a huge disservice. Just get NGS to get the information. And so I think that's still very actionable and equates fairly well with really high IHC loss of PTEN. So that's how I'm going to view it internally is I just need to try to do a good job of getting upfront somatic testing, NGS, IHC if available on basically all those metastatic patients, and then go from there.
Neeraj Agarwal: And Neal, regarding the NGS testing, you mentioned that you are doing NGS testing for pretty much all patients with advanced prostate cancer, but we consistently see in the US, at least from in 2023, about 35% of our patients were tested with NGS testing. What are the reasons you think out there why patients are not being tested, NGS, IHC or both?
Neal Shore: Yeah. Well, yeah, like I said, I do it at 100%. And if it's not 100%, it's because we somehow missed it. It was an error on our part. So if anybody has metastatic disease, de novo recurrent with sensitive biology and everyone with resistant, everyone gets germline, even if their somatic is positive, because I want to know to potentially inform their family members. And now I've asked all the commercial folks that I work with, Foundation, TEMPUS, Guardant, Caris, you got to do IHC. And if I'm sending it, I want IHC PTEN, I got to know it. And so for really those great curves that Dan showed you in the CAPItello's trial, there's the 90%, 95%, 100%, it's pretty dramatic. So I think some of the reasons, Neeraj, has been, the earlier reports were a bit excessive, 15, 30-page reports and too many acronyms, little words, word salad–like for the busy community practitioners.
I think those companies have gotten better with just their reports and the structure, sort of simplification of it. It's a complicated topic, but giving the clinical utility. And then as David said earlier, we now have precision personalized therapies. If you have homologous recombinant mutation, if you have really high MSI or TMB high, and now potentially, if it gets approved, PTEN loss, you now have clearly biomarker correlative therapeutics by the FDA and covered by Medicare. So you need to be doing that if you're considering yourself a center of excellence, whether you're in community or in a tertiary academic center or you're med-onc or a rad-onc or a uro-onc. I mean, the burden is on us to do this. I mean, otherwise when people get their second opinion, they're going to say, "Well, wait a minute, Dr. Shore, why didn't you send the testing because Dr. George did?"
It's the first question he asked when I walked in the door. I'm not kidding a little bit, but not really. So I think that our colleagues really need to get on this. And I recognize that sometimes there can be some expense for some patients in different parts of the world. And sometimes there's concern about your level of education to do it correctly, to give that dialogue with patients. Some people start to fret about, "Well, do I need to have a certified genetic counselor?" And I think the answer to that is clearly not. It's a nice to have. It's not a have to have.
Neeraj Agarwal: Yes.
Neal Shore: Some people still are bothered by some medical legal issues regarding this. And I think these are all surmountable issues.
Neeraj Agarwal: Yeah. Dan, before we get to the conclusion of the trial from you, any suggestions for us on how often we should be doing imaging studies in these patients who have high degree of PTEN losses?
Daniel George: No, I think this is super critical. So I think the main reason to do PTEN testing now is to identify this group of patients. Like I said, this is not a needle in a haystack. It's 25% of our population. This is a big chunk. But for the folks that don't have it, probably it's going to be overkill to keep doing that imaging every three months. For the people who do have loss of PTEN, especially for the first three years, that's what we're seeing a lot of these events is in the first two, three years. So for the first three years, I would do that even with that low PSA. And it's not that they had zero PSA. Most of those patients had a very low PSA, but maybe it didn't really start going up and yet they had disease progression. So they hit a nadir usually within six to 12 months, and then it's usually fairly flat, but that's the time that this progression is likely to happen.
Sure, there are going to be some that happen right off the bat, and those are bad, and we kind of know who those people are, but it's these patients that get a good PSA response that we get lulled into missing. So year two, year three are really the critical years not to lapse on your imaging if you've got loss of PTEN. And if you don't, then that's probably a group that's 95% PSA is going to be the leader indicator of progression first. So this gives you a way of dichotomizing your population so you're not overburdened with all this imaging, but it's really critical on this group of patients not to give up the imaging in those later years.
Neeraj Agarwal: Any final take on the results of the trial before we conclude our discussion today?
Daniel George: Yeah, just two things. One, this is still early. This was an interim analysis looking at the primary endpoint, RPFS and it was positive. This is a population we've never prospectively studied before. It's the first drug. It's an on target drug and it was positive. It does have side effects: diarrhea, anemia, and some hyperglycemia. But these are things that are easy to know. You know when you have diarrhea, it's easy to measure anemia, it's easier to measure blood glucose, and they're fixable and reversible. You take the drug four days, you're off at three days. So this is a manageable toxicity. Now, there's 20% of patients that had to come off, 18% that had to come off treatment, 6, 7% on placebo. So some patients won't tolerate, but the majority of patients, you figure out that dosing and then they're good and they can stay on. So this is not like giving docettaxel chemotherapy where we're going to look at some of our patients as unfit for it.
This is tolerable therapy, particularly in this population that's at risk. So if you see it, if you have access to the drug, we're going to want to use it. Even if it's not a home run, it's a building block for other things. And I think the second thing with this, Neeraj, is what you brought out is the degree of loss of PTEN because I think if within that 25%, there's 10% that have this very high degree, 95 to 100% loss of PTEN. That's the group where we're going to get an outsized benefit in OS, and we absolutely don't want to miss that population. So I think recognizing this is not an all or nothing. It's a graduated thing. It's a drug that we can learn to manage. The breast oncologists have done so, we can do so as well, and this is the only opportunity to use it. You can use chemotherapy later, we can use our other radioligand therapies later. This would be the only setting that you could use, the Capivasertib, that AKT inhibitor.
Neeraj Agarwal: Thank you, Dan, for such a nice summary of the results of the trial and the implications of these results. Well, that concludes our discussion. Thank you, and congratulations to you all for conducting such a large phase three trial, positive phase three trial. Kudos to you. And I think we already are seeing changes in our practice, at least in terms of testing. So testing, let's test for these patients, look for these patients because they need more often imaging studies. This will alter their counseling, prognostication, and hopefully a therapeutic intervention soon to be available, hopefully. So thank you again for being-
Daniel George: Thank you, Neeraj. Happy to be here. Thanks.