Elisabeth Heath: Thanks, Neeraj. I'm Elisabeth Heath and I'm at the Mayo Clinic.
Hope Rugo: And I'm Hope Rugo, thanks for having me here. A little bit of an outlier, I would say misfit. I'm a breast cancer oncologist and a Professor of Medicine at the City of Hope Comprehensive Cancer Center. But I was one of the co-PIs of the CAPItello-291 trial using capivasertib in breast cancer, so that's my role here. Thanks for including me.
Neeraj Agarwal: And we are going to learn a lot from you today. Dan?
Daniel George: Yes. Hope, you're never a misfit. Hi, I'm Dan George. I'm a GU medical oncologist at Duke, and part of the steering committee for the CAPItello-281 study, which was the pivotal trial in prostate cancer to evaluate capivasertib plus abiraterone and prednisone versus abiraterone, prednisone, and placebo in prostate cancer.
Neeraj Agarwal: Thank you, everyone, for being here. That's a very nice segue to talk to Elisabeth, about your viewpoints on how you approach treatment of patients with metastatic hormone-sensitive or androgen modulation sensitive prostate cancer.
Elisabeth Heath: Yeah. It's such an exciting time because we went from here, let's just get rid of your testosterone, to, we have to intensify, intensify, to now a more precision approach. I mean, this is really exciting. Obviously with many of the work you did with BRCA2, for example, we're starting to figure out, who are these patients that can be evaluated in different buckets? PTEN loss is really bubbled up to the top now as a major driver of aggressive cancer. So, I think having that as part of the conversation about what to do when you're newly diagnosed metastatic, is critical.
Neeraj Agarwal: So that reminds me to, before we go to the CAPItello-281 trial, historically we have been ordering CT scans and bone scans and then PSMA PET scans recently came, but molecular markers are still not emphasized enough. We just saw recent real-world data. Less than 30% of patients had NGS testing and now we are talking about immunohistochemistry. Before we get there, how do you approach biomarkers, especially the aggressive ones in your care, and what has been your experience?
Elisabeth Heath: We have definitely spent so much time, as I'm sure Dan and many of our colleagues just figuring out, what does that algorithm look like in your practice? And this is us in oncology and it's also us in urology, since a bulk of these patients are seen there. So, the messaging at least is out. Now, whether implementation as you mentioned is out, that's still a lot of work. Now there's this push though for immunohistochemistry, which is much more widely available. Many pathologists do IHC stains on many different things. PSA for one, as an example.
So I think we still have to do the proper messaging of not just genomics or genetics and even that's interchangeable terms, but what else about the cancer? So it's a heavy lift, I think, for our prostate cancer community to really get this right, because we're barely just getting the messaging of testing right. So, we have to go all in to really get this right.
Neeraj Agarwal: Absolutely. Unless we know, we don't know how those patients' cancer are going to behave down the line. From CAPItello, first of all, congratulations, Dan, for a positive phase-three trial, CAPItello-281 trial, and hopefully we'll have the agent, capivasertib, be available soon for our patients. Before we get to the trial and design, why PTEN matters, PTEN loss matters for patients with metastatic androgen pathway modulation sensitive prostate cancer?
Daniel George: Yeah, Neeraj, this is a really important point. When we treat patients with metastatic prostate cancer now, and we use these combination therapies that we've talked about, in general, we're seeing really good PSA responses in these patients, but there's a subset of patients that get an incomplete response. And what's concerning for us is when we look at the early progressors in these trials and patients on these drugs, there's a disproportionate number of these patients who are progressing without a significant PSA rise. Their PSA is still low, and yet they've got progressive disease. It's a somewhat different phenotype than what we saw when patients were getting ADT alone. And what we're finding is 25% of patients that present with de novo APMS, this androgen pathway modulation sensitive disease, these patients have loss of PTEN, and we're defining this as 90% or greater loss of PTEN on their immunohistochemistry.
So, it's going to be important for our pathologist now to really have either a reflex or a standard for checking PTEN status immunohistochemistry on our patients that have metastatic disease at diagnosis. And if we find that, recognize that that's associated with a worse prognosis. Retrospective data has shown those patients are more likely to progress sooner and die quicker from prostate cancer. And in CAPItello-281 in our control arm, we saw exactly the same thing. Patients treated with abiraterone, prednisone and ADT had a median RPFS of just 25.7 months, which is much shorter than the 33.7 months or so.
Neeraj Agarwal: Which basically validates our experience.
Daniel George: Yeah, absolutely, absolutely.
Neeraj Agarwal: So before we get to the CAPItello-281 trials results, please tell us, could you tell us more about the design of the trial and-
Daniel George: Absolutely.
Neeraj Agarwal: ... how you randomize patients and so on?
Daniel George: Yes, yeah. So CAPItello-281 was an ambitious study, recognizing that to identify these patients with 90% or greater loss of PTEN in this de novo metastatic setting, that if that's going to be 25%, we had to screen a lot of patients. So we screened 6,000 plus patients to identify what ultimately turned out a thousand patients, 500 and 500 in each arm, randomized to either ADT, abiraterone, prednisone, and placebo, or that combination plus capivasertib. And capivasertib is an AKT inhibitor, it inhibits all isoforms of AKT, it's very specific for AKT, and it's given in a very important dosing schedule. We take it twice a day for four days, and then three days off. So it has a built-in washout for these patients every week. And it's designed that way to really maximize the therapeutic index, that this helps to mitigate the cumulative toxicities with this drug, while maintaining that inhibition of AKT long enough in the setting of standard active therapy, in this case abiraterone and ADT.
So these patients were randomized one to one, blinded placebo controlled, and followed until radiographic progression-free survival. And that was defined as obviously radiographic progression-free survival or death. And that was the primary endpoint. And then overall survival was a key secondary endpoint, along with some really important secondary endpoints, including time to first symptomatic skeletal event, time to metastatic, what we call castration-resistant prostate cancer, now APMR disease, or time to subsequent therapy first chemotherapy. So, these were really important secondary endpoints to help solidify the results we saw with the primary endpoint RPFS.
Neeraj Agarwal: And how many patients were randomized?
Daniel George: 1,000 patients. So 502 or 504, around there. And the results were first presented at ESMO last fall, and again at PCF, and then this spring we had an FDA ODAC meeting.
Neeraj Agarwal: Could you share the results with us?
Daniel George: Yes, absolutely. So the primary endpoint, the radiographic progression-free survival, as I mentioned in the control arm was 25.7 months shorter than even what we had anticipated in this population. In our capivasertib arm, it was 33.7 months, so a seven and a half month improvement in the median progression-free survival. It was a hazard ratio of 0.81, a P value of 0.03, statistically significant P value. So, it met its primary endpoint.
Neeraj Agarwal: And did the radiographic progression free survival benefit increase in magnitude with increasing PTEN loss?
Daniel George: So this was a really important, interesting post-hoc analysis. And it was based upon the ipatasertib, another AKT inhibitor that was developed in the castrate resistance setting that showed efficacy but didn't change overall survival, was not ultimately FDA approved, was used in all comers, but they also looked in patients for PTEN status. And for patients with 50% or greater loss of PTEN started to see greater proportional benefit with that agent. So we looked post-hoc in our population in CAPItello-281, and what we saw was exactly the same trend. We saw improvements in the progression-free survival, not because the cipatasertib arm was doing better and better as we went up to 95 or 99 or even 100% loss of PTEN, but because the control arm was getting worse and worse. The higher that loss of PTEN, really, the worst the prognosis in these patients. And it's important to recognize that, that this is a relative prognostic factor and starting at 90% is bad enough, but going higher really suggests that these patients are doing even worse.
So, this is the population we focused on as really, as Elisabeth said, precision medicine, not just with the target, but with this level of loss of target is really associated with the worst actors. And even with those worst actors, adding this single AKT inhibitor changed the course of this disease.
Neeraj Agarwal: Absolutely. Elisabeth, when we saw the results presented by Dan and Dr. Fizazi and colleagues, there was something striking. Not only there was almost a year, if I remember well, year delay in radiographic progression-free survival with PTEN loss 100%, there were many patients who were experiencing radiographic progression without meeting the PCWG3 defined PSA progression. So if you relied on only PSA progression, then we would have missed the radiographic progression in these patients. And concerning thing is, with that observation in these patients, even in my clinic, I'm not doing CTN bone scans every three months. So, what are the implications in your clinic or for our clinic?
Elisabeth Heath: Yeah, and we may not know this right off the bat. All I know is it scares me that I don't know what's going on with the patients. Right? We rely on that PSA in this particular part so closely that we almost are like, "No, no, you don't need extra imaging. In fact, what will that really tell us?" And yet knowing this upfront, the answer is, "Oh, that's going to tell us a lot." So, we are probably going to have to be much more proactive, whether the question is CAT scan, bone scan or PET scans, I'm sure that will be yet another episode on UroToday, but what's critical for us is, you need some sort of imaging. I think the standard CAT scan, bone scan makes a lot of sense.
I think also the aggressiveness of this cancer, we're doing everything we can to delay that castration state, and that's really impactful I think from a patient standpoint because we can buy this much time upfront and getting that sort of course done correctly and then monitoring. I think patients are probably best served by us being that aggressive in its management for PTEN loss patients.
Neeraj Agarwal: Absolutely. Your paper, one of the most well-cited paper in clinical genitourinary cancers, showed that half of the patients with metastatic prostate cancer do not get second line of therapy or subsequent line of therapies. And this in my experience, if patients experience disease progression, there is a pretty high likelihood of them succumbing or dying because of their disease and not being able to receive next line of therapy. So, I think it's so important to recognize that radiographic progression without having to rely on PSA progression, because many of patients may present with cord compression if you are just relying on PSF progression. Was this your experience, Dan?
Daniel George: Absolutely, and I think you said it well. I mean, we see a earlier onset of these symptomatic skeletal events associated with the control arm. And it's frightening because these are, these pathologic fractures, these core compressions, even if we treat them, these patients aren't the same. They're debilitated by it, they're not going to get back to where they were. And if you have one, you're more likely to have another and subsequent one. So it's really telling us that this is a more aggressive, clinically aggressive course associated with this disease. And we see that capivasertib used upfront in this population is changing that course. So it's a really important thing to recognize.
And I think to your point as well, we tend to be reactive in this space and ultimately, I think what we're going to need are tools to help us really predict how to be a little bit more proactive in this.
Neeraj Agarwal: And we won't know in which patients we need to be proactive unless we know the status.
Daniel George: That's right, that's right.
Elisabeth Heath: Right, yep.
Daniel George: Of PTEN loss.
That's the first proactive thing is to check for loss of PTEN.
Neeraj Agarwal: Before we go to Hope, would you like to tell us about the side effects you encountered, you experienced while treating these patients on this trial?
Daniel George: Yeah, Neeraj, this was really concerning for us because there is a profile that is well known associated with loss of AKT, that's what we call an on target side effect profile. And this includes hyperglycemia, rash, and diarrhea. And so there's no avoiding it when you block this target, you're going to get some of these side effects in most patients. So, recognizing that it's really important for clinicians starting this medication to educate patients about, to prepare them for that with Imodium, with topical steroids and with blood checks, even if they don't have diabetes, but especially if they're pre-diabetic or have some history of diabetes. As long as they weren't on insulin, they were allowed on the study as their hemoglobin A1C was less than 8.0. So, we did take some of those patients. I don't want to make it sound like we can't treat diabetics, we can, but these are the patients we have to be monitoring from day one very closely. And we learned on this study.
Unfortunately, it's an international study, it was launched during the pandemic. These are patients that probably weren't monitored close enough. And I think we can do a better job knowing now what we do, and Hope can speak to this even better, but I think that these are the three side effects that... There are other side effects, but the most concerning ones. And if we control them long term, these patients can tolerate this drug.
Neeraj Agarwal: Learning from you, we are looking forward to learn how to manage side effects of capivasertib. I have not, like any other prostate cancer doc, we don't have experience with this agent. But in breast cancer, capivasertib has been approved.
Hope Rugo: It's been approved for some time. And I think capi is unique in many ways. We've struggled with managing the side effects of PI3 kinase inhibitors, including alpha specific inhibitors that seem to have a more efficacy than what we looked at before, which were more broadly active drugs. And there's a whole new class of drugs that where really the focus has been hyperglycemia. Are these drugs have as much hyperglycemia and rash?
So the thinking behind using capivasertib in this setting was really to block the pathway more effectively by starting after PIK3CA activating mutations and starting at AKT. And it's great to see that success in this trial. But there are notable differences, and I have to say that having spent quite a bit of time talking to people about managing the toxicity of capi, I can see there are some unique challenges to discuss with prostate cancer because unlike breast cancer, you give prednisone along with it, and we all know what happens with prednisone. So, we're always struggling with people who don't have hyperglycemia becoming hyperglycemic on prednisone.
So, I think you already mentioned these specific areas, which I think we all hold true, is when we're using these drugs that have specific pathway associated toxicities, we know them upfront. What we want to do is be proactive in management. So the first thing you need to do is to assess patients for their risk of hyperglycemia since about a third of the US population and similarly or higher in many other parts of the world actually have prediabetes even without a known diagnosis of diabetes. These patients have elevated hemoglobin A1C and often slightly elevated fasting glucose. We know those patients are at higher risk of hyperglycemia.
One of the hallmarks that capivasertib for us in breast cancer was less hyperglycemia than we saw with the PI3 kinase inhibitors. But when you add in prednisone, I think you really have to be very careful because you're giving two hits, we all know what that risk is. So here, you evaluate a patient at baseline and if they do have elevated risk, you need to involve an endocrinologist, at least I do. Unless you're really good at all those names they're singing on the television all the time, all the new oral hypoglycemic drugs, I think we really do need to involve a diabetics doc or an endocrinologist depending on where you are. And some of these patients who aren't even on the drugs are on three agents already. So that's the first thing, is just really be proactive and make sure you have good control when you're starting out if you're giving the drug with prednisone.
And then I think the advice to the patient is so important, you mentioned the education. These patients have cancer that is not curable, and so we want to be careful about telling them they have to eat some ridiculous diet that they can't manage. So you have to really talk about what the specifics are with diet that create issues for the patients, large carbohydrate loads and sugar, etc, that they have to be cautious.
The other thing that we've noticed in post marketing with breast cancer is that although we didn't see a lot of hyperglycemia, when patients got sick, so they had urosepsis, which women get, but it's a much higher risk in this population. So they're on prednisone, they might have indwelling catheters. They are often patients who may have reflux and other issues which increase the risk of infection. What we saw was patients who had infection could get, they start out with normal hemoglobin A1C, they could become hyperglycemic and get DKA, even. So what I started telling patients was, having had one patient whose glucose went to 450 because she had a minor illness, is, "You don't feel good, don't take the pill, then you call us." So actually, that works really well. These are short half life drugs. So you don't take the pills and you call, you can manage this really well with partnering with your staff and your patients.
The other two issues that you talked about, the other issues that are actually more common than hyperglycemia in a lower risk population are rash and diarrhea, because it's an on/off drug four days on, three days off, the rash I have found very easy to control. But some people aren't even here talking to my colleagues that some people have found rash a bigger challenge. So, I would think that maybe the steroid would reduce your rash. I don't know because we use steroids to treat the bad rashes, for example, associated with PI3 kinase inhibitors, like alpelisib or even the MTOR inhibitor everolimus and it works well. So, it may be easier to manage this population, but we use non-sedating antihistamines twice a day. And then you have to tell the patient, "We know the bottle says once a day, but it's okay to take it twice a day," starting the day before they start the pill, because I do think that stopping the rash before it starts or controlling it reduces the recurrence issue. It's like your cells, they've seen that antigen so they're ready to have a rash again. So, that works well.
And then the diarrhea again, we talk to patients. "Okay, you know if you're going to go out and eat a really big salad and it's the fourth day of your treatment or the fifth day, you might get diarrhea. So, take a half a Imodium beforehand, that works really well. You don't have to get constipated because you're not taking a whole pill." And then for some people, it's a small percentage who get more diarrhea, than we do have them take prevention. So for example, I know the patients on their two or three weeks of treatment, they're getting diarrhea on day three and four or four and five, we have them take that morning a half a Imodium. So that works really, really well.
And I think as we all know, the cornerstone to managing toxicities, which are difficult for patients or making their life more difficult and it's harder for us to manage is to dose reduce, hold drug, dose reduce. And it's not a trial, you can always go back up again. So, I found that those strategies work really well.
Neeraj Agarwal: Thank you, those were really helpful. Regarding dose reduction, how often did you have to reduce the dose on this trial?
Daniel George: Despite the fact that we had about 18% of patients have to stop the drug completely, we had about 30% of patients, we had 28% of patients that dose reduced. Even with that, Neeraj, we still saw this clinical benefit.
And I think as Hope said, this isn't a permanent thing. If you need to do that in the first six or 12 weeks to get the patients worked out and how to manage, and then you can ease back up if you need to. But these patients, I mean, this is going to be chronic therapy. And I think here having some therapy on board rather than none is really the goal. And you know you're hitting your target if you're having some of these side effects. So, it gives you some peace of mind to say, "Okay, we can do this even at these lower doses, we can still be effective." So, I don't look at that as like a compromise or a defeat. I look at that as just personalizing this to what the patients can tolerate.
Hope Rugo: Absolutely.
Daniel George: Where they are, meeting them where they are.
Neeraj Agarwal: Personalizing is a great word. Personalizing the dose to attenuate toxicities while maintaining efficacy-
Daniel George: That's right.
Neeraj Agarwal: ... if you will. And that's a great thing about on target toxicities, Hope, that we can reduce the dose and toxicities get better. And another really attractive aspect is, it's not a continuous therapy. So four days, after four days, patients feel better. And if they're really not wanting to start at the full dose, they can always reduce the dose.
So coming back to Elisabeth, after all this discussion, if I'm starting a patient on capivasertib once it is approved, what kind of test I should be looking at? Like we are already ordering CBC, CMP, CT scans and bone scans or PSMF head scans, what test we should be looking at doing in addition to everything else we do?
Elisabeth Heath: Yeah. I think again, the package insert here is very helpful. We've already learned from our breast cancer colleagues, the package insert's already been updated and really has that very proactive, frequent monitoring of that blood glucose upfront. So until you get some stability, and then you can peel off depending on how they're responding. But I think also in our community, we're used to not starting everything at one time. So one could start the ADT and the abiraterone, the prednisone, see how you do, and then you can weave in the capivasertib. So I think there's a lot of strategies to do the onboarding.
I think with immunohistochemistry, that might require a conversation, for example, with your pathologist and say, "How do we do this from an organizational standpoint? How do we want to work this in into the flow, because that is a different flow? Genetic testing is not adequate for what we need, it's really immunohistochemistry." So I think conversations have to be had in your own practice, but realistically, it's quite easy to implement and move forward.
Neeraj Agarwal: I was going to say, that IHC, we do it all the time.
Elisabeth Heath: All the time, all the time.
Neeraj Agarwal: So it's much easier to IHC than NGS testing, actually. Yeah.
Elisabeth Heath: Absolutely.
Neeraj Agarwal: Dan, any last words on how to monitor these patients for our colleagues? I know prognosis is worse, but we need to monitor them very closely.
Daniel George: Yes, yes, yes. So a couple of things. One, hemoglobin A1C, not necessarily routine for us to check, but we want to check that in these patients that we're going to be starting so we have a good baseline to compare it to.
I think number two, recognize the average time of onset was two weeks. So, have these patients checked every week. They could be checked by your nurse, they could be a phone call, a video visit, whatever, but touch points early on in the first month or so. Have your endocrinologist on speed dial, because you want to be able to make sure that they're ready, they can get them in if we need them, especially if it's somebody that does shoot up. And sometimes it's unpredictable, a non-diabetic can shoot up. So, just be ready for those kind of things.
On the cancer side, keep that imaging monitoring going. Even if we stretch it out a little bit over time, but just keep that going, especially for the patients whose PSAs don't go all the way down to zero. That's the group that I'm concerned about. We see that at a higher percentage of incomplete PSA response in this population of PTEN deficiency. So be aware of that and recognize that that's not a safety, having a PSA that's low but not zero. That patient still has residual disease, they're at risk for these kinds of symptomatic progressions.
Hope Rugo: And you made a really good point, which just to emphasize, which is that this is a more of a front loaded toxicity, not only, but mostly. So you see most of these side effects right up front and can manage them and the package insert talks about checking the glucose more frequently in the beginning and then you taper off once you know that it's stable. But then you remind the patient, they come back in a month. "Great, we're back for the month. You get a fever, don't take your pill." So, because I think that that's important because that could happen a little bit later in the course of the treatment.
Daniel George: That's exactly right.
Neeraj Agarwal: Those are great insights. Well, thank you so much for being here and sharing your knowledge and insights.
Hope Rugo: Thanks so much.
Elisabeth Heath: Thank you.
Daniel George: Thank you, Neeraj.