Read the Full Video Transcript

Alicia Morgans: Hi, I'm so excited to be here today with Dr. Mark Markowski, who's joining me from Johns Hopkins University, and Dr. Emmanuel Antonarakis, who is joining me from the University of Minnesota. We are so excited to talk about a new publication. Thank you both so much for being here with me today.

Emmanuel Antonarakis: Thank you.

Alicia Morgans: Wonderful. So, guys, you've both been participating in some really exciting work around giving testosterone in high doses to patients with prostate cancer and trying to understand how we can really reset things and hopefully keep people feeling well while we're controlling their cancer. Emmanuel, tell me, what's the motivation for this particular trial?

Emmanuel Antonarakis: We've talked about high-dose testosterone on this show before. It was the invention of Dr. Sam Denmeade, and it's the paradox of using high doses of intramuscular super physiological testosterone to induce responses in a subset of patients with metastatic prostate cancer. In this case, Alicia, the notion came from a clinical observation, and it was that we had three patients between us collectively at Johns Hopkins who had received bipolar androgen therapy, this high-dose testosterone on a clinical trial, came off of the trial for whatever reason, and then were subsequently treated with an immune checkpoint inhibitor. In one case, it was nivolumab; in one case, it was pembrolizumab; in one case, it was a clinical trial. And we had these three anecdotes where we saw very profound and durable responses to immune checkpoint blockade, which followed accidentally prior treatment with bipolar androgen therapy. We looked carefully at the genomics of those patients, and none of them had microsatellite instability or a high tumor mutation burden.

It got us wondering whether there might be something that was being induced by the bipolar androgen therapy that was subsequently sensitizing those patients to an immune checkpoint, and we decided to design a trial to prospectively test the hypothesis.

Alicia Morgans: Fantastic. And I should say that this was published in Nature Communications, so it didn't just get you interested. I think it got a lot of people really interested. And so, as we kind of get to that, I'd love if you could, Mark, can you just explain the design of the COMBAT trial, where it was conducted, how it was funded, so that we know kind of how patients were treated before they were given these immunotherapies?

Mark Markowski: Yeah, absolutely. This trial was a single-arm phase two study where we enrolled 45 patients that had advanced metastatic castration-resistant prostate cancer. The few requirements that we had were that patients needed to have had prior novel AR-targeted therapy. So, all patients had to have that. We allowed prior chemotherapy, and we had no limit to prior therapy to go on study. The other requirement we had was that patients needed to have biopsy-able disease, and what do we mean by that? We preferred soft tissue disease, so we were able to get kind of serial sampling of tissue to better understand what we were doing on the study. And as I said, it's a phase two study, and all patients received bipolar androgen therapy for three cycles as a lead-in. Everyone got testosterone alone for three months, and then we added nivolumab, an immune checkpoint inhibitor, for the rest of the study. So, for cycles four and beyond, patients would get testosterone and nivolumab. The primary endpoint of the study was a PSA50 response rate, and we had other secondary endpoints such as rPFS and overall survival, which we reported in the journal.

Alicia Morgans: Fantastic. Emmanuel, really, what were the results from this study, which of course, again, I think people find really compelling and interesting to be published here and also to be focused on this bipolar androgen therapy?

Emmanuel Antonarakis: As Mark mentioned, we had three months of high-dose testosterone, and there wasn't progression. After the third month, beginning on cycle four, we continued the high-dose testosterone, added the nivolumab on a monthly dose. We saw a PSA 50% response rate of 40%. And in our prior studies of bipolar androgen therapy alone, the PSA response rate was about 25 to 30%. It looked like it was numerically higher, although not double. We also saw an objective response rate of 24%. 42 of the 45 patients had measurable disease. As was mentioned, a biopsy was required. And of those 42 patients, 10 of the 42, in other words, 24%, had an objective response. Now, one thing that we asked ourselves and also the reviewers at Nature Communications also asked is how many of those responses occurred anyway with the first three cycles of testosterone? In other words, did the nivolumab add anything?

And of the 18 PSA responses, 16 of the 18 occurred anyway before the nivolumab, but two occurred after the addition of the nivolumab, which had not happened by that 12-week mark. In fact, in one of those two, the PSA was rising on the high-dose testosterone and then only began to decline after the nivolumab was added. And with the objective response rate, there were also three patients. So three of the 10 patients who met the objective response criteria. That only happened after the nivolumab was added. So the conclusion to me, at least, was that although there was a robust response to the bipolar androgen therapy alone, we could slightly augment that a little bit more by the addition of nivolumab at cycle four. The last thing, of course, is the safety. We can't talk about efficacy without hearing about the safety. We had 11% grade-three events, which was favorable in our opinion. And most of the grade-three events were fatigue, peripheral edema, and a syndrome of inflammatory musculoskeletal pain, which is a known reaction to the super physiological testosterone.

We did have one patient who developed pericarditis, which we believe may have been an immune pericarditis, so it was one out of 45 patients. But again, with the 11% grade-three, grade-four AE profile, we thought that was acceptable.

Alicia Morgans: So, Mark, I want to just ask you, one of the things that Emmanuel mentioned is that patients had to get biopsies within this study, and that can be actually really hard in clinical trials, particularly in prostate cancer clinical trials where we find that these are older patients, they may have bone-only disease. Tell me a little bit about that because, as I said, it can be challenging and really congratulate you and the study team for making sure that that got done so we could get all of the correlative information that was necessary.

Mark Markowski: Yeah, absolutely. I think what was key to this study were the collaborators that we were working with. So this study was open at Johns Hopkins, obviously, but also open at UCSF with Dr. Aggarwal as the PI and also Dana Farber with Dr. Taplin. And both of those sites were very supportive of the study and worked hard to find patients to meet that eligibility and talk to them about the pros and cons of treatment, and were able to find patients that fit the eligibility, and kudos to those patients for undergoing those biopsies, serial biopsies. So they had to get biopsies before they started on study, and then three months in, which is a big ask for patients. And so we're grateful to all the patients that participated, that were willing to do it. The collaborators were very important. And then on the Hopkins end, Dr. DeMarzo and his pathology team were very supportive of this study, and they would go to the biopsy suite and process the tissue kind of immediately right after the biopsy.

We were able to get formalin-fixed tissues and flash-frozen tissues and really preserve all that DNA and RNA and get as much data as we could. I think the key to this study were the collaborators that we were working with and their enthusiasm. And then obviously, the Prostate Cancer Foundation supported us throughout this with the challenge award. We were able to do all the correlates that we wanted to do. And I think together with the financing and the collaborators and the scientists involved and obviously the patients, it ended up being a successful venture, and we were able to accrue those 45 patients in about two years. And that was during peak COVID when it was hard to get patients into the clinic and into the biopsy suite. So there were so many people that helped us out with that, and I think it was a huge team effort, and we couldn't have done it without everybody involved.
Alicia Morgans: Fantastic, fantastic work, and congratulations to you again as a study team and certainly thank you to the patients. Just as we think about moving on to biomarkers in just a moment, I would love to hear from you, Emmanuel, because I think it's probably a pretty select population that went on to this trial. And I would love to hear how you think about selecting patients to engage in a trial like this and generally to engage in treatment with bipolar androgen therapy when we know that there can be concerns, complications, there can be a downside, there can be risk. How do you think about engaging patients in this kind of treatment? In this kind of trial?

Emmanuel Antonarakis: Alicia, this was the eighth bipolar androgen trial that we did. And in the first seven, we really didn't know why the 30% were responding. Our big mission in this trial, in addition to the combination with immunotherapy, was really to take a deep dive and learn as much as possible from a smaller number of patients. And so that was truly at the heart of our desire here. And you are correct, in prostate cancer, the majority of patients, I would say 60 to 80%, would have bone-predominant disease where they either don't have a measurable lesion at all, or the measurable lesion might be one that they're not willing to undergo biopsy, let alone a second serial biopsy 12 weeks later. So, are these patients reflective of the standard typical metastatic castration-resistant patient? Probably not. But on the other hand, when we spoke to the patients about the importance of understanding the underlying mechanism, how this worked, many of those patients wanted to be our partners and took pride in participating in something that would advance the field and also help us understand how this paradoxical therapy works.

Alicia Morgans: Great. Well, Mark, back to you. I'd love to hear what you learned. What were the insights that we gained in terms of biomarkers of response or nonresponse? What did you learn from this study?

Mark Markowski: I think one of the best parts about the study was that we did collect a lot of specimens. We did the biopsy specimens obviously, but we collected peripheral blood, rectal swabs. I mean, anything that we thought was going to be important we wanted to collect so we could study in the future. And I think we're really just scratching the surface now, kind of learning what we've got from the COMBAT study. I think at least the initial impression, and a lot of work was done by Laura Sena, David Sanin, Srinivasan Yegnasubramanian in our next-generation sequencing core to get RNA-Seq and whole-genome sequencing to kind of really try to nail down some mechanisms here, not just for the combination, but for the bipolar androgen alone, as Emmanuel alluded to. And I think some of the key findings, at least right now, and I think we're still learning, as I said, was that bipolar androgen was quite the potent MYC inhibitor.

Okay? We didn't realize that, I think, at the time until we were working with Angelo DeMarzo, looking at the IHC where all the pretreatment samples had very high levels of MYC that you can detect at a protein level, and then three months later that protein level was virtually gone and you don't need to be a pathologist to kind of see those differences. I mean, it was quite profound. I think the effect of testosterone on MYC suppression was quite impressive, and I think we're going to build on that finding. And then also understanding it's not just the androgen receptor expression and testosterone levels that matter, it's kind of how that androgen receptor signals. We've kind of worked with an androgen receptor signature to kind of better predict who may respond to bipolar androgen. I think those are two very informative findings, at least initially upfront.

And then we haven't even gotten into the immunotherapy part of this, and we saw very profound changes in an inflammatory gene set across responders. I think one of the other interesting findings we had was looking at the immunotherapy component to this, and we found that in those patients that responded, so those patients that achieved a PSA or an objective response, they had pro-inflammatory expression changes in their RNA-Seq data, which we didn't expect but was quite profound. We also found that those patients that responded had very high levels of CD4 and CD8 T cells, particularly those that were PD1 positive. So, there is some effect of the bipolar androgen on the immune system. And what we're trying to figure out now is how much are we harnessing with the addition of nivolumab? And I think additional studies are going to help clarify that.

Alicia Morgans: That's really, really fascinating. And I'm going to just kind of shift gears and go to Emmanuel now and just say, I know you've touched on this and it sounded like you were kind of hemming and hawing a little bit, but bottom line, did the nivolumab add anything?

Emmanuel Antonarakis: We want to put on our objective scientist hats sometimes, and it's not clear-cut, Alicia, I wish it was. The interesting data that Mark presented is that if you have high PD1 expressing T-cells in your biopsy at baseline, those patients are going to do better, but they may do better simply with that alone. And that we didn't see a direct correlation between an increase in any of those T-cell subsets, which only occurred in the responders and not in the non-responders. In fact, those analyses were not statistically significant. I think that we have some ideas about future clinical trials that can answer the question clinically, but I don't believe that we have, as of yet, unequivocal molecular or immunologic evidence to say for sure that the immunotherapy added something.

Alicia Morgans: I have to say, and just sort of put it out there, it was an unfair question, but I so appreciate you answering it because it's a difficult one to answer, and I knew that when I asked. So thank you for being a good sport, Emmanuel. You always are. And Mark, let's just round this out. Where do we go from here? What other combinations are you thinking about? And as you were thinking about that in general, potentially being utilized in clinical practices because this is something that medical oncologists could use if they wanted to just write a prescription tomorrow. What are some of the things that you're cautious about? What are some of the things that you have to think about as you're cautiously trying to deliver this type of care in practice right now?

Mark Markowski: I'll start with the latter first. And I think over the last several studies that we did with bipolar androgen, there are a lot of safety regulations built into it that there are certain patients we are not going to put on bipolar androgen. Those patients that have very large visceral metastases, we would not put them on. Patients that have pain requiring narcotics, those patients would be ineligible because we're worried about safety, we're worried about the possibility of tumor flare. Although we haven't seen that in clinical practice, we want to make sure that we're being safe. So, large visceral metastases, pain requiring narcotics, any potential clinical scenario that may be made worse with testosterone, urinary obstruction, right? If you're pending urinary obstruction, we're not going to put you on study. There's a lot of safety built into that. So if providers are out there thinking about doing it, we're largely thinking about asymptomatic patients that have progressed through a number of standard-of-care options, and we think this is a reasonable safe option under the right supervision.

I'll say it that way. In terms of different combinations, I think we'd like to piggyback off of the COMBAT study and do a randomized study to say, look, this was single-arm, the response rate was good, but is it any better than some standard-of-care option for patients? And I think that's our next step here, is to see whether or not the bipolar androgen with an immune checkpoint inhibitor can be better than some control. And the one reason I think why I'm optimistic about that approach is that it's not necessarily the response rates, but it's the overall survival data that we saw coming out of this study. I mean, we had an overall survival over two years, and many of these patients were pretty heavily pretreated, so half of the patients got chemotherapy. A third of patients went through two lines of AR-targeted therapy and chemotherapy.

It's kind of a heterogeneous population in terms of treatment, but the overall survival was, I think, impressive based on that population. Can we do a study where we're looking at overall survival as opposed to response rates or even rPFS and look at this more definitively to say, maybe we cannot quantify the effect of the immune checkpoint inhibitor in a traditional sense of response rate. Maybe there is some benefit on the backend with survival. I think we'd like to do it in a randomized fashion. And then we're always interested in novel combinations with bipolar androgen therapy. I think that the finding of MYC inhibition or MYC suppression with testosterone is quite exciting for us, and we're looking for a combination where we can kind of take advantage of that and maybe even accentuate that effect on MYC. I think those are some of the areas that we're looking to move forward in.

Alicia Morgans: Well, I really congratulate you, and I also just want to emphasize the importance of the ongoing investigation of how we can use a different approach, a different kind of mantra to treat patients with advanced prostate cancer, that maybe we don't have to use castration in every patient. This is a complete shift in mindset, and you and your team are helping us think about this and to do it safely in a way that I think will really help patients in the long run. Congratulations on this wonderful publication and certainly congratulations to the patients who made this work possible. I appreciate both of your time.

Emmanuel Antonarakis: Thank you.

Mark Markowski: Thank you.