Alicia Morgans: Hi, this is Alicia Morgans. I’m a Medical Oncologist, a GU Medical Oncologist at Northwestern University. I am so excited today to have my colleague and friend, Dr. Maha Hussain, on the line to share with us some information about the PROfound study.
Dr. Hussain is a Professor of Medicine and the Deputy Director of the Robert H. Lurie Cancer Center at Northwestern University. And was the PI of the PROfound Phase III clinical trial that was looking at Olaparib in patients who were selected for various DNA repair defects. And thank you so much for taking the time to talk with me today, Dr. Hussain.
Maha Hussain: Well thank you Dr. Morgans. And you can call me Maha.
Alicia Morgans: And I will. So Maha, can you please share with everyone what the design was of this trial? Why did you do the trial, and how did you design it to answer the right questions?
Maha Hussain: Certainly, and thank you again for the opportunity.
So PROfound was designed with the intent of evaluating the role of PARP inhibition in preselected patient population. When we designed the trial, a lot of the data that was available with regard to the different DNA repair defect pathways were pretty much centered around BRCA1, 2, and maybe ATM. And this certainly is the case with other cancers where there is more experience with PARP inhibition such as breast and ovarian cancer.
So when we designed it, we wanted to look at the spectrum of a potential DNA repair, defective pathways, but at the same time we did not want to kind of dilute the results or confound things. And so we designed it with the intent of looking at two separate cohorts, what I would call the canonical cohort, which is BRCA1, 2 and ATM and then all of the other potential DNA repair defect genes in cohort B. And in total there were 15 DRD genes that we looked at.
The trial was intended to look at the second line setting because obviously in frontline setting we already have active treatment and the intent was to capture patients who may be less likely to respond to AR targeted drugs or are not potential chemo candidates or interested in chemo and so on.
So this was a second-line trial at a minimum and we allowed a fairly liberal eligibility criteria as long as a patient had progressed at least on one prior AR targeted agent. And the specifics were, of course, abiraterone with prednisone or enzalutamide. But in some… We wanted to keep it kind of liberal because there could be other AR targeted drugs that a patient might have received, second generation sort of AR targeted agents there.
And so that was sort of the general schema. And the intent was basically to stratify by certain prognostic criteria including prior taxane exposure and then the presence or absence of measurable disease, which will, of course, include visceral disease. And then the patients were screened first by a central FoundationOne next-gen sequencing assay. And then they were stratified if they had the DNA repair defect, qualifying mutation or alteration, they were stratified by prior taxane and measurable disease and then allocated to the respective cohort, cohort A which you had the BRCA1, 2 or ATM, and cohort B, which has the other alterations.
Alicia Morgans: Great. And, so these patients randomized then, to treatment with the other AR targeted agent that we would normally use or Olaparib and were followed for the primary endpoint of radiographic progression-free survival, right?
Maha Hussain: Correct. So basically the patients were randomized two to one, so to Olaparib versus physician choice based on the agent they may have received previously, and then followed with scheduled imaging and the primary endpoint was radiographic progression-free survival by a blinded into independent central review for a cohort A.
And of course, there were a series of secondary endpoints that looked at efficacy and other issues including adverse events, confirmed radiographic progression-free survival in measurable disease, time to pain progression, overall survival and so on.
Alicia Morgans: Great. And one of the things that I thought was so interesting is that unlike some recent Phase IIIs that we’ve seen, and now these were in the metastatic hormone-sensitive setting where we may not have had a lot of patients who actually had exposure to taxane, there were a fair number of patients who had exposure to taxane in this particular trial.
Maha Hussain: Correct.
Alicia Morgans: Is that true?
Maha Hussain: Yeah. It was actually not just taxane. In fact, it was a fairly heavily pretreated population beginning with basically obviously everybody had seen either abi or enza. And then in fact we’ve had a somewhat around 20% of the patients or so who’ve seen both abi and enza and then across the board, around 65% or so of patients have seen taxane chemotherapy and somewhat about a fifth or so, somewhat about 20% of patients have seen not only one taxane, they’ve seen two taxanes. So I think that is a fairly heavily pretreated patient population.
Alicia Morgans: Absolutely. And then when we saw the primary endpoint of radiographic progression-free survival, and I know we heard about this even in a press release months ago, but you presented actually a significantly prolonged rPFS and there were other endpoints like you said, even time to pain progression and others that were very clearly positive in this trial.
Maha Hussain: Yes. And so the radiographic progression-free survival, again by independent review, was pretty impressive. Where there is… The median was seven point almost 7.39 versus 3.5 but actually when you look at the hazard ratio, it’s 0.34 with a P value of less than 0.0001 which really when you look at the hazard ratio, this is like 66% reduction in the risk of death or disease progression because by the way, death from any cause counted towards the endpoint. And so it’s a pretty impressive delay.
The other part, which I thought was fascinating is that the separation of the curves, and I don’t know if your audience will have the chance to review the graphs, but the separation of the curves began very early in the process, literally at the first evaluation and continued even past a year into treatment. So I think that was really very impressive, considering-
Alicia Morgans: very high-
Maha Hussain: -a heavily pretreated patient population.
Alicia Morgans: I agree. And highly, highly meaningful. And in your presentation too, you really emphasized the clinical meaningfulness with the time to pain progression separating early as well. And those curves just remained apart for the duration of the time on treatment done. I mean it was dramatic.
Maha Hussain: Yes. And I think the other important point for the audience is the fact that based on the general, and again we’re, all of these comments that we’re making are in relationship to the primary endpoint in cohort A. But in that regard, essentially patients benefited irrespective of whether they’ve seen taxane or not, whether they’ve had measurable disease, visceral disease or bone disease, performance status, age, continent, which I think is important also. And obviously PSA level and things like that.
So I think that is the part that I think is interesting and important is that there is no subpopulation that did not benefit overall. Obviously no two people are alike, but the trends were across the board.
Alicia Morgans: Absolutely. So the one thing that everyone always asks about of course, but I don’t know that it’s quite mature, is overall survival. So can you tell us a little bit about that?
Maha Hussain: Yeah, so the data is not mature at this moment, but overall there is a trend for both the primary cohort, which is cohort A and the combination cohort with the, basically, the trend is about the three month difference in median survival with a hazard ratio of 0.64 in cohort A, 0.67 in the overall cohort. But again, the P value has not reached the significance level because the data is not mature. At this moment we have about a 38% maturity and the final analysis is planned at 60% maturity.
Alicia Morgans: Okay. Well, the other important thing I-
Maha Hussain: And I… I’m sorry I should point out this is despite the fact that over 80% of the patients in the control arm in both cohorts actually crossed over to Olaparib so that’s a… To see that kind of trend in the overall survival curve and again, obviously while it’s not mature, it is very interesting because this is despite crossover.
Alicia Morgans: I agree and thank you for emphasizing that point because we talk a lot about subsequent therapies, and it was very important in this trial that there was access for patients to actually cross over or within the context of the study and have access to Olaparib. And there have been multiple trials that have actually failed. It failed in the Phase III setting because of this crossover that the benefit was, I guess one could say two weak to actually withstand that crossover. But clearly earlier treatment makes a difference even when patients are receiving the PARP inhibitor at a later time point after progression, so really I think very powerful message.
Maha Hussain: I fully agree.
Alicia Morgans: And so just to… I know we talked a lot about the primary endpoint, which was kind of based around, or I believe was based around those DNA repair defects that you were calling the canonical defects, but other patients did respond. And I thought that was really interesting. You gave a nice breakdown gene by gene and showed that, and these were small numbers in some cases I think, but did show that there could be differential effects on these endpoints on radiographic progression-free survival at least by DNA repair defect. And I’m wondering if you can comment on that.
Maha Hussain: Oh absolutely. So I think when we talk about the genes that seem to have the majority of, or the more profound response, again, it’s the cohort A and primarily BRCA1 and 2, clearly cohort B, there were trends certainly in favor of Olaparib therapy, but the magnitude of the benefit and the hazard ratio is not the same as is the case in cohort A. Having said that, I would say the interesting analysis was the fact that this isn’t, again explanatory analysis because we did not know what we’re going to be seeing in terms of the rates and responses in other genes, but we included that in there because I think it’s a very important information. And what was interesting is that in reality for the whole patient population that was screened the frequency was that the 97% of the patients were randomized based on alterations in 8 of 15 genes. So 7 of 15 genes had very low frequency to even be counted from a statistical perspective, less than five patients. And so the eight genes are the BRCA1, BRCA2, CDK12, ATM, CHECK2, RAD51B, RAD54L and PPP2R2A.
Now these are, again, some of these numbers are very small and what is interesting is this BRCA2 continues to be impressive from the perspective of the number of patients and the rate of response. Certainly to my surprise was the CDK12 patients. We had quite a bit of patients with CDK12 and they seem to have better response compared to Olaparib than the control arm.
There are other trends in other genes that are highlighted in the presentation. Obviously at this moment this is very much a sort of post hoc type exploratory analysis and I don’t think one can make very strong conclusions because the numbers are small, but I think it’s interesting and I think it highlights the fact that when we design future trials, one has to not lump everybody together just because I think there are differences in responses and outcomes and I think it’s important to kind of break down the genomics upfront in terms of the eligibility criteria.
Alicia Morgans: I completely agree and I think that was one of the most fascinating pieces of this work that will also hopefully be translated into a future study so that we can look more closely and maybe enrich for some of these genes. Though of course, they are occurring in very small numbers. Some of them seem to have pretty pronounced responses and so I’m definitely looking forward to continued work in understanding the way that differential mutations may respond to different therapies.
Now, one more question I wanted to ask before I ask for your take home message for this study. People have been asking me, and I don’t know the answer to this, but they’ve said, “Look, if your patient has not had a taxane before and you’ve had progression on an AR targeted agent, they’re less heavily pretreated than the patients in this trial”, just say. Would you choose a PARP inhibitor as one of their earlier lines of options for treatment for metastatic CRPC? What, what do you think?
Maha Hussain: I think, like anything else, the decision is going to have to be a shared decision based on balancing risks and benefits and based on other factors, confounding factors like comorbidities, concurrent other medications, patient’s desire, do they want to take a pill every day and continue on therapy for months versus get a treatment every three weeks and for a finite period of time and so on.
And what I actually find is that sometimes we as physicians think we know what a patient wants. And yet when you sit down and have a conversation with the patients, they have different preferences. So to me, the good news is the patient has choices and there are some treatments that might be feasible earlier and when the disease gets out of control, it may not be easy to give because of issues of side effects and things of that sort. You know what I’m saying?
To give chemo you have to feel that it can be given safely. Somebody may be able to get a pill, but not necessarily undergo chemotherapy if their health is not great at that moment. So this is the kind of stuff that one has to balance.
Alicia Morgans: Absolutely. I agree and I appreciate all of those thoughts and really a message that we try to make sure that people hear is that when we do have choices, engaging patients to hear what they actually want is going to be, I think, one of the most powerful things that we can do as physicians. So, thank you for that. Now any closing thoughts or take home messages that you want the audience to know about?
Maha Hussain: No, other than I think the critical part I think is we’ve entered a new era and I call it a new world order in prostate cancer and that prostate cancer has lagged behind other solid tumors when it comes to precision medicine and targeted therapy. Obviously AR therapy is targeted, but we obviously do not pre-select. And this is the first time that we’re actually in a mode of targeted, based on pre-selection for potential responsiveness to treatment.
And as you see, not everybody responds even though when you pre-select them. So I do think the next steps is going to be looking at combination treatment and potentially moving therapy into earlier phases of the disease. Where the potential return on investment, so to speak, in terms of benefits might be higher. So I do think it’s very exciting times.
Alicia Morgans: Wonderful, and congratulations on a wonderful study. I know you were involved with the design from the beginning. Congratulations on finishing the first, like you said, the first study that was able to target these DNA repair defects or any personalization by DNA in a Phase III style in prostate cancer. It’s just a phenomenal effort and really exciting work, so thank you for taking the time to talk with us about it.
Maha Hussain: Thank you Dr. Morgans, and thank you for your audience.