Transcript:Raoul S. Concepcion, MD: Moving on to our next segment, one of the interesting articles that has been published this year is the Pritchard article that was in The New England Journal of Medicine, which basically looked at DNA repair gene mutations in over 600 patients with metastatic prostate cancer. And what he reported on—this compared to data that was in TCGA—was this tremendous, close to 12% incidence of germline mutations, and these were patients that were nonselected for history of prostate cancer. What they found was obviously there seems to be this higher incidence of primarily BRCA2 mutations that exist or that occur in patients who have metastatic prostate cancer. So, I think what this opens up now is, as we move forward with genetic counseling and we are looking more toward screening controversies, what role does BRCA1/BRCA2 testing play as we see either newly diagnosed patients with prostate cancer, people that come to us who have a family history of prostate cancer? Maybe they have a family history of breast, ovarian, colon, or pancreatic. What’s going to be the role of this, Ash?

Ashley E. Ross, MD, PhD: Well, I think a couple of things, and we’ll get to it down the road, too. One is, it’s important for the audience to know that the cell cycle regulation and DNA repair is really a cascade. BRCA1 and BRCA2 are involved in homologous recombination. There are other genes that they were looking at, too, because there was a really deep sequencing approach with ATM, NVS-1, MRE11, etc. And they had this surprising finding, like you said, 12% of metastatic men had a germline, so every cell in their body had some mutation there. What I found extremely surprising myself is just like you pointed out, family history of prostate cancer didn’t seem to play a big role, but history of any cancer seemed to play a role. I would have expected the men to be younger. The younger men would have the mutations, but that wasn’t the case either, giving up the idea that there are probably multiple hits or genetic interactions that are necessary for this.

And so, there are 2 places where knowledge of the BRCA1 or BRCA2 or DNA break repair status are going to be important. One, which I’m sure we’ll touch on in a little bit, is in the person with castrate-resistant prostate cancer who’s progressing through therapies. Is there a role for PARP inhibitors or platinum-based therapies in that gentleman? But the second is the concept, and Peter Nelson has talked about it, of using these men as sentinels. So, when a man comes in with metastatic prostate cancer, regardless of whether they’re hormone-naïve or not, we’ve institutionally been moving toward testing them for their germline status to allow alerting their family members of having a germline mutation that might put them at higher risk for malignancies in general, particularly breast cancer for the female family members. What do you do with the man—and we are still debating this—who comes in and who has low-risk prostate cancer and they’re 70 years old with a small volume, Gleason 6 tumor, but you’ve identified them as having a BRCA1 mutation or BRCA2 mutation? We don’t really understand that at all. I think that the one thing that we’ll do as this is becoming more commonplace and easier, and less expensive, to test the germline status is we’ll have to collect that data in the registry and see if it does play any part in what’s going on.

A final interesting note is about 6% of the high-risk men had germline mutations. Somatic mutations might be even more common. PARP inhibitors are not exceptionally well tolerated, but fairly well tolerated. Can we start to think about moving them up earlier into space and doing combination trials? These, I stress, should be small phase II trials looking at efficacy, maybe combining them with radiation, which is also damaging DNA.

Raoul S. Concepcion, MD: So, Jorge, for the urologists who really have got 0 experience with PARP inhibitors, give us an overview and your experience, just in general.

Jorge A. Garcia, MD: The first thought I have is tumor genomics and prostate cancer. I think that the challenge with tumor genomics, and when we’re trying to go from personalized medicine to precision medicine, is the fact that you may get a list of genes that you have alterations on, but you may not be able to act on those changes we find with whatever platform you use, like foundation medicine. So, I personally think that for me, if you have a hypomethylation or a DNA-deleterious mutation, whether it’s ATM, BRCA1, or BRCA2, or the Fanconi anemia gene, at this point, it doesn’t really change my therapeutics. And the reason why is because in the PARP trial—which is the trial that our French colleagues actually published them first before Pritchard’s data—we know that there is a response rate when you use a PARP inhibitor. The only FDA-approved PARP inhibitor in the United States right now is olaparib, which is approved in ovarian cancer. Now, they have a breakthrough registration in front of the FDA for prostate, but they are not yet there.

AstraZeneca has a trial right now in prostate and is selecting patients, which is a challenge because of what you mentioned. Certainly, in the Prichard data, actually less than 6% have BRCA1 and BRCA2. If you look at all the mutations, it actually was around 11%. Screening 100 men just to find 10 patients is very costly, and you can swing that bill to some extent. We are not adopting germline screening for everybody. And to some extent, as you mentioned, Ash, I was surprised because I would have predicted if you were 40 years old and you have really aggressive disease, maybe that’s the guy that I want to look into. But at this stage, if I see a patient, let’s say, that comes with a report and says, “I’m ATM-mutant” or “I have a BRCA deletion,” I simply tell them, “Your cancer is going to be far more aggressive, and more than likely I’m going to be able to expose you to a PARP inhibitor.” At this point, the biggest question for me clinically is, if I haven’t treated you yet with abiraterone, with enzalutamide, with radium, or with docetaxel, would I change those treatments and put you on a PARP inhibitor? I’m not sure the data we have would support that approach. Having said that, if you have seen every single treatment and you happen to have one of these hypomethylations, somatic change, or a germline change, then there is no option, and maybe we can actually resort and try to use a PARP inhibitor.

There are over 6 or 7 companies right now that are looking at different PARP inhibitors. So, as you can imagine, the next wave of agents that we’re going to be testing in prostate cancer are not going to be the PD-1 alone, they’re going to be the PARP inhibitors. The biggest challenge that we are facing as a group is are we going to be selecting out? We believe that up to 90% of those patients who really have these mutations are the ones who truly benefit from therapy. Then, the question is going to be if it is useful in a small niche of patients, what happens to the other 90% of patients? I think that to the extent of where we are, unless we have randomized phase III data, this is probably the only space where we may be able to swing a placebo trial again in the postheavily pretreated patients to get a PARP or to get nothing. Those aren’t the trials that we’re conducting right now.

Raoul S. Concepcion, MD: Alicia, what’s your experience with PARP inhibitors?

Alicia K. Morgans, MD, MPH: I’ve actually used them in ovarian, but I have not used them in prostate. And I think one of the important things that I would stress with all of this is that the de Bono/Mateo paper actually looked at mutations that were somatic mutations—so, mutations that were in the tumor after exposure to prior treatment. I think we may find that we’re able to incorporate PARP inhibitors in that space much more easily, as Jorge is saying, because the mutation rate is much higher or at least it appeared to be so in the Mateo paper. So, we’re talking closer to 30% rather than the 12% that we see in germline, and, conceivably, we should see a similar benefit from the PARP inhibitors in that space. And in the Mateo study, I think we saw response rates in those patients who had mutations, around 88% or so. It was a high response rate.

I think germline mutations are important, but it’s much more important we are sequencing more intensively, actually, biopsy samples, because the somatic mutations may be more useful. And as we move through all of this, I think it’s important to think about how we’re going to support these patients and their families in terms of genetic counseling. Because if this is the direction in which we’re moving, we are going to find things that are going to affect more than just the patient, and we need to have a strategy to support them. But, in general, PARP inhibitors are, I would say, not as easily tolerated as things like abiraterone and enzalutamide. And so, I am inclined to stick with those tried and true therapies first. But should I have a patient who has a mutation and has already received all of the approved agents, I think a PARP inhibitor is great. The other thing I would think about is something like carboplatin because it does look like these patients may be more sensitive to platinum agents and carboplatin is already approved, so we can get that. We can get it pretty easily off the shelf.

Raoul S. Concepcion, MD: So, what does the side effect profile look like, just in general, with some of these PARP inhibitors?

Alicia K. Morgans, MD, MPH: I’ve seen fatigue as probably the most challenging issue. You can see cytopenias as well and some GI distress. I’m not sure if Jorge has other issues that he’s come across, but I would say fatigue and the cytopenias are probably the biggest ones from my end.

Jorge A. Garcia, MD: I think it’s cytopenias. Obviously, it becomes an issue when you have a patient who has been heavily pretreated with chemotherapy. If you have seen 6 cycles of docetaxel or if you have seen more than 6 cycles and then Jevtana, you come back with a PARP inhibitor, I’m going to double that radium, if you will. If you use it earlier, those patients may have more myelosuppression than the average patient. I actually have used PARP inhibitors quite a bit because I was able to pin down a couple of patients who had aggressive disease, in 3 to 5 patients that I have, one of which that I actually knew when I met him had a BRCA deletion. He actually progressed on abiraterone. He got CHAARTED type of therapy, hormones, chemotherapy, and progressed biochemically. We repeated scans, and there was no evidence of disease because his initial volume was soft tissue and that was melted down by therapy. So, he progressed. I put him on abiraterone because I knew he had metastatic disease, he progressed within probably 9 months or so. He didn’t want more chemotherapy. And then, the PARP data came out and I said, “You know what, maybe you’re BRCA. Let me just see if I can actually get you a PARP inhibitor.” I was able to get him a PARP inhibitor. He had a PSA response, a 90% PSA decline for around 5 to 6 months, and then progressed. And then, the question was what else to do at that time.

I don’t think these patients are actually in complete responses. More important than that, I think that when you look at the PARP data, the question would be, would those patients who have visceral disease—the ones that we may want to actually use a PARP inhibitor maybe in combination with a platinum-based therapy to increase their response—make it more durable? Because we know for those patients, there may be a neuroendocrine behavior, but it may also be driven by BRCA.

Transcript Edited for Clarity