Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: Chris, turning to you and thinking about your personal approach, as well as the timing and the choice of therapy for a patient with metastatic castration-resistant prostate cancer—chemotherapy and targeted agents or bone-targeted therapy, what’s your approach to deciding on the timing? What is your choice of drug in an individual patient? What type of things do you consider?

Chris Parker, MD, FRCR, MRCP: It’s obviously a big question. To simplify the question, should we take men who are managed with ADT [androgen deprivation therapy] alone up front, so they’ve not seen abiraterone and they’ve not seen docetaxel?

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: Exactly.

Chris Parker, MD, FRCR, MRCP: Then we’re left with a choice between the AR-targeted agents, chemotherapy agents, and radium-223. I think it’s fair to say this is an evidence-free zone. So, we can do what we think is best in terms of clinical judgment. In that scenario, I typically use the AR-targeted agent first and the chemotherapy second, on the grounds that it’s less toxic and leaves the more toxic approach for later. I understand there are counterarguments.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: We’ll hear them in a moment.

Chris Parker, MD, FRCR, MRCP: And then, if we’re going to use radium-223—if the patient has bone disease and does not have visceral disease—then my preference is to use that with the AR-targeted agent, not to delay it until they do get visceral disease, because that means they’re never going to get radium-223. And that seems to be the worst mistake you could make. Although there isn’t evidence to support that, that would be my preference.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: That’s your personal preference. What about you, Johann, in that situation?

Johann de Bono, PhD, MB, ChB: I would agree with Chris. I would fully agree with what he said, actually.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: Bertrand?

Bertrand Tombal, MD, PhD: I also agree with what Johann already mentioned. What we can say in terms of sequence is not more what we can do, but rather what we probably shouldn’t do anymore. Something I see a lot—and I don’t think there is a role for this—is if a patient is 65 years old and progressing on enzalutamide with a quite rapid progression, they are actually denied chemotherapy, but receive another shot of abiraterone if they had enzalutamide for 3, 6 months. I think that at least for the first 2 or 3 lines, we should exhaust the drugs with proven benefit, meaning abiraterone or enzalutamide, docetaxel or cabazitaxel. And then, maybe if you have nothing, try to find a place for radium-223 on top of that. It’s like rechallenging with docetaxel when they received docetaxel: Yes, it works, but if the patient can receive cabazitaxel, shouldn’t we still do that? I’m not sure.

Chris Parker, MD, FRCR, MRCP: Can I just bring up AR-V7?

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: Please.

Chris Parker, MD, FRCR, MRCP: I was in a debate this morning, and there was audience voting. The majority of the audience would use, or would want to use, AR-V7 status to decide this question, whether to use chemotherapy or an AR-targeted therapy. That was in spite of the fact that the panelists in the front were unanimous in that they would not advocate for that. The panel was unanimous that there isn’t enough evidence to support using AR-V7 to make this decision.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: People want to believe that it works, don’t they?

Johann de Bono, PhD, MB, ChB: I want to support the panel fully. I don’t think AR-V7 testing should be routine. In fact, there’s very strong evidence that many cancers, up to 30% with AR-V7-positive disease, respond very well to abiraterone and enzalutamide. So, I think that actually is a poor interpretation of the data. In fact, what I can tell you is that we do a lot of AR-V7 testing in tumor biopsies with CRPC. Not many do this. And, if you look at a biopsy for AR-V7, you get islands that are AR wild-type—full-length AR, no V7—and islands with V7. So, when you test V7-positive, most of the cancer is still AR wild-type with some V7. And, actually, it really worries me. We have evidence that V7 positivity implies lots of CTCs [circulating tumor cells]. That’s what it tells you. If you are V7-positive in blood, what it means if you have a lot of circulating tumor cells. We published years ago that a lot of circulating tumor cells is a bad prognosis. The reason that being AR-V7-positive is highly prognostic is that you have a lot of CTCs. We have no evidence for it, and we’ve got a paper that is just being submitted. It worries me that those data have been hugely misinterpreted.

Chris Parker, MD, FRCR, MRCP: What worried me this morning at the debate is that there was voting before and after the debate, and, in spite of the fact that those arguments were put forward extremely clearly, the results of the vote were precisely the same. People were just as enthusiastic about AR-V7 testing after the debate as they were before.

Johann de Bono, PhD, MB, ChB: That is misplaced, I think.

Bertrand Tombal, MD, PhD: Yes. On top of that, this is a discussion that was generated in the late use of abiraterone/enzalutamide. Actually, we’ve got a good example, which is the galeterone trial that was run similar to the enzalutamide PREVAIL scenario, where you use this for early CRPC in asymptomatic patients. First of all, you’re going to test a lot of patients and use a lot of resources to pick up a few patients who have a very, very poor prognostic. So, the test that you believe is predictive is actually prognostic in a very small subgroup of patients, and otherwise useless. You would probably have the same benefit by giving 1 or 2 months of enzalutamide and monitoring patients and being, once again, more attentive to the PSA and the radiological response, rather than to do an expensive test that, yes, this company will do, but it’s one sequencing. So, I would fully support not doing it, and we never do it.

Johann de Bono, PhD, MB, ChB: And can I say—because I run this test in my lab, so I know this test intimately, with great detail—this test does not have a good dynamic range. Actually, what I mean is that very rapidly in these patients, you come to a level of AR-V7 that is very noisy. All assays have this, but with this mRNA blood assay, it is particularly a big problem. If you do the assay 3 times, the same blood test 3 times, you get different results every time. So, I have great concerns about the use of this assay. We now have data that with many patients who truly have AR-V7 in their tumor, in their tissue, in their biopsy, you don’t find positivity in the blood. And they have a high burden of disease, a lot of CTCs. So, actually, I don’t think this is a good assay, and I’m really concerned that the community has been misled. That’s my concern.

Transcript Edited for Clarity