Dan George, MD: Joe, I’m curious. This is going to require a culture shift in how we think about prostate cancer, how we think about germline, and how we think about somatic. How is that going to translate into the real world? How do we convince our colleagues around that this niche, however big or small it is, deserves screening in the entire population? Or can we pick out clinical disease states or patient phenotypes that we should be definitely screening versus people who are going to be a little less suspicious?

Joe M. O’Sullivan, MD, FRCPI, FFRRSCI, FRCR: I guess we’ve seen data emerging gradually over the years, but I agree with Chuck Ryan. I think these data, which have shown a survival benefit for a mutation-targeted therapy, are really going to accelerate that involvement. We’re really going to have to get to know our molecular pathologists very well now.

There was a fantastic session at ESMO [the European Society for Medical Oncology Congress] with Mark Rubin, Johann de Bono, and Rosalind Eeles giving a tour—the full state of the art with regard to genomics and prostate cancer generally. Of course, predicting response to therapy is 1 part of it. Prognostication is the other. But it’s very clear now that, as Rosalind Eeles says, “we have a tsunami coming our way.”

We’re going to have to be ready for this, first to decide which patients will benefit from DNA-damaged targeted therapies but also to understand that in much more detail. Johann really gave a great talk yesterday talking about the complexity. As you said, Chuck, we need to look at whether it’s single allele loss or biallelic. It’s not just a question of how it is in breast cancer—ER [estrogen receptor] positive, ER negative. It’s way more complex than that. I think we’re just seeing the beginning, and even that’s much more complex.

Charles Ryan, MD: The other factor is, rarely do you see a patient with 1 mutation, right? If you’re sequencing a lot of your prostate cancer patients, you may get 3 with BRCA2 alterations, all pathogenic BRCA2 alterations. But 1 has RB1 loss, 1 has P53 loss, 1 has PTEN, P53, and RB loss. Those are 3 different patients molecularly. We’re just beginning to scratch the surface of what all that means.

Dan George, MD: But I can’t emphasize enough that you don’t know any of that unless you profile.

Joe M. O’Sullivan, MD, FRCPI, FFRRSCI, FRCR: Exactly.

Dan George, MD: I think this tsunami has been there. We’ve just been ignoring it. We haven’t had the tools. We haven’t had the tools to treat or the tools to diagnose.

Joe M O’Sullivan, MD, FRCPI, FFRRSCI, FRCR: They didn’t have the tools to treat that I think will really open…

Dan George, MD: This is the thing now. We’ve really got to change our mind-set. This becomes the norm, not the exception, in our patient.

Bertrand Tombal, MD, PhD: I don’t think, at least from the European perspective, that we’ve been ignoring it. But you have to understand that so far, it was confined to a very privileged world of academic institutions. Actually, that tsunami comes with a financial cost.

There is a caveat around that in that I don’t think we haven’t been screening for science reasons. We’ve haven’t been screening because it comes with a huge cost. If you look at most healthcare systems—except for the big 5 in the US—if you can pay for it, you’ve got it. If you don’t have money, you don’t pay for it. In many European countries, for instance, we are reaching the limit for the sustainability of the healthcare system. When you do a quick calculation introducing this drug, it means introducing in the community the testing.

What scares me about a presentation like Johann de Bono’s is that we are very far away from having an easy, cheap point-of-care test for DNA-repair mutation, including ATM. This means we still need to go to complex, expensive genetic technology. When people add the price of that to the price of the drug, we’re going to have to make a crucial choice. For instance, when you see the difference in response between BRCA2 and ATM, it’s very likely that in many countries the payer will ask for 2 models. They’re going to ask a value-based model for BRCA2 and a value-based model from ATM. So there is quite a bit of complexity.

The second thing is that when you look at it, you think, “OK, that’s great, because basically it’s the first in class to precision study medicine.” But then you take time, you sit down, and you say, “Wow, my God, there’s still a lot of frustration.” First, we’re going to have to cope with all the patients who don’t have these mutations. That’s 4 of 5, best-case scenario. The second is that even for those who have the mutation, the benefit remains modest.

Especially when we speak to the patient, I would say, “Yes, that’s a great trial.” Probably in 15 years this is going to be the reference as we enter the field of precision medicine in prostate cancer. Still, we should restrain from being overenthusiastic. I think that trial result is generating a lot of very important ethical and practical questions.

Dan George, MD: It’s much when CHAARTED was recorded. It was the first study. Then others came, and more will come. I would just say that I agree with you on the cost, Bertrand. But we’ve seen on the research laboratory side the cost of doing this kind of profiling had come way down. I think clinically we could see those costs come down as well. You know I’m not convinced that we’ve lost the value-care argument at this point.

Charles Ryan, MD: I’m not worried about the cost of the test, because if the test costs the same as about 1 month of therapy with the alternative drug that you’re not going to use, you’re saving money, right? If you’re not choosing second-line enzalutamide or abiraterone for a patient, but rather you’re putting them on olaparib because they have a BRCA2 alteration, you’re not only helping the patient but avoiding unnecessary, ineffective care. You’re probably also delaying or avoiding complications of the disease that may arise.

I do think there’s an important point about the patients, the spectrum of response in the patients with the mutation. It used to be that I thought when I picked up a BRCA2 mutation I was going to prescribe a PARP inhibitor with so much optimism. I still am very optimistic, and it’s really great to make that call. But what we’re seeing is the response proportion to these drugs is about 50, not 100%. So that’s an interesting story. Why is that the case? What are the mutations? How can we do better? What are the other targets?

Christopher Sweeney, MBBS: Actually, I want to pick up on 2 points. One is what we hear in the early reports, and this is the same with Lutetium 177 PSMA [prostate-specific membrane antigen]. We hear about the extremely good responder, and we think that’s going to happen to everyone. We’ll see what happens with the randomized phase III trials, and we see this.

The other thing I think when we’re talking to patients about, “I’m going to biopsy you to get a specimen. We’ll try to work out if you have, what we think are the 3 lead candidates,” and it’s probably going to be biallelic, BRCA2 right now. That’s going to be the main predominant one. Some ATMs maybe, and some real BRCA) ones. They’re the big ones. There may be some other smatterings, but let’s actually say you’re going to biopsy 100 patients, and probably 10% of them actually have something that may be relevant, of which 50% will actually have a response. Even with the point of biopsy, I think we have to start the patient’s expectation that we’re going for it, but it’s no guarantee.

Charles Ryan, MD: Hopefully all that biopsying will lead to some discoveries that will help the 80% who aren’t currently benefiting.

Christopher Sweeney, MBBS: For the next…

Charles Ryan, MD: The other point, as a minor consideration, is these MSI [microsatellite instability]—high cases that you pick up. That’s the difference between life and death. When you identify an MSI-high case, that makes many biopsies. That makes that biopsy worthwhile, and you don’t know it until you see it.

Christopher Sweeney, MBBS: But the incidence again is still…

Charles Ryan, MD: Three percent.

Dan George, MD: But it’s 0% if you don’t check.

Christopher Sweeney, MBBS: No, I’m saying check it, but I actually don’t say, “Look, you really need this. This is the difference between life and death when you’re starting the 100%.”

Dan George, MD: That’s right.

Charles Ryan, MD: Consider the cost and the morbidity of the biopsy versus putting everybody on pembrolizumab, for example.

Christopher Sweeney, MBBS: I’m 100% in there. This issue of desperation oncology, as you said, pembrolizumab on hospice because you got a shot.

Transcript Edited for Clarity