Charles Ryan: Hello. I'm with Colin Pritchard from the University of Washington, where he is the Director of the Genetics and Solid Tumor Laboratory, and emerging as one of the really top voices in the area of genetics in prostate cancer.

I want to take a moment, and I want to talk about some of the more recent results that we saw at ESMO 2019. It was a great meeting in terms of our understanding of PARP inhibitors. There was the PROfound study that was presented, then the niraparib data, GALAHAD was presented, and there was data presented with rucaparib as well. And there were a couple of themes emerging from this data.

One thing I wanted to ask you about, which is monoallelic versus biallelic BRCA2 alterations and some of the studies require biallelic, and... Does that really matter? Is the question.

Colin Pritchard: I think it does. It's something where the devil's in the details. So, I think sometimes it's safe to say that there is biallelic inactivation of BRCA2, which I think really is the critical biological feature when only monoallelic is detected. Right, so we're thinking about what's detected versus what is truth. Those are two different things, right?

So when, for example, a germline mutation is detected and a somatic second allele inactivation is not detected, in the context of BRCA2 and prostate cancer, I think most of the time it's safe to assume, if you have an advanced prostate cancer I'm talking about, that there is biallelic inactivation even if you don't detect it. But that's just one example. Every example's different.

So for BRCA1, it may be a different story and we don't have enough data to know the answer to this. This is just complete speculation, but what we do know about BRCA1 is that when you have an inherited pathogenic variant in BRCA1 that is less penetrant for prostate cancer. That means that you're less likely to get prostate cancer than if you have an inherited pathogenic variant BRCA2. So that's fairly well understood, in terms of the lifetime risks that we really know now.

Charles Ryan: Right, right.

Colin Pritchard: But we don't know the exact numbers. So what that means is when you detect a germline mutation BRCA1, well then, maybe sometimes it'll be true-true unrelated, maybe sometimes your prostate cancer might be unrelated to that. And so, maybe the context, maybe it is more important to prove that there is biallelic inactivation because the tumor may not be related to that mutation and therefore no reason to think it'd be PARP sensitivity.

So every gene is going to be... The details are going to matter for every scenario like that. I've given you a couple of scenarios. ATM is another one I know that people are really focusing on them, and the data's been mixed there to say the least, in terms of PARP responsiveness. And ATM is is an even less penetrant gene probably by penetrant, I mean in the germline context, how frequently when you have the mutation, does it cause a cancer? In this case, prostate cancer. So I'd argue for ATM it probably is important and there's data emerging there from a number of groups on that-

Charles Ryan: If you'd argue for ATM, what is important? Biallelic?

Colin Pritchard: That is important to prove biallelic. I'm sorry, I should've said that specifically. Yeah, but again, we speculate about this, but we still have relatively limited data. It's not like there's one paper we can hang our hat on that says this proves we have to do it. Now there was a study that came out of Memorial Sloan Kettering that showed over a number of cancer types when you have an inherited germline mutation in BRCA1 or BRCA2 how often the tumor had biallelic or not.

So we have that data, so that gives us sort of a landscape what fraction of time tumors are driven by BRCA1/2, if you have an inherited mutation. Prostate's up there as surprisingly prostate, breast, ovarian, and pancreatic are the main ones, and then it kind of drops off from there. So that gives us some sense for, again, that context of when you have a germline mutation, what's the probability when you have a man with a tumor with that mutation that germline mutation caused the cancer.

It gives us some information there, but it still doesn't give us the data we really need about, you know, which context do we need to really prove biallelic, and which context we don't. But it's something that I, and a lot of other people are really interested in studying more.

Charles Ryan: Right. Well, as you remind us, even though we are now getting maturing clinical data on the PARP inhibitors and whatnot, it still is early days, as they say, for this story in prostate cancer. And thank you for all the work that you've done in leading the way and enlightening not only the community, but myself. I always learn when chatting with you about these. So thank you very much for joining us.

Colin Pritchard: Well, it was my pleasure.

Charles Ryan: Okay.

Colin Pritchard: That's right.