Metastatic Prostate Cancer: A Practical Review of Therapy - Episode 3

Metastatic Prostate Cancer: Intrinsic Genomic Subtypes


Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: Chris, turning to you, what do you think has been learned about prostate cancer intrinsic genomic subtypes? Really, what are their prognostic significance? What’s their use in the clinic?

Chris Parker, MD, FRCR, MRCP: I think we all know that not all prostate cancers are the same. We all accept that there are distinct molecular subtypes, but I don’t think there’s yet any consensus as to how we should define those subtypes. We look at our colleagues in breast cancer, for example, and they’ve got it all worked out. They have different subtypes, they have different treatments for each subtype, and we’re not there yet. I think, at the moment, it’s not of any clinical utility. The various subtypes that are being proposed certainly appear to be prognostic, but they’re not in a position yet to guide management, and, of course, that’s what we want.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: What about predicting response to radiation or to ADT, do you think we’ve made any progress there?

Chris Parker, MD, FRCR, MRCP: It depends what you mean by predicting.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: Predicting a better response than average.

Chris Parker, MD, FRCR, MRCP: Yes. Well, I think we know that certain genomic subtypes are likely to respond better to radiation than others, and certain genomic subtypes are likely to respond better to ADT than others. But I would go back to saying that it doesn’t amount to a predictive biomarker, because even if you can predict that you’re less likely to respond to ADT, you’re not saying ADT is futile. You’re still going to use ADT regardless.

Bertrand Tombal, MD, PhD: As a surgeon, I see in my corporation that there is an interest in using these tests to say, “OK, that’s a bad prognostic signature for radiotherapy, so we should use surgery instead.” I say, “Yes, OK,” but then you look at surgery. They don’t do much better. This time we have, with all the questionnaires, a lot of prognostic markers that make sense: if you’re young—you’re 49—you’ve got PTEN loss, RB loss, or AR loss. They don’t need to be very smart to know they’re not going to do well. But when it comes to indicating whether or not we should omit ADT surgery, radiotherapy, and move to more sophisticated treatment, we don’t have that information yet.

Johann de Bono, PhD, MB, ChB: I think a very important point is that we probably don’t need more prognostic biomarkers. We’ve got plenty of those. The urgent need now is for more predictive biomarkers.

Bertrand Tombal, MD, PhD: Yes, and also, when we design clinical trials, not everybody is very enthusiastic with all the radiation oncologists, and you have plenty of combination trials. But, unfortunately, most of them are still shotgun experiments. Maybe it’s not yet the time to stratify the patients based on their signature, but we should at least obtain tissue and blood to create this because, otherwise, we’re going to be left with exactly the same observation.

Joe O’Sullivan, MD, FRCPI, FFRRCSI, FRCR: That’s a very good point. We have to collect these data and collect this tissue.

Bertrand Tombal, MD, PhD: And we need to get organized to get that tissue.

Johann de Bono, PhD, MB, ChB: I would like to postulate that in the very near future, in the next decade, we will be saying that certain patients should get surgery, not radiation, and vice versa, based on genomics. I really believe that will come.

Transcript Edited for Clarity