Alicia Morgans: Hi, I'm so excited to be at ASCO 2022 with Dr. Ashley Ross of Northwestern University, where we have the opportunity to speak about Decipher testing, both Decipher Biopsy and Decipher for prostatectomy specimens and how we integrate that into standard clinical practice. Thank you so much for being here.

Ashley Ross: Thank you for having me.

Alicia Morgans: Wonderful. So, Ash, I know you've been involved with Decipher for a long time, using it in clinical practice, and actually working on a lot of trials and investigations to better understand how we use the data from the Decipher test to really inform our practice. Can you talk about the Decipher test as it was first developed for prostatectomy specimens and how you think about that in your practice?

Ashley Ross: Thank you again for having me. Decipher is actually Affymetrix 1.0 ST array, and they actually look at the whole genome and expression patterns on that genome. The genomic classifier score, which is the most studied in what we're used to seeing in our clinical reports, is a prognostic score that goes from 0 to 1, and it was first developed on risk of metastatic progression following prostatectomy. Now there's been over 30,000 patients with long-term outcomes that it's been studied in, and across the board, it's shown to have independent prognostic ability for outcomes like prostate cancer-specific mortality and metastasis. But as you mentioned, the tool was first being used clinically in the post-prostatectomy setting. And there we're looking at our patients and we're wondering, particularly if they had adverse pathologic features, whether or not they need to have adjuvant radiation, whether or not it can be early salvage radiation, and whether we should add a systemic therapy to that radiation. Sometimes if you're thinking just in general of adding systemic therapy in the post-prostatectomy setting, it could also be a question that it depends on disease risk.

For a long time before the three clinical trials came out differentiating early salvage and adjuvant radiation, the adjuvant question was very difficult and we were using Decipher risk to decide who needs that radiation before the PSA gets to 0.1 or 0.2. Even now with the publication of three randomized studies that have shown that early salvage radiation is probably equivalent to adjuvant, those studies were not representative of patients in the higher spectrum of disease risk, seminal vesicle involvement, high Gleason grade, and so that's still a sweet spot for me to use Decipher.

Right now, when my patient has adverse pathologic features, I'll typically get a Decipher score and then I'll watch to see what the PSA does. I use ultrasensitive PSA, so I'm often looking at PSAs that are in the 0.01-0.02 range. If the Decipher score is high and there's one other feature that I find concerning, high Gleason score, seminal vesicle involvement, then I usually will go and refer to the radiation oncologists before the PSA reaches 0.1. So even if it's 0.03, 0.04, I'll refer them in that scenario. If the Decipher score is lower, so this is lower than 0.6, I'll often wait till the PSA rises to about 0.1, 0.2, as long as they don't have high Gleason score and seminal vesicle invasion.

One way of use is, when do I pull the trigger on radiation? The second use is, when we're going to do the radiation, do we add ADT or not? In the 9601 trial, which was the original trials looking at using bicalutamide in those cases for 2 years with radiation, it showed an overall survival benefit, but it was sort of limited to people who were being salvaged later. And when they did a sub-analysis of people with earlier PSAs, what they found was the benefit was strongest among people who had high Decipher scores. So the genomics give you a high-risk, adding the ADT, or in that case, antigen receptor blockade with bicalutamide, improved overall survival. But if you had low Decipher scores and you're doing salvage radiation plus the bicalutamide, you actually had a detriment to overall survival.

Now the SPPORT trial just came out, RTOG 534, which basically showed that radiation to the prostate bed, lymph nodes, and 6 months of androgen deprivation therapy, in this case, it was an LHRH-based therapy, seemed to have the best, at least progression-free survival, outcomes. And I'm still using Decipher to decide, am I going to add that ADT or not?

So, long-winded answer, but two scenarios in the post-prostatectomy setting. I'm using it to gauge, when am I doing my radiation? I think there's still a role for quote-unquote adjuvant, or let's call it early, early salvage before the PSA gets to 0.1, and the high Decipher teased me off to say, these are guys that I really want to put in that bucket. And then across the realm, whether I'm salvaging before the PSA gets to 0.1 or before the PSA gets to 0.5, do I add androgen deprivation therapy or not? High Decipher, I'm going to add it. Low Decipher, it's a really good look at the risk and benefits and understanding that I might be having a detriment to their overall survival, and maybe I'm omitting the androgen deprivation therapy there.

Alicia Morgans: Those are really practical points and I'm glad that we raised them and that you really talked that through. Deciphers actually used in another setting though, Decipher Biopsy is something that you can use even before a prostatectomy. How are you using that test?

Ashley Ross: Yeah. Traditionally, next thing that they moved into was a big decision making about who should we do active surveillance in or not? A lot of us were using genomic tests, and there was multiple tests in this category, Decipher, Prolaris, Oncotype DX Prostate, that was helping us decide who with higher volume, low-risk disease or with favorable intermediate risk disease, was an appropriate candidate for surveillance. For all of those, if the genomics were in the higher risk category, so the RNA expression of the cancer was showing metastatic potential or potential for adverse pathology, they probably should not be on surveillance. More recently, particularly with the genomic classifier and Decipher, there's been intense investigation into intermediate-risk prostate cancer. And there, if we're going to treat someone with unfavorable intermediate-risk disease, or even with high-risk disease, for surgery, our treatment doesn't really modulate much. We're going to do surgery, and if you're taking someone in the operating room because the disease has risk, in most cases, we're going to perform a lymph node dissection.

So, the question is, well, what if you don't do a mechanical therapy? I often tell my patients, surgery is like a weed in your garden. You're going to pluck the prostate cancer out like pulling a weed out, but radiation is lik, you're going to spray it with weed killer. We're really good at targeting the radiation, we can spray it in the right place, but will that weed die without weed killer? Is it going to be sensitive to the radiation? Does it need more to sensitize it? And androgen deprivation therapy may be sensitizing to that radiation, but it has more morbidities. Even with the newer androgen deprivation therapies that are out there like relugolix, it still can have lingering effects. It can affect bone health, sexual function, metabolic health, obviously.

The idea is, what if we have another tool, like genomics with Decipher, for example, and we can risk stratify many even more and decide who's going to do well with radiation alone, who's going to need the androgen deprivation therapy. There's been studies, both retrospective series from databases and also now evaluation of prospectively collected data in clinical trials, that have shown that if you have a low Decipher score and you have unfavorable intermediate-risk disease, if you get radiation by itself to that patient, they do quite well. Good metastasis-free survival, good overall survival from the RTGO 126 trial.

And so, that actually means two things. Right now in my practice, I'm a little bit ahead of the curve in the sense of, I'm using it to decide, do I add the ADT or not? Index patient would be the 70-year-old, maybe with a little bit of cardiovascular history, has 4+3=7 cancer, but maybe in a few cores, PSA is under 10, T1c exam. And the question is, do I add on antigen deprivation therapy for 6 months or just give them radiation alone? Then I'll get the Decipher. If Decipher is low risk, I think they do quite well with just the radiation. If Decipher is high risk, I think they really need the ADT to do well. There was almost a five-fold difference in metastasis-free survival from Decipher low-risk to high-risk in that context if you didn't give the ADT.

Now, some people say, "Well, I think that's okay in certain cases. I really want to see that in randomized prospective data." And the National Radiation Oncology Group has been very proactive in that way. They've just opened at multiple sites across the US, the GUIDANCE trial. And that's actually saying two things. One, stratify unfavorable intermediate-risk by Decipher. If you're Decipher low-risk, you can get ADT for 6 months or randomized to just radiation alone. And also importantly, there's some data from Paul [inaudible 00:08:51] and others that shows that for some of these guys who are unfavorable intermediate-risk with high Decipher, they have poor outcomes even with the 6 months of ADT. And so the other arm is if you're high Decipher, ADT for 6 months plus the radiation, which is now standard of care, or 6 months of ADT plus darolutamide. And that would be interesting to see if that improves outcomes like metastasis-free survival.

On the higher risk case, there the question is, do we really need 24 months of ADT? If the Decipher is low, can you get away with less, 12 months? And I have to say, in my patients, I really do see a difference between 12 months and 24, particularly some of the cognitive and psychological effects of the ADT, I see that as a point. And then for the high Decipher risk, high-risk guys, is that enough? Do they need intensification with something like abiraterone or apalutamide? And that's being studied perspectively in the PREDICT-RT trial, also opened at multiple centers. So it's really becoming a big part of my practice across spectrum, and I think that as those trials evolve, as more data develops, I think a lot of us who are using it routinely, I think it's going to become, quite frankly, a needed standard of care, but we'll see how those trials shake out.

Alicia Morgans: Well. That's wonderful and we'll have to make sure that we have links to those trials available for folks and the clinicaltrials.gov information, because the NRG is really doing all that they can to help us answer these questions of how we can apply the treatments in a way that's not one-size-fits-all. Because even though we try, as you said, to talk about favorable or unfavorable intermediate risk, and then of course high risk, we know that these patients don't have disease biology that puts them in such sort of large buckets. So, this is fantastic.

Ashley Ross: Absolutely. And there's always a question I ask myself sometimes, and sometimes I talk to colleagues about, which is, if you look at those trials, and thank you for putting the links in for the audience, they're really pragmatic trials. They're things that we're doing right now, but they want to make sure we're doing it the right way, and so they're studying it. And the question is, do we pull the trigger on these things in our practices? And I think that what the index patient I was talking about maybe demonstrates is, for a lot of us, we're not ready to routinely use it on all patients until the trials come out, but there's certainly patients, and it might be even a quarter of our patients or more, where we really are at a crossroads of what exactly do we want to do to them.

And like you said, having more personalized medicine allows us to make the decision. Free to comment as well, that in my practice, it's like, where are you practicing? Are you reactive to the data that's coming out and waiting until it's fully finalized, or do you think that there are places that you can be a little bit ahead of it and use the tools that we have because they're available before you know the full answer? I think that the latter is what I do, particularly with the patients where I'm struggling a little bit. I'm not sure how you approach it.

Alicia Morgans: Yeah, I think the same exact way. In my current practice, I actually see a fair amount of patients with localized disease and we're using these kinds of tools all the time to try to make those decisions, and of course, to enroll patients in those NRG trials, because those are so important as well. So, if you're in your practice and you don't feel comfortable making the choice based on the score alone, it's really, really helpful to know, "Hey, right down the street I can send patients who are going to get radiation to this trial." You may be able to even be a site to provide the ADT, if that's possible within the trial context. But really giving patients the opportunity to enroll in those trials is wonderful, so I appreciate you raising them. And I also appreciate you talking us through these really nuts-and-bolts, concrete examples of how we can use these tests to do the right thing for our patients. It's extremely helpful.

Ashley Ross: Thank you.