Deciphering Prostate Cancer: How Genomic Biomarkers are Shaping Treatment Decisions - Paul Nguyen

December 5, 2023

Neil Desai and Paul Nguyen discuss the integration of the Decipher genomic biomarker in prostate cancer treatment. The conversation centers on two trials, NRG-GU009 and GU010, which utilize Decipher genomic classifiers to tailor treatment intensity, particularly in hormonal therapy. Dr. Nguyen emphasizes the shift from traditional metrics like PSA and Gleason scores to more precise genomic classifiers that predict distant metastasis with high accuracy. Both oncologists discuss the future of personalized treatment, including the potential for Decipher to influence not just hormonal therapy but also radiation techniques. The trials aim to optimize treatment duration and intensity, thereby improving patient outcomes while minimizing side effects.


Paul Nguyen, MD, MBA, Genitourinary Clinical Center Director, Radiation Oncology, Dana Farber Cancer Institute, Boston, MA

Neil Desai, MD, Director of Clinical Research, Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, TX

Read the Full Video Transcript

Neil Desai: Hi. So we're here with Dr. Paul Nguyen, professor of radiation oncology at Brigham and Women's Hospital in Boston, Harvard Medical School. And my name is Neil Desai. I'm a radiation oncologist at UT Southwestern in Dallas, Texas, here at the UroToday group to talk a little bit about the integration of a new or novel genomic biomarker called Decipher into the treatment selection of prostate cancer radiotherapy and adjunct hormonal therapies. And so thank you first, Dr. Nguyen for joining us today here at ASTRO 2023.

Paul Nguyen: Thanks, Dr. Desai. It's great to be here.

Neil Desai: Well, it's a real pleasure because I think we're seeing the culmination of a lot of work from you as well as others in these two trials, NRG-GU009, GU010, in which we're using the Decipher genomic classifiers to select treatment intensity regarding hormonal therapy, which has been the focus of a lot of efforts for many years from you and many others. From our perspective as a junior investigator in my own shoes, to know how you got to this point using genomic classifiers and the day-to-day treatment selection when it wasn't that long ago that we were preoccupied about Gleason scores and percentage core positivities.

Paul Nguyen: No, thank you. And I do want to emphasize there were so many people that did this work, and I was just one of the people that did it. But as you know, back in the day, we were focused on PSA, T, Gleason score, and then all of a sudden the Decipher classifier came out for patients after radical prostatectomy and showed that it could predict who's going to develop distant metastasis with a very high degree of accuracy beyond what you can get from the clinical factors alone.

And I think the challenge for us as radiation oncologists is, "Hey, that's after radical prostatectomy. Can we actually bring that earlier to the biopsy space so that we can use it for our patients and help them with radiation decision making?" So as you know, we were able to do some of that work with the Brigham samples and that was the first radiation paper that came out looking at Biopsy Decipher and showing that it actually was prognostic for distant metastasis and even prostate cancer death.

And then others contributed, made a series with patients who got treated with prostatectomy, but using the biopsy samples as well. So now, we have this Decipher Biopsy Test that we can use on patients after diagnosis and predict basically who's going to develop a high rate of metastasis versus a lower rate of metastasis. And so thinking about that, we really wanted to incorporate that into future trials because it's great that it's prognostic, but the question is it going to be predictive? And that's what led to the work I think that you're alluding to.

Neil Desai: Yeah. No, that's fantastic. And I think it's very important also what you brought up is what are we actually predicting for? What do we try to influence? And it's distant metastasis, that's the impact on salvage hormonal therapy, rest of everyone's life or death and other events. And from my perspective, as a resident, everything I learned was more about biochemical recurrence prediction and that was a shift. To tell me maybe in my shoes five, eight years ago, why should I care about this metastasis more based on this classifier, specifically predictive benefit or prognostic influence rather than biochemical recurrence.

Paul Nguyen: Yeah. As you know, a lot of men will develop biochemical recurrence but not die of their prostate cancer, whereas those who develop distant metastasis, they have a much higher risk of actually dying of their prostate cancer. And because of that, the FDA actually considers metastasis free survival as an FDA approvable endpoint for drugs. And so I agree with you, it's so powerful to have a classifier that actually predicts for distant metastasis.

Neil Desai: Excellent. So I think obviously, like you alluded to, lots of work went into not only the suggestion of this Biomarker, Decipher as an additive aid in decision making the clinic, but should be something beyond that, something that influenced actual treatment decision making and care in a trial platform. And we all know nothing happens fast for NRG. So tell me a bit about the story that led to how you guys thought about the cut points you're using or how to use Decipher in this high-risk population, NRG-GU009.

Paul Nguyen: Yeah, I have to give a lot of credit to Dr. Felix Feng, who was the leader of the NRG GU group and I worked with him looking at previous randomized trials, 92-02, 94-13 and 99-02. And these are trials that were done 25 years ago where patients actually from 25, 30 years ago gave their tissue to be saved for things that we didn't even know what we could test them on.

We tested these using the Decipher Biopsy Tests and were actually able to find that on archival tissue you can predict for distant metastasis using Decipher Biopsy scores. And that's what led to the incorporation of that into the current randomized trials that we're looking at now, where we're really trying to figure out is the Decipher score going to help us personalize therapy for patients getting radiation for prostate cancer?

Neil Desai: And I think that's exactly the personalization aspect, I think that reflects the deep unease many of us have. Knowing that we're probably, would you say over or undertreating a lot of men using binary categories of say CN or other categorizations clinically?

Paul Nguyen: Yeah, I think so. I think right now, let's say high risk prostate cancer, which is the GU009 PREDICT-RT trial, we currently have a one size fits all for the hormone therapy. Most international trials use two years of hormone therapy for those patients. And that's exactly what the trial is trying to get around. We're using the Decipher test and stratifying patients, and for patients who have higher Decipher scores, we're going to see whether we can benefit from intensifying their therapy. So half of them will get two years of regular hormones and half of them will get two years of regular hormones plus two years of apalutamide. And of course, both arms get standard of care radiation.

Whereas for patients with lower Decipher scores, we think two years might be too much treatment for them. And in those patients, we're randomizing to the De-intensification study where half of them get the standard of care, two years of hormones plus radiation, but the other half get 12 months of hormones plus radiation. And we're hoping that for those patients, 12 months will be just as good. So the trial is accruing. Well, as you know, thanks to so many NRG investigators really putting their time and effort into that study.

Neil Desai: Yeah. And these are actually fantastic questions. We can't wait for the answers to this. Obviously, none of us want to give hormonal therapy longer than we want. None of us want to give next generation hormonal agents unless we have the benefits to justify the cost. This is really, really helpful in our own clinical practice. So when do you think we're going to get data? Is it accruing well?

Paul Nguyen: Yeah, I think we're getting close to, we're around 50% accrual now. It's 2,500 patients. The accrual is just outstanding across the country from investigators really working hard to get patients on. So we're really excited about that. And actually, I must say I'm very excited about your trial, the GU010 trial. Now, looking at the intermediate risk space. I'd love to hear some more about that.

Neil Desai: Yeah, so thank you Paul. And obviously thanks to you and many others who mentored us in the GU committee. I was a beneficiary being able to be in position to ask this question along with my co-PI, Dr. Ale Berlin at Princess Margaret, in which we are randomizing men just like in the GU009 high-risk cohort, but instead now an unfavorable intermediate-risk cohort of men with localized prostate cancer such as those with higher genomic risk and randomized to standard of care, six months hormonal therapy and radiation, or the addition of darolutamide and more intensified hormonal therapy similar to what you described in high risk.

Conversely, in those with lower genomic risk, we randomized the omission of hormonal therapy, which is also a pertinent question for men from a quality of life standpoint. So really following your lead, just taking it further down the spectrum to men of different risk categories, this is clearly a question of relevance to all men, which is data personalization. So got to give a tip of my hat to you, Felix, Dan Spratt, and many others who worked on all this data. It's an exciting time. Like your trial, we're enrolling pretty well as well.

Paul Nguyen: Yeah, I saw the accrual numbers. It looks fantastic. So Neil, if you project ahead 10 years from now when these trials read out, how do you think we'll be treating these intermediate and high-risk patients? What are we going to be doing in the clinic every day?

Neil Desai: And so I think to your original point, to some extent it matters and doesn't matter in that metastasis is king, whether we influence a person's chance of dying from prostate cancer will always be important. And so it is future-proof to an extent because you're talking about baseline risk at a microscopic level before you can see an imaging that will always be helpful to patients even 10 years from now, just like the trials from 20 years ago informed your trial and our trial today.

At the same time, we know radiation techniques are changing, integration of PET PSMA and other advanced imaging is getting more integrated into treatment decision making and treatment planning. For instance, one area of hot topic I want to highlight to your point is the use of integrated or micro-boost of higher dose of the dominant MRI lesion or defined lesion, which appears per one randomized trial of flame to reduce biochemical failure without increasing toxicity.

And the question is whether that influences our outcomes. There is some suggestion of impact on the metastasis rates, how does Decipher play into that? There's at least some data based upon work that you were involved in, I believe, as well as Dr. Spratt and Dr. Michalski looking at the RTOG-126 data that Decipher appears to associate with which men actually need higher dose radiation.

So can we look post hoc at that cohort of patients being increasingly enrolled on GU010 specifically that we allow to be on this trial to also ask that question in the future, not only is this metastasis prediction important and hormone therapy important, but will that be robust or influenced at all by the type of radiation and specifically micro-boost. So we are trying to quote unquote "future-proof" it in a way that is relevant to younger investigators like myself, eager to use these newer techniques.

Paul Nguyen: Yeah. That's really cool. So it sounds like in the future, we'll not just be getting PSA and Gleason score, but also Decipher and not just using it to think about the type and the duration, and the intensity of hormone therapy, but maybe also what kind of radiation techniques we should be using as well.

Neil Desai: We hope so. It's a really exciting time and again, just thankful to be part of the process and all the work that's gone before us in this.

Paul Nguyen: Yeah. Well thanks for talking about this today. This is really fun.

Neil Desai: Thank you, Paul. The pleasure's all mine.