Incorporating Genomic Testing for Prostate Cancer into Your Practice - Daniel Lin

June 24, 2022

Daniel Lin joins Alicia Morgans in highlighting a guideline-directed practical course on how to use genomics in clinical practice. Dr. Lin was part of the instructional course faculty at the AUA annual meeting. The learning objectives of this course include a discussion on the research studies that led to the approval of genomic testing for prostate cancer, ordering appropriate genomic testing based on a patient's unique clinical situation, the state of the NCCN and AUA guidelines for genomic testing for prostate cancer, discerning the different prognostic clinical endpoints provided by various genomic tests, and how to recognize candidates for, and implications of, germline testing for prostate cancer.


Daniel Lin, MD, Professor of Urology, Cheif of University of Washington Urologic Oncology, Pritt Family Endowed Chair, University of Washington, Seattle, Washington.

Alicia Morgans, MD, MPH, Genitourinary Medical Oncologist, Medical Director of Survivorship Program at Dana-Farber Cancer Institute, Boston, Massachusetts

Read the Full Video Transcript

Alicia Morgans: Hi, I'm so excited to be at AUA 2022, where I have the opportunity to speak with Dr. Dan Lin, about the course he's presenting this year. Thank you so much for being here, Dr. Lin.

Daniel Lin: Thanks for having me.

Alicia Morgans: Wonderful. Tell us about the course.

Daniel Lin: I'll state first that I'm one of four faculty. There's Matt Cooperberg, David Penson, Joseph Wagner, and me. And we all put this course together as sort of a practical course about how to use genomics in clinical practice. Just practical aspects. I call it soup to nuts because we started, really, in the pre-diagnostic setting. You might think about germline factors. You might think about genomics to diagnose. And then post diagnosis, in other men that have a diagnosis already, how you're deciding whether to do surveillance or not in localized disease. But we're also touching on what you do for a living, which is more advanced prostate cancer. The impact of genomics, in that perhaps right after surgery, to decide adjuvant therapies or so forth, as well as, even in very advanced, as you know, the PARP inhibitors and so forth with germline and or somatic defects in the DDR pathways. We kind of did the whole scope. Obviously we're urologists, so we're more focused on localized disease, but that's kind of the outlay of the course itself.

Alicia Morgans: Great. So why don't we start going through the nuts to bolts? Why don't you start from the beginning and tell us how are you thinking about it?

Daniel Lin: I think that we're in a new age, no doubt, a new age with genomics in prostate cancer, amongst the whole breadth, the whole spectrum of disease. In the early disease, I think that we have a multitude of tests that help us diagnose, but then actually help us with what's the risk.

As you know, the BRCA one and two variants and their associated other variants germline what we inherit from our parents, certainly confer risk. We know that people that with those variants in the DNA damage repair pathway have younger age of diagnosis, worse cancers, worse outcomes. And so we're really trying to screen those people in a different way. There are multiple studies that are ongoing. We really discussed, in that course, quite a bit about that aspect about germline, what you have no control over versus somatic, which is defects after birth.

We discussed that quite a bit, as well as diagnostics. There are multiple tests out there that aid us in the diagnosis of prostate cancer, usually for men that have already an elevated PSA. Come in with elevated PSA. Maybe they get a negative biopsy. There are tests for the tissue like ConfirmMDX that look for methylation. There are urine tests that aid us to biopsy again. And that's kind of the beginning.

The middle is where I think the debate is, a bit, and I know you know this, with regards to after the diagnosis. Should we do more markers? Should we do risk stratification based on clinical variables like PSA grade, age, number biopsies? Or are there some DNA signatures, some genomic classifiers. There are multiple on the market now. I think therein lies a little bit of the battleground as well. Maybe that's where I'll start. I think that there are multiple, now, guidelines that are out there to say how to use them.

Alicia Morgans: There are. It's interesting though. The AUA localized guidelines have come out and they don't necessarily direct you to use this, that, or the other thing in each setting. They kind of talk about it as a concept and the fact that it's an evolving space. The data's just not really concrete in any one direction at this point. How do you deal with that?

Daniel Lin: Great question. So I'll just quote right from the guidelines as the disclosure, "I am part of the guidelines." James Eastham And Steve Boorjian really led the guideline for the AUA. I was a panel member. But I can even take a step back even more, like Scott Eggener and others. I'm not sure you were on that paper, but there was an ASCO guideline that was published in JCO two or three years ago, that directly said they don't recommend it for routine use, although they're select groups. I think that the word or the phrase "selectively used." I think that's what we're going to all adopt. The AUA guideline said, "Don't use routinely, but use selectively." Who were those select patients? They gave examples in the ASCO guideline of high volume Grade Group one, low volume Grade Group two; maybe some discordancy in clinical characteristics to pathology.

Those are the ideal patients to use the initial at the time of diagnosis, not post-treatment, which we're going to get into, I hope. But the ones after diagnosis. Because do we need a biomarker for a single core of Gleason 6 disease with a PSA of six? Probably not. There's multiple case series of men doing universally well, even randomized clinical trials that pivot, or the SBCG, showing that they do very well. It's the ones that we think that's going to change our decision, like high volume Gleason Grade Group one, like high volume 3+3. It's technically low risk, but maybe there's some badness in there on a genomic signature level. Those are the ones I think that, if you can take anything away from this, and I think the AUA guidelines would like that to be just selectively, not nobody get them, but not everybody get them. We just want to selectively use them.

Alicia Morgans: Really being judicious, because they have an expense, of course, added to them. But they can be so highly useful in those select populations.

Daniel Lin: Yeah.

Alicia Morgans: Thank you for driving that home.

Daniel Lin: Yeah.

Alicia Morgans: When you're thinking about a postoperative patient population, when might these classifiers be useful then?

Daniel Lin: I think that in the post-op setting... There are decisions that we make, such as should we give early adjuvant radiation or not? There have been some very nice papers done, primarily by the decipher platform, for their genomic classifier, looking at if someone has a high genomic score. Certainly, if a patient has a high genomic score, they do have a higher chance of metastasis, higher chance of death from prostate cancer, higher chance of recurrence. That's pretty much very well accepted, even taking into account disease factors, "Are they extra [inaudible 00:06:05] ? What grade?" You put that all and dump it into a model and it spits out... The genomic classifier does predict outcomes. The true test is, "Can it be associated with what we give the patient?" And so, should we give them radiation? Should we give them hormonal therapy? Should we give them chemotherapy? That remains to be seen.

We know now, in the post-op setting, that we're trying to decide, maybe even to pull the trigger early, despite some of the salvage radiation data. I think a genomic classifier can at least tell us... It's sort of the rule out. If they have a very low score, then I'm much more willing to say, "Probably no." A high score.... I'm unsure still. Honestly. It's the NPV. I always say, The negative predictive value is king. I tell all my trainees, "NPV is king." We take the Hippocratic Oath, "We do no harm." A "Do no harm" means, if a test is negative, you're pretty sure it's negative. That you're not going to harm somebody by not acting. I put a big amount of weight in the NPV and there are a lot of tests with good NPV, along the whole spectrum. I think the decipher looks pretty good, for the low ones are pretty much more indolent.

Alicia Morgans: I totally agree that the patients, when they're sort of on the border, and you get that low risk, it just makes you feel even more confident in monitoring closely and maybe early salvage if you need it. But, maybe, you actually won't even need that. 

Daniel Lin: That's right.

Alicia Morgans: That's really helpful. What would your message be, as one of the directors of this course, to folks who are trying to think about how to rationally integrate these into their practices, whether it's early, middle, or late on in their care of men with prostate cancer?

Daniel Lin: I think that a practical bit of advice is basically... It's all guideline directed. If one starts... Certainly at the beginnings. And there are guidelines around screening. I think family history is incredibly important. Look for not only prostate cancer in relatives, but also breast and ovarian cancer, certainly colon and pancreatic cancer. Again, it's not quite in the guidelines yet, all of them. But if a person has that, before they've ever been diagnosed, I think that there's some element of screening or referral to genetic counseling to look for some of these variants. In the next setting, I just tell people, :Look at the guidelines."If you see somebody that you think might have more aggressive disease... Again, it's written in the guidelines, which ones are approved and when to start thinking about it. That would be used selectively, like in the patients I said, "Higher volume Grade Group ones." Maybe those post operated that you're wondering about changing your management and only do it if you're going to really act on it.

I'll tell you Alicia. There's so many papers written on what are those criteria that predict risk. And genomic classifiers, genomic tests add to them. But how do we incorporate them all together? That's going to be, I think, the $100,000 question, which we're going to get into. I know you're doing this in some other trials that you do as we incorporate those markers, even before stratifying for treatment. I think that's where we're going to go. I hope that people take some practical advice out of our course.

Alicia Morgans: Wonderful. Thank you so much for doing the course, first of all. But then speaking with us about it and sharing some of the highlights with the UroToday audience. I really appreciate your time.

Daniel Lin: My pleasure. Thank you.