Michael Leapman: Thank you so much. I'm really privileged to be here and really excited to talk about our study.
Zachary Klaassen: Yeah, some really cool stuff we're going to get into it. Before we get into the data you presented, just talk about how Decipher genomic classifiers really helped us risk-stratify these patients with active surveillance.
Michael Leapman: Absolutely. So what Decipher is, it's a genomic profile, and it gives us information about tumor biology. And so this is an assay that is used, a test used from biopsy specimens or radical prostatectomy, and it gives us a risk of events, progression, metastasis, recurrence after treatment. And so I use it in my practice, I think many people use in their practice to give information and communicate to our patients about the risk of that cancer progressing.
Zachary Klaassen: Right.
Michael Leapman: So one of the biggest use cases, I think, has been in patients with low and intermediate-risk prostate cancer who are considering active surveillance rather than initial treatment. And so there's been a big question about how do we actually use these tests, right? How do we translate this information into a coherent decision, and what are the right thresholds to act? How should we monitor people differently?
Zachary Klaassen: No, I think it's great. I mean, it's been huge in my practice. A three plus three, you don't want to do anything with. You want to put them active surveillance. Or that three, four that you maybe think you should, and you find at a low-risk Decipher, maybe you watch them a little closer. So I think the data that they've generated over the years is really robust. We're going to talk a little bit about the population-level data that you looked at. So maybe just lay out for our listeners what exactly SEER linked to Decipher is and how that came about.
Michael Leapman: Okay. So SEER is a national cancer registry. What it does is it pulls from cancer registries all throughout the United States. And this is information that hospitals and registries pass along centrally.
Zachary Klaassen: Right.
Michael Leapman: So this has very rich information about cancer grade, stage, PSA level, and initial treatment. And this has been the backbone for a lot of population health research in cancer for many, many decades.
Zachary Klaassen: Right.
Michael Leapman: So Decipher, the company that makes Decipher, did a very cool thing about a year or two ago, which is they created a linkage between their results and the SEER registry. So they inputted the two. They linked those two data together securely and published on it. Okay. And over the summer, it became released publicly. And so we could look at this data set and begin asking questions about how Decipher genomic classifier results relate to patterns of initial management.
Zachary Klaassen: Which is really cool because we've all seen SEER. It's been around since 1973. Now you have this powerful genomic data linking to this population-level data. So maybe just set up your study design and some of the key results you had that you presented at GU ASCO.
Michael Leapman: Absolutely. So our question really was to focus on people with low and intermediate-risk prostate cancer and understand how their genomic classifier results related to their initial treatment. Basically asking the simple question, "Okay, are people with higher or lower Decipher scores, are they more or less likely to be treated, rather than get active surveillance?" Very simple question. We're trying to ask a question of how these tests are actually being used in practice. We know that they're prognostic. We know that they give us the right signal, but it's a totally separate question about how do we actually act on this information.
Zachary Klaassen: Right. And what did you guys find? Were we consistent with what we should be doing?
Michael Leapman: Okay. Well, no suspense here, but basically what we found is that they are very very much aligned.
Zachary Klaassen: Yeah.
Michael Leapman: Higher Decipher scores were independently associated with greater odds of treatment rather than surveillance. And that held up for the low-grade, grade group one, and grade group two, low-risk and intermediate clinical risk groups. And it was actually quite strongly associated with getting treated versus active surveillance.
Zachary Klaassen: Which is good. I mean, if you think about it, we use this on a day-to-day basis. Every patient's a sample size of one. We see it now at a population level for I think the first time, where we're appropriately putting people in active surveillance as should be, and probably treating patients that have a higher genomic classifier score. So how do we convey this to patient? How does this translate to the clinic?
Michael Leapman: So I think there's lots of potential ways to communicate this, right. And I think this is something I've always struggled with. We have data from ProtecT saying that even in the unselected group of people with low and intermediate-risk prostate cancer, their outcomes are going to be really good with active surveillance or conservative management. So I think this kind of is one of the missing links in that story, right. It's saying, we know we have a biologic signal. We know the overall outlook is really good. And I think this is a way to sort of calibrate how we're doing in practice with what those signals are. So I think that's been a kind of a reassuring finding that we are directionally aligned. We're putting more patients on surveillance that have more favorable profiles and vice versa. And so simple message is I think we're reacting to this data. We are putting people in the path that seems to be suggested by their results.
Zachary Klaassen: Yeah. No, very encouraging for sure. Great conversation. Anything we haven't hit on, you want to leave our listeners with? Any take-home messages?
Michael Leapman: No, I think this is, I think, a lot more work to be done on this question. I think we have other... The secondary and tertiary questions are, well, what happens afterward, right? Are we making the right decision? How well do we monitor these people? We have shown that there are gaps in the quality of active surveillance in the United States. Many patients get initial testing, but that tends to fall off over time. So we're very interested in understanding how we can make that better, pick off the bad actors, and leave people alone who don't need to be treated.
Zachary Klaassen: Awesome. Congrats on the great work, Michael. Thanks for joining us.
Michael Leapman: Thanks for having me.