TRANSCRIPT VIDEO ID 3382 I 042823 - AUA23 AUA23_COOPERBERG_LIN_1
Daniel Lin: Thanks for having me.
Matthew Cooperberg: A lot to unpack in the update. I think there's a lot to like in the update, but before we even get into the details, maybe just talk a little bit about what the process has been, what the AUA does with these guideline updates. Maybe contrast that a little bit, with the NCCN or some of the other content panels.
Daniel Lin: Sure and I'm a Vice Chair of an NCCN panel as well. It is quite a bit different and that's the reason why we take years to do an update, usually and I'll tell you, I'll first just call out again John Wei, who was the chair and then it was Team Science as you know. I'm a big believer in Team Science. A huge group of people. I'm going to specifically call out Sigrid Carlsson, Dan Barocas and Peter Pinto, who led work groups, but then a huge group of people.
I'd encourage everyone to look at the authors, because they're all people that put so much effort into it. Then Lesley Souter was our epidemiologist-type biostatistician, and then we had Sennett Kim and Erin Kirkby, of course, just the back end, making sure everything was going on to the AUA, so Program Managers, but I can tell you it's a long process and you've been involved in these before too.
There's a systematic review that occurs, that is of course hundreds of papers and we have a very high level, so we pull out all the data and then it's winnowed down of course, like any systematic review would and then there's of course PICO, questions, key questions in writing that takes the majority of the time. What people don't realize on the backend, is that we then send that out to over 150 reviewers. You were one of them. I saw your name on the list and then from there over 80 of them returned comments and we had over 750 individual comments and John and I had to go through every comment.
Now of course, some of them overlapped, some of them didn't. We had to address every single one and then turn out the product. There were some comments that were incredibly relevant, that we had to actually literally change guideline statements and go back to the panel. The process was iterative, going back to the panel, back to the reviewer, back to panel. Very different from other guideline processes. We do give levels of evidence, of course and strength of evidence. Very similar to the European guidelines, but a little bit different in the way that we look at our evidence. It is a long process.
Matthew Cooperberg: And just compare that, for those who are not familiar with it, to the NCCN.
Daniel Lin: Yeah, the NCCN is a bit different. It is more of a consensus. There is not always a literature review, that's with each update. Members of the panel bring papers. The chairs obviously, are tasked with looking at the relevant papers to bring to the committee and the panel. The panel meets usually once or twice a year. Not all panel members can make it to the panel meeting, so it's very different. There are now evidence blocks within the NCCN. I'm going to speak for the NCCN and wear my NCCN hat a little bit, there is levels of evidence that they provide, but it's much different from a systematic review and it's much different from sending it out to external review.
Matthew Cooperberg: Walk us through some of the major highlights, in terms of this update for screening.
Daniel Lin: This update as you said, it was a lot. If you look at the last couple of prostate early detection, there were less than 10 statements. We have 35 statements. The previous guidelines generally just addressed when to start screening, perhaps how to do screening a little bit. This one had some major advances. First, we discussed at-risk groups, which we might want to talk about, with regards to Black, ancestry and germline variants. We discussed of course, when to start screening and how to do it and then we discussed this whole issue of initial versus repeat biopsy. We decided to divide it in that manner and we did for the first time, the AUA guidelines address imaging and biomarkers and then biopsy technique. Those were the key highlights I think, that we added in this guideline.
Matthew Cooperberg: Yeah, great. Walk us through what is the recommendation now? When should we start? When should we... How should we do it? When should we stop?
Daniel Lin: Yeah, and so we specifically stated for a patient at average risk, to start between 45 and 50, which was a little bit earlier than our earlier ones, and you know the data better than anybody, is that if you look at Goteborg and ERSPC, they were at 50 and 55, respectively. We said even 45, because there are data looking at very early PSAs, maybe one PSA at age 40, that kind of thing. You've seen the Malmo data and we felt like 45 to 54, in that range, we gave a range, for an average risk patient was the time to start screening.
Matthew Cooperberg: Can we talk about that? It's a methodologic decision, to include those papers. The last time around, the reason we got that 55 to 70 recommendation, was the AUA panel back then, made a decision to only include the ERSPC core age range, which is the same decision that US Preventative Services Task Force has made in their [inaudible 00:05:50]. Why now include Malmo and the Harvard Cohort and these other cohorts?
Daniel Lin: Yeah, the Goteborg was part of the ERSPC. That was actually a fair game. That was at age 50. I think that, that... Yeah, then now we received that, looked at it with higher rigor and said we can include that one. That brought it down to 50 and then with regards to modeling, we had a long discussion about whether to include modeling papers as well as the earlier papers for earlier PSA and again, the decision was as a consensus at a high level of evidence to say that we could, particularly the Goteborg, of course.
Matthew Cooperberg: All right, we start at 45 or 50. You mentioned the high risk groups. Is anyone, we should start at 40?
Daniel Lin: Well, yeah, we actually say in there very specifically on a statement, men with Black ancestry, strong family history, germline variants, to maybe start five years earlier. That'd be 40 to 45. As you know, there is evidence to that. If you look at the lead time for particularly Black ancestry men, it's somewhere in that. Well, the number I think is three to nine, but we chose five to 10 years, so five years earlier for lead time and we felt like saying 40 to 45, was a reasonable time period for those at higher risk.
Matthew Cooperberg: Yeah. Let's walk forward a couple scenarios. The 48-year-old with a PSA of 0.4, how long can he now wait before checking again?
Daniel Lin: Great question. We describe intervals and we said a two to four year interval. That's based on modeling evidence. If one looks at, and there's a nice table in there looking at modeling, but basically saying every other year biopsies, to every four year biopsies, depending on the baseline biopsy and there are now calculators that look at risk, yeah.
Matthew Cooperberg: PSAs not biopsies, right?
Daniel Lin: No, no, but yeah, exactly. One can look at that and then there's a number needed to screen, number need to detect and those numbers actually get pretty far down. A lot of people quote the ERSPC of hundreds of patients and treat many patients, to save that one patient, but if you look at really what's coming out now for screening, and you say that the number needed to screen versus the number needed to detect, those numbers are getting pretty low.
Matthew Cooperberg: They're starting to compare favorably to anything else we do in preventive oncology, right? I mean,
Daniel Lin: Yeah, or preventative medicine. Preventive medicine.
Matthew Cooperberg: Absolutely. Absolutely.
Daniel Lin: Yeah.
Matthew Cooperberg: All right. We should start around 45 to 50 for most men. When do we stop? Should a man who's never had a PSA and is 72 and healthy, get a PSA? Should a man who who's had a PSA every five years up to 70 and it's always been 0.4, can he stop?
Daniel Lin: Yeah. There used to be as you know, just some of these arbitrary age cutoffs and I think now, we talk in life expectancy rather than just in age, because as you know, you said a 72 old man, that's healthy. What does that mean? There are life expectancy calculators. We put them in our guidelines saying, "Hey, calculate life expectancy." If it's over 10 years, we think that actually screening could be reasonable. That's what we're saying.
Matthew Cooperberg: And what thresholds should drive further work up, once you've had a PSA test?
Daniel Lin: Yeah, that's a great question. Again, there's no magical threshold. I think that we put in age ranges, and this is really good for our primary care colleagues. I don't think most of our primary care colleagues realize that a man who's 45 with a PSA of 2.5, his normal PSA, but it's high.
Matthew Cooperberg: That's really high.
Daniel Lin: It's very high and we see that quite a bit, and that's a message to our primary care colleagues that we can all give, is that there are age associated medians and means, and we should have them in our minds a bit and those all play into calculators. We actually promote some calculators looking at age, PSA, DRE, and so forth, and we actually list them. There's all various different ones, as you know, and they're difficult to do and as you've been championing, putting them into EPIC as they automatically feed out, I think someday we could do that. As of now, we're plugging it into an online interface. It's a little bit awkward.
Matthew Cooperberg: Do you use the PCPT or a different one?
Daniel Lin: Well, there's that, and I use PCBG, but I look at them both and I say something's halfway between, because if you look at PCPT and you look at some other ones, there's a disparity there and the truth is probably halfway in between. That's how I've been looking at it.
Matthew Cooperberg: Yeah. Okay and then what does the guideline have to say about MRI and biomarkers?
Daniel Lin: Yeah, we're splitting MRI into use in the initial versus use in the repeat and I think the sound bite would be, the easy one, is if a man has had an initial biopsy with no MRI, and they have reason to get a repeat, they've got to get an MRI. We're saying that pretty hard and fast. In the initial, it's a difficult situation. If you read our statement, it just says that men may use MRI. It doesn't say they should, okay and there's a very long description about why, and if you look at really the strength of evidence, even the European zone evidence, and there's the analysis first author's [inaudible 00:10:36], in a Cochrane review, in the initial biopsy setting, it's actually not significant for MRI for detection.
Now again, we can debate this and there's NPVs that we can talk about and so forth, but for the initial biopsy, we're not mandating it for a repeat. If they haven't had an MRI, we're saying yes. That's MRI. For biomarkers again, we're taking a very similar line to ASCO and others saying, use a biomarker if it's going to change your decision making. There are some of the calculators that incorporate biomarkers, most of them don't. Most of them go incorporate 5 or 4K or any of the urine biomarkers or anything like that. Confirm MDX is not included in, but we mentioned them. We give the evidence that you can use them, but they're not mandated. We do say that there shouldn't be like a reflex. I don't think there should be a reflex, everybody that has an elevated PSA gets a biomarker and I think that there are some practices that are doing that. There's very little data to back that up.
Matthew Cooperberg: Well, we've talked about that. If you are going to screen young men with a low threshold, again, let's take the 48-year-old, now he's got a PSA at 1.1 instead of 0.4, he's high for his age. He's already 90th percentile. Does he go straight to biopsy or should we think about imaging or marker?
Daniel Lin: Those are the situations that one would think about, about a biomarker and imaging is, again, if it's going to change you and push you in a confidence level one way or the other. I do say, I tell everybody, life is a bell shaped curve. Most people are in the bell. If you're on either end, it's great for a test. A test is going to maybe show you either end, but most people are here, which you already know they are, based on clinical information.
Matthew Cooperberg: We're not ready for the UK model of nobody gets a biopsy without an MR lesion?
Daniel Lin: Well, great question. The paper that you're referring, one of the recent papers and so forth. Yeah, and not yet. We're not.
Matthew Cooperberg: Yeah, and I think a lot of us would agree, MRI consistency is probably not where it needs to be in the US.
Daniel Lin: Not yet. I think I'm a MRI optimist though, because I think there's going to be some AI and maybe a computer rated methods to make MRI a little bit less variable. I think that, that will happen.
Matthew Cooperberg: We'll get there.
Daniel Lin: And I think we'll get there eventually. I'm hoping that our healthcare system will look into either biphasic MRI or some MRI, that will cost a little less. When one looks at the, how many Euro it costs in the UK versus us.
Matthew Cooperberg: 300 versus a couple thousand.
Daniel Lin: It's difficult.
Matthew Cooperberg: In your own practice for that patient, do you think imaging or do you think biomarker as the...
Daniel Lin: We've been going towards imaging more.
Matthew Cooperberg: Toward imaging?
Daniel Lin: Yeah. Except for if it's a repeat biopsy. We have been looking at the tissue based ConfirmMDX and other biomarkers in that setting. If they've already had an MRI, of course, but usually they have.
Matthew Cooperberg: Do you think reflex testing could have a role in primary care at any point?
Daniel Lin: I think that it would be interesting.
Matthew Cooperberg: Yeah.
Daniel Lin: I think if your vision of having it baked into EPIC or Cerner or whatever one uses, that though might produce this calculated risk that overcomes some of the biomarkers. We know that biomarkers work a little better than PSA. We do. We don't know how necessarily they work within a really multinomial model, sometimes.
Matthew Cooperberg: Yeah, another question is what do you do with the number? I always remember a comment by Ian Thompson years ago when he published the calculator, that you can take too well-educated men with a 7% risk of high grade disease. One decides to get a biopsy, one decides not to and those are both perfectly good decisions as long as they're educated decisions.
Daniel Lin: Yeah, perfectly good decisions.
Matthew Cooperberg: Just to shift gears a little bit, let's talk about this question of transperineal versus trans rectal biopsy, which the new guideline takes on explicitly for the first time. Of course, the European guideline came down down very strongly in favor of transperineal . I think the AUA is a little more neutral with the trials still pending. There's been some chatter already on social media about this. Why wasn't the AUA more pro transperineal? Where are we at?
Daniel Lin: Yeah, it's an interesting idea to see some of that on Twitter. Yeah, I would say I agree with you. The AUA came down more neutral or lukewarm, more or less saying one can do either and we didn't come down heavily on one, based on evidence and I think that we looked at the evidence at again, a very high bar.
The highest bar, of course, is randomized clinical trial, of which there really are none, at least reported out, until on Sunday of this meeting, there will be a randomized clinical trial reported by Biedarman, and that was over 750 patients and I'll tell you about that in a minute, but I think that we looked at the strength of evidence mostly. If you look at the European data, a lot of the European and Australian data, usually largely single center, or there was one meta-analysis, based on historical controls that did look like the infectious risk was much less and I think it is probably.
Although we knew and we quoted two ongoing randomized clinical trials, one here in the United States by the PCORI group, Jim Hu I think leads that one and then one in the UK. They're over a thousand patients each. They're ongoing. I always say to myself, if there's random clinical trials trying to answer the question, should we say that it's a standard? And we decided not to, albeit the UK one is based on detection with a secondary endpoint of infection.
The United States one, the primary endpoint is infection and then as I said at this meeting at the AUA, in a few days, there will be a randomized clinical trial, over 750 patients, and it's already released. I can say this, is that over 750 patients randomized it to transperineal versus trans rectal and the primary endpoint's infection risk, and they were equal or not at least significantly different. Which again, begs that question whether we should say definitely do one or the other. As you know, there are a lot of strategies to mitigate some of the infectious risks, which we do, whether that be rectal swabs, those types of things. We actually do give an IM injection at the time, as well as an oral. We were doing other strategies, and many people are.
Matthew Cooperberg: Yeah. What are the unanswered questions?
Daniel Lin: Yeah. Well, that's one of them.
Matthew Cooperberg: Yeah. Yeah, sure.
Daniel Lin: I think that one of the unanswered questions shortly, is this transperineal versus transrectal, and I think that'll be answered in the randomized clinical trials between now and the next update. I think that there are definitely some unanswered questions with regards to use of other markers that are coming out. We didn't have a chance to talk about the PRS signatures that are coming out now, the polygenic risk scores, in combination with other markers and I think that might be the UK way and eventually being a reflex test, so then you do maybe a lower cost PRS or biomarker of that, promotes doing an MRI and then if that's something, and then doing a biopsy. You can see how you can... A little more step-wise than we are now. I think those questions might be answered in the future.
Matthew Cooperberg: We can talk a little bit about those decisions. When you use a PRS, versus a 4K, versus a [inaudible 00:17:28] verses a MI Prostate Score, the list is longer every year. The PRS is mostly competing in that space, unless you believe we're going to use it to decide who needs to get a PSA test in the first place, which I don't think we're there, but how do you see PRS slotting in with those other post PSA markers?
Daniel Lin: Sure. I think that right now, at least the way I look at it, we look at some of the strengths of the test and what they're useful for. PRS at this point as you know and Burcu Darst, at the Fred Hutch, with us is really doing a lot of work in this and it's right now a binary yes or no cancer. It's not yes or no cancer and then if it's bad cancer and once we get to that, then maybe you do that first and then you use a test that's more geared towards grade group two or a higher cancer, whether that be MRI or biomarker.
You choose which one and then at that point, you decide the biopsy. It's interesting, this whole thing of MRI, when I'm thinking about it, across the pond, they're thinking PIRADS one, twos and threes are normal and I don't know about you, but I think a lot of people in this country still include the threes in their [inaudible 00:18:42]. Whether they're using it to try to biopsy, I don't know, what are you doing, but actually there are some people in my group that ignore the threes, and there are some people that don't. I do still, but if you look at the data, it's a little bit variable.
Matt Cooperberg: Well, part of the problem, is you get two radiologists and they're going to give a different number, even within a good center.
Daniel Lin: Yes.
Matthew Cooperberg: Geoff Sonn had that terrific paper from Stanford, a courageous study. I'm sure it's no better at any other high volume MR shop and Antonio Westphalen had the paper looking at center versus center, 25% to 75% positive predictive value for PIRADS five, in finding high grade cancer and those are high volume centers.
Daniel Lin: Those are high volume centers. I've had friends say the first thing they do when they look at the MRI report, is who read it and unfortunately, that's how we are now, but again, I'm an MRI enthusiast. I think we'll eventually figure that out over time.
Matthew Cooperberg: I can't remember the last biopsy I did, without having an MR to look at first, but it's a reflex test. I'm not sure we're ready in this country.
Daniel Lin: I don't think we're ready in this country and again, for sakes of completeness, because we did mention this, there are those that are using other imaging modalities, micro-ultrasound, I think you have one at your shop, but there are micro ultrasounds out there, that there's going to be a criteria for that as well and that might be actually within the hands of the Urologist. A little bit more facile.
Matthew Cooperberg: Yeah, this is a big update from the last one, how does AUA get this out to the ones who are actually doing the screening, which is not us, it's the primary care world.
Daniel Lin: Yeah. Well, I'll say Martin Minor, who's a primary care physician, was on our panel, and he offered great insights on this. It's difficult. I think that they're running a big battle. They have to think about colorectal screening and diabetes and cardiac. It's one of their many things, and I'm sure they're relieved at the initial USPSTFD, and then they've got to bring it back in again. I think part of it might be this reflex of what we're talking about, with it baked into EPIC, but until then, well, it's things like what you're doing today, but also publicity within their own congresses and other meetings. It's going to be a hard task and as you know, as you said before, there's NCCN, there's AUA, there's their own family practice guidelines. There's USPSTF, and where do they go if their information's tough?
Matthew Cooperberg: We did a poll on this question years ago, and the answer was overwhelmingly USPSTF, when it comes to prostate cancer screening, for better or worse. We can talk about that guideline process for hours as well. It does seem like we need engagement by Urologists with their own community referring docs, to try to get the message out there. This may be a situation of all politics is local, but does AUA have any strategies to engage with these other organizations directly, to try to get convergence to the guidelines?
Daniel Lin: Yeah, I think that the Urology Care Foundation, I'll shout them out, full disclosure, I'm part of their Prostate Health Committee, has a campaign for that. There are really nice web interfaces, that we're updating now. The guidelines are out. We have to update them again, but their urologyhealth.org, it's a really easy thing. Just look at urologyhealth.org. We're going to have everything on there. Will the primary care doc do that? I'm not sure. I don't think the AUA at this point in time that I know of, have this massive plan to the Family Practice Providers in the nation and will be something that we'll have to battle. Yeah.
Matthew Cooperberg: Interesting. Anything else you want us to take away?
Daniel Lin: No. Read the guidelines. Yeah. Thanks to everyone who helped, and thanks to all the reviewers and the support.
Matthew Cooperberg: Yeah, thanks for all your work on it and thanks for joining us today.
Daniel Lin: Thanks.
Matthew Cooperberg: Great.