Active Surveillance for Low-Risk Prostate Cancer - Laurence Klotz and Peter Carroll
March 13, 2020
Laurence Klotz is a Professor in the Department of Surgery at the University of Toronto. He serves as President of the Canadian Urology Research Consortium, Chairman of the World Uro-Oncology Federation and Editor Emeritus of the Canadian Urological Association Journal. He served as Chief of Urology at Sunnybrook Health Sciences Centre and is a renowned clinician in the area of Active Surveillance. Dr. Klotz was awarded the Order of Canada, the highest civilian award in Canada for his prolific, distinguished work.
Peter R. Carroll, MD, MPH, Professor, Department of Urology, Ken and Donna Derr-Chevron Distinguished Professorship in Urology, Taube Family Distinguished Professor in Urology, University of California, San Francisco.
Matthew Cooperberg, MD, MPH, FACS, Professor of Urology; Epidemiology & Biostatistics, Helen Diller Family Chair in Urology, The University of California, San Francisco, UCSF
Conference Coverage: ASCO GU 2020: Implementation of Novel Tools in the Active Surveillance Pathway
Matt Cooperberg: Good morning. I'm Matt Cooperberg from the UCSF Helen Diller Family Comprehensive Cancer Center in San Francisco. This is the inaugural video in a new UroToday Center of Excellence focused on localized prostate cancer in which one of our major areas of interest is going to be active surveillance and evolution of management for low-risk prostate cancer.
I can think of no better way to launch this new center than a conversation with two of my true heroes in the field, Dr. Peter Carroll, who has just finished a 25-year tenure as Chair of Urology at UCSF, and Dr. Laurie Klotz, who's Chief of Urology at the Sunnybrook Institute at the University of Toronto, who together with Dr. Ball Carter of Johns Hopkins really launched active surveillance in North America and brought it from an initial academic curiosity to a real academic endeavor and now finally to the standard of care for men with prostate cancer across the country.
So thank you very much for joining us and taking the time this morning. I think we really do want to start with just a story. We all see these early reports from Willet Whitmore and others about do we need to treat prostate cancer, et cetera, but active surveillance really launched in the '90s, and I'm really curious about how it started for you both looking back, how the idea originated and how this became integrated into your formal practice.
Laurence Klotz: Sure. So I say the intellectual precedents were a few papers. One that influenced me was by Nick George from Manchester. That was conservative management, no grading; it was really another era, but a 4% mortality at seven years. And also, we knew that micro focal prostate cancer developed with age, and there had been a consensus in the '80s, '70s, and '80s when I trained that, for example, T1a prostate cancer should not be treated post-TURP and that in fact, you didn't even tell the patient they had cancer. So in retrospect, that was a very advanced, modern idea, and then somehow PSA came in and we forgot that, and all patients were being treated. I would say the pundits really felt that this was the right thing to do, and Pat Walsh aggressively promoted radical prostatectomy for Gleason 6 because you could achieve excellent nerve sparing without risk of compromising the oncologic outcome.
But I think a lot of people felt that over-treatment was a problem, although there was not a consensus about that. So to make a long story short, I went out with two radiation oncology colleagues from my center for lunch. We assigned ourselves a project to come up with a way to reduce overtreatment, and so we came up with this idea: why not watch these guys with low PSAs and low grade? If the PSA rises rapidly, we'll treat them. If their grade increases, we'll treat them. Seemed like a simple idea, had no idea that would really go anywhere, and so we started and after five years we had 250 patients, no deaths, first publication, and that really created a storm of controversy, which is another story, but it came out of multidisciplinary collaboration, which I think was a key part of it.
Matt Cooperberg: Yeah. Sure sure. And what about your experience?
Peter Carroll: Well pretty much like Laurie mentioned, the early history of it. For me, it became almost a moral issue when I saw the impact of PSA. As a busy surgeon, I saw the impact of PSA, this issue of overdetection. I wrote an editorial many years ago in the Journal of Urology, "Early stage prostate cancer--do we have a problem with over-detection, overtreatment or both?", and I said at the time urology will be judged on how they treat this disease and it was clear to me that we were over-treating and over-detecting it, and of course I felt the right thing to do was consider surveillance, and of course we had to document our outcomes, and then we rapidly accrued a large patient population we have today.
Matt Cooperberg: I think one can say we have been judged for what happened in the '90s through the task force machinations. Were you in communication with each other and with Hopkins or did these efforts really start in parallel and separate protocols?
Peter Carroll: I think we all started, the patient populations were a little bit different, how we evaluated them. We were obviously ... Laurie and I are close, as well as Ball Carter. We were actually all in New York at the same time, actually. He was a resident, we were fellows. But we all started a little differently but kept in close communication over the years.
Laurence Klotz: Yeah, I think we weren't talking to each other at the start, but it rapidly became apparent that this was a good idea that it occurred to more than one group at the same time, and because there was a storm of opposition, it was very comforting to have allies. We felt a little bit like the Résistance; we were fighting some large forces. Obviously there were a lot of vested interests in continuing to do surgery and radiation for these patients.
The one other point I think is probably worth commenting on, we did have some philosophical differences. So we were quite inclusive at the start, included 3+4s, PSA over 10, and particularly Ball at Hopkins, he deserves a lot of credit because he's promoting surveillance in the temple of radical prostatectomy, but very restrictive.
So we had kind of a different philosophy, and initially we kind of looked at each other like ... I thought, "Why is he being so restrictive?" He thought, "Why are you guys being so liberal in our inclusion criteria?" And ultimately actually we converged in the middle. So we got along much better after that.
Matt Cooperberg: On the topic of active surveillance for Gleason 3+4, interestingly enough, literally just yesterday Brian Chapin put out a poll on Twitter, of course, the platform for most academic discourse these days, asking urologists if you had prostate cancer as the urologist, you had 3+4 in a single core, what treatment decision would you make with active surveillance, focal therapy, surgery, radiation, et cetera? Seventy-five percent of the respondents, over 200 urologists, said now they would choose active surveillance, and I think that's a real sea change from where we were even just a few short years ago. So where do you think the field is evolving specifically in terms of active surveillance for 3+4?
Peter Carroll: Well, I think urology has been held hostage by the grading system for too long, and I think in the era of precision medicine we have to step back a little bit. I think grading has lost some of its prognostic utility over time based on changes in grading by a pathologist, and I think we just need to be a little bit more thoughtful. Look number one for low volume 3/4, be sure you've had it graded correctly. Two, look at the amount of pattern four.
For us it's actually volume more than grade predicts outcome for 3/4 patients, and now we have these other markers of MRI and genomic testing, and we have to think, like all of our colleagues in the field, that may be precision medicine means that we'll interrogate these tumors a little bit more to better understand where we are in the spectrum of disease.
And remember for the CAPRA score, which you developed and tested around the world in tens of thousands of patients, going from a 3+3 to a 3+4 alone gets you one point on a 10 point scale. So you have to be careful that we're not over assessing risk. Now I do agree with Laurie that a higher volume 3+4 is a different story. I think that's when you have to be a little more cautious.
Matt Cooperberg: Have you started re-accruing 3+4s?
Laurence Klotz: So listen, there's no question many patients with 3+4 have indolent disease and can be managed with surveillance. The question is how you pick them. We did not pick them too well and our own experience was discouraging, so when we looked back with 15-year followup, we had about 120 3+4s, and their actuarial metastasis rate of 15 years was around 30%, way too high. But I think it was all about patient selection and pre MRI, pre biomarkers.
I think surveillance for three plus small amounts of four is reasonable, but this is clearly the sweet spot for molecular markers. And I have to say, to me, Peter, what you're saying is a little bit sacrilegious because the percent pattern for going back to Tom Stamey has been such a powerful predictor of outcome and I think it still is, so I'm not ready to say to abandon Gleason grading. I don't think that's what you're saying either, but it's still so powerful, but clearly it can be augmented by molecular genetic tests and imaging and so on.
Peter Carroll: Yeah, the one caution I have to that, we're in a different era nowadays. That was a six-core biopsy or 12 -ore, and now we've gone to MR targeting, repeated biopsy and over-sampling, so I worry a little bit now the risk is being inflated because we're using new technology to over-sample the dominant tumor.
Matt Cooperberg: We're now entering an era where molecular testing is becoming more and more widely available and advanced imaging likewise is being used more and more frequently, but I think there's still a lot of uncertainty in terms of which biomarkers and which imaging tests are the most useful and can offer the most value at the point of clinical care for men on active surveillance today. Of course, the ASIST trial, which was a wonderful trial, did not necessarily support routine use of MRI for all men with active surveillance, but I'm curious where we are heading, which of the biomarkers since we now have more than we really know what to do with, which of them may be the most useful and in which contexts, which imaging tests may be the most useful and in which settings.
Laurence Klotz: So first of all, the ASIST trial actually, with followup, did show value of the MRI. That was just published. The whole thing is a little confusing. Won't get into that now. So clearly MRI has ... Let's say imaging because micro-ultrasound I think is emerging as possibly an alternative, but, accurate imaging of prostate cancer is clearly a game-changer. It's made a huge difference, but not perfect, and there are all kinds of data emerging now about the limitations of imaging; it misses significant cancers and so on.
The genetic biomarkers I think clearly have a role but not in the routine, particularly the very low-risk to low-risk patients. These patients have such a low-risk of metastasis, particularly if they have favorable imaging or targeted biopsy that doesn't show anything. So I don't promote the use of these tests which are fairly expensive for the routine, very low-grade to small volume Gleason 6, grade group 1, but for sure the 3+4s. If I was going to go on surveillance, I would want to know that my risk profile was favorable.
Now having said that, you're going to get these guys with grade group 1 with an adverse risk profile, and I'm concerned that may be a trigger for treatment, which it shouldn't be. That's saying there's a risk of adverse pathology; look for it, find it, but don't treat the patient because they might have it. So that I think is one of the limitations.
Peter Carroll: Yeah, I'd say the same thing. I think imaging has value. We do a lot of testing on biomarkers, as you well know. For us in multi-variable analyses, the biomarker appears to predict better than the MRI, and I think part of that's because we're not getting as much information from the MRI as we could.
I do agree with Laurie that the advantage to molecular testing is there are these patients who are on the fringes of surveillance, they're low or intermediate-risk patients, but I also agree with him totally that there is no biomarker result that absolutely predicts or excludes progression, so you use it as one more bit of information, and that doesn't necessitate treatment, but you might actually want to interrogate that patient more carefully, do a confirmatory biopsy a bit earlier than you would otherwise.
Laurence Klotz: If I could just add one more thing. This field is on fire. As you mentioned, new biomarkers, they're getting better and better, so I'm really confident, and this is paralleled with advances in other areas of oncology, particularly for example the hemato proliferative diseases. It's like dynamic risk profiling over time using liquid biomarkers.
So I am almost certain we are going to have a test in another few years that'll be very useful for following the patient. If favorable, suddenly there's a switch and you have a signal for intervention or at least for a biologic change in the phenotype of the tumor. That is where this is going, to me, very clearly.
Matt Cooperberg: So now that active surveillance has become widely endorsed by virtually every major international guideline as the preferred standard of care option for men with low-risk prostate cancer, I think a lot of attention is now shifting to how we should do active surveillance. On the one hand, it seems to be increasingly clear that not all men need a biopsy every year or even every two years, but on the other hand, there are data from the MUSIC Consortium in Michigan and elsewhere that many men who are initiated on what is meant to be active surveillance do not actually complete even the minimal requirements for surveillance. We know from the ProtecT trial that an inactive approach to surveillance is not really adequate. So where are we heading in terms of how we should really conduct active surveillance?
Peter Carroll: Yeah, we clearly have to make surveillance less intense for patients, so you'd get out of the periodic biopsies frequently that are being done. So I think good predictors of non-progression are, as you well know and you've submitted papers on this, on low PSA density, low PI-RADS, favorable genomic score, anyone who has a negative biopsy, 20% of patients in surveillance will have a negative biopsy, all these predict for non-progression, so these are the patients who can be followed less intensively versus a patient who has high PI-RADS, high genomics, unfavorable genomics, unfavorable MRI.
Laurence Klotz: These patients need to be followed, so I'm not ready to say there are patients who are so favorable that you can just say goodbye, and in fact, we have some data that patients being followed, particularly having at least one confirmatory biopsy and then periodically over time, actually translate into mortality difference. The ones who were diagnosed and then refused further diagnostic follow-up had a higher mortality. So I think it really makes a difference because some of these people have bad disease and you have to find it.
But yeah, it's exactly what you said. It's risk stratification using whatever tools are available. Now it's mainly MRI, periodic biopsy. Clearly the MRI is replacing some of the biopsies. We're now at about a five-year biopsy interval for the average patient, and I think you have to individualize this. We're not at a point where you can have easy rules.
Matt Cooperberg: I want to come back to your comment, Laurie, about men with T1a prostate cancer not even being given the diagnosis of cancer back in the 1990s because now we're in a situation where men may have 5% of Gleason 3+3 in a single core out of a systematic biopsy, the MRI is negative, the confirmatory biopsy is negative, maybe a biomarker test is negative. Are we ever going to get to the point where we can undiagnose some of these men or give them a label of something other than cancer? One of my favorite controversial questions.
Peter Carroll: I don't think so actually. We've shown that even some people who have repeated negative biopsies can progress in the future. And I think we have to approach cancer like we do heart disease. Not everyone who has heart disease gets a cardiac bypass, and so I think the public has to understand that cancer is a disease of increments, and of course, with all of our advanced imaging and molecular testing, we're seeing cancer rates in different patients go up over time, so I think we have to just let people know that cancer is something that happens to them and it doesn't mean that it's a death sentence.
Laurence Klotz: I would say not calling it cancer would make our life easier, although actually our life's become easier because I think the concept that you just referred to has become actually more widely known amongst laypeople, such a thing as indolent cancer. But really the nomenclature lies with the pathologist, so they're the ones we have to convince, and my impression, having talked to a lot of people all over the world about this, is they're not ready to abandon the word cancer because they see features of cancer even in Gleason 3+3, occasional invasion, et cetera, seminal vesicle involvement. So, it's different from PUNLMP in that regard, and therefore I don't see it changing, but obviously there are some pragmatic grounds to make that change.
Matt Cooperberg: Yeah. So the last question I'd really love to get your thoughts on is where the field is heading. Where are we going to be five years from now? Where are we going to be 10 years from now? What will active surveillance look like for men with low-risk prostate cancer in North America?
Peter Carroll: Well, I think we'll see more of it in some situations. I worry a little bit now on the move towards focal therapy that we could see another round of over-detection occurring, so I think we have to be a little bit more thoughtful about how we apply focal therapy. I also think we have to be cautious about ... We talk about surveillance in North America, but I was in China recently and they felt that they really couldn't apply surveillance to their population. People coming from a long-distance, they're not doing MRIs, things like this. So we have to be a little bit more thoughtful, and we will not get out of detecting low-grade disease. Even when we use these new markers as specificity, it's still about 35% of our biopsies at UCSF, even when you use MRI and 4K and others, we see that we detect low-risk cancer in about 30% of patients.
Laurence Klotz: Yeah, I think clearly there's a move away from systematic biopsies towards initial molecular liquid biomarkers, MRI, so we're probably going to be seeing less grade group 1 than we have been, but it's not going away, and so I think the basic issues will still be there. The cliche of personalized medicine I think really applies here. We are going to have better tests that allow us to really stratify patients accurately.
The marginal utility in terms of mortality reduction is small. Hopkins mortality was 0.5%, ours was about 3%, so I think the gain is going to be less over-treatment. We're still treating a lot of these guys because we think they have higher risk cancer who probably don't need to be treated.
Matt Cooperberg: Well thank you so much again for your time today and for all, you've done for the field and for men with prostate cancer over the last 25 years. It's really been tremendous, tremendous, unparalleled contributions, so thank you.
Peter Carroll: Thanks for having us.
Laurence Klotz: Thanks.
Matt Cooperberg: Thank you.
Laurence Klotz: Thanks, Matt.