Chapters Transcript Assessing Oncological Outcomes Following Focal Ablation Course: Advances in Screening, Detection, and Treatment of Prostate Cancer So, is, and I think this is a very uh important topic because there's very, very, very little guidance in terms of how do we assess uh oncologic outcomes following uh focal therapy of prostate cancer. Uh, so disease can recur, uh locally in field, and that can be the margin. Uh, or just we didn't adequately treat the lesion. It could be out of field. That could be due to preexisting disease, uh, the MRI invisible disease, or the development of new disease and you can get metastatic recurrence could be the preexisting uh disease that was missed or the development of the metastasis from a local recurrence. So of course, what are the oncologic testing? We do PSA, DRE imaging. Is MRI good enough? Do we need to do PET imaging and biopsy? And there's no consensus as to testing and timing uh of testing for oncological surveillance. So, uh, the PSA, uh, when Jim and I started our, our program, we said we're gonna do PSA at 3 months, 6 months, and every 6 months, uh, uh, uh, uh, inevitably, that we would do an MRI at 6 to 12 months, at 2 years, 3.5 years, 5 years, 7.5 and 10 years. And additional MRI's at the discretion of the surgeon. And actually the DRE we only do, uh, if they're a scheduled visit because many of our visits are remote. And we said, OK, we want to be rigorous, we're gonna do a biopsy at 6 to 12 months, so we're gonna do the biopsy, the target, the ipsilateral systematic uh biopsy, uh, anything that showed contrast enhancement. Uh, as I mentioned in August of 2020, uh, we abandoned the, the, the protocol biopsy because of very low rates of clinically significant disease. Uh, at two years, we actually did the target, we did 12 systematic biopsies, we did, uh. Uh, uh, another biopsy of any lesion, we actually took the, uh, uh, the original MRI and segmented the lesion on the MRI and ultrasound at 2 years. Uh, and because of the low rates of actionable significant disease, we actually abandoned that. And at 5 years where we're now moving forward, we got 100 patients at that time point. Uh, we biopsy is no longer protocol mandated. We biopsy of a rising PSA, a new target, uh, a pre-treatment contralateral low risk disease because our two year data showed that that patient had a higher risk of recurrence, but at 5 years, that's no longer the case. So, the decision to perform oncologic testing with PSA imaging biopsy must be driven by the prevalence and the detection rates of clinically significant prostate cancer. So the true rates of clinically significant cancer and the performance of the diagnostic test to detect clinically significant cancer requires a protocol biopsy on everyone on all subjects, independent of the presumed uh risk for clinically significant cancer. And the testing and biopsy, low yield, uh, if low rates are clinically significant prostate. So that's why we abandoned the, the testing, uh, at 6 months, uh, in, in 2 years. So the performance of the test. Cannot be extrapolated from one center to another due to inconsistent patient selection and aggressiveness of treatment. So you can't take our data and say, you know what, I don't need to biopsy at 6 months or 2 years. You gotta show yourself that uh that in fact, uh, is relevant to, to your cohort. I showed this. Why do we abandon 6 month biopsy because we only had 2% risk of uh clinically significant disease. Why did we abandon uh the uh the biopsy at 2 years, uh, and that we was because uh uh clinically significant cancer-free survival, uh, in field was 96%, uh, out of field was 86%. Um, and the time to failure in terms of freedom from failure was 97%. And then when we looked, I think there was an important concept of of actionable recurrences. So if we actually looked at that rigorous protocol where we biopsied virtually every single patient that we were following at 2 years, and I think that was close to 75-80%, we only found 6% of cases that we actually did something about. Because a good percentage of those clinically significant cancers was remember simply having any pattern 4. But in a good number of those, it was such a tiny percentage of pattern 4 we didn't do anything. So now we're at 5 years, as I showed you, uh, and we look at in field, we're still 86% uh have no uh clinically significant cancer. Uh, as out of field about the same, so we're now finding in field is sort of catching up with the uh with the with the at the still low, and then we look at everybody, it's 70, uh, it's 70%. So let's take a half a second and talk about the utility of PSA testing. Who wouldn't get a PSA, right? Simple. If you are following patients for active surveillance, you get a PSA. There is no validated longitudinal PSA uh expression triggering an active surveillance biopsy and upgrading of disease. So look at all the guidelines, look at all the papers. Is there ever a single um guideline that says how you should use PSA in in determining who needs a biopsy on a patient of active surveillance? Uh, and so we're fortunate, one of our meds, one of our residents was a med student, uh, at Cornell. He had his degree in biostats. He had his own company before he's joined us as a resident. And so I charged and was saying, OK, here's all of our data. You take any parameter that you can think of for PSA absolute change, percentage change, PSA velocity, consecutive change, in other words, two consecutive changes above 0.5%, 3 consecutive changes above 0.5, or the slope of the regression line, and we know who had a significant cancer. We can't predict. So we get PSAs, but I can't tell you. Uh, how you should use PSA in determining who should have a biopsy, and we really can't figure it out. And so here's just a, a patient, he was 58, um, he had a rising PSA and this was pre when he came to see me, 0.67, 2.79, 4.9, you know, that's sort of a rate of PSA with others in between. I will show you that's a PSA velocity. I'd worry for cancer. He had a 4K test, it was 20%, uh, and he had a, a, a pyre 3 in the transition zone, uh, apex. Um, and this is, you can see that, uh, that, uh, uh, high signal intensity, OK? Uh, and so, uh, the target, almost all the target biopsies were positive. The, and you would expect the random biopsies to be negative because this is an anterior lesion. So we talked about all the different options, radical radiation, emphasize the data with surveillance and our and our data with the partial gland uh uh the cryoablation uh and he uh elected to, um, uh, he elected to uh uh undergo uh and uh uh and uh an ablation. I'm looking to see where is his, uh, um. Uh, data, and let me just say that he had a very progressive rise in the PSA. OK, it was pretty compelling, almost like what I showed you before. So I figured if there's a guy with 3 consecutive rises in the PSA, he's a guy who's likely gonna be has disease if I'm using PSA as a, as a, as a um a predictor. Uh, and when we biopsied him, uh, he simply had one of 4 cores, uh, that was in his prior lesion that had disappeared, so about 1 millimeter of, uh, maybe 100 millimeters of biopsy. Uh, that was positive for Gleason gray group was so even in this patient, uh, who I would expect, uh, the PSA to give me a signal, uh, it didn't. So, uh, let's talk about the imaging. So again, we have 311 men and I can, we have 500 or so in the database. I'm just gonna limit it to the guys that had intermediate risk disease. And we detected clinically significant cancer, uh, in 11%. You always have to be careful when they talk about your overall rate because if most of your patients are 1 to 2 years, well, those patients, uh, they're, they're, you're, they're included in your uh in your rate of disease, uh, but that doesn't reflect uh those uh uh along, and if it's always front loaded with uh patients who only have 1 or 2 years, your, your data looks pretty good. So as we show, we're 30% at 5 years, but in this cohort, we have 11%. And so what I looked at is I said, OK, let's look, because this is the whole concept is MRI visible and invisible disease, uh, are they, uh, are they biologically different? Uh, and so if we looked at our 33, uh recurrences and then get intermediate risk disease. Uh, 2/3 were visible and 1/3 was invisible, and I hope that the resident who is here is actually working on this paper, so I assume, uh, if he's not actually in the audience, I assume that he's not, uh, uh, even though this is outside of his 80 hours that uh he's, he's working on this paper. But so let's look at the 11 invisible lesions, and remember, all these guys have clinically significant disease. Uh, and of these, 10 were Gleason grade group 2. Uh, one was Gleason gray Group 43 were infield outs, 8 were out of field, uh, but the mean linear amount of Gleason pattern 4 was only 0.6 millimeters, OK? And we look at visible disease, there were 12 that were Gleason grade group 27 Gleason grade Group 3, actually 2 that were Gleason grade Group 4, almost equal between in and out of field or both. And again, that's a significant amount of Gleason pattern 4. So if I actually looked at MRI invisible disease, what did we actually do with those guys? 7, we followed, uh, 3 had focal, uh, salvaged focal therapy, and 1 had whole gland treatment, whereas if you had the MR visible disease because of the volume, they were all treated. Now, what I don't know is uh uh in terms of this rep how many MRIs did we do in all of these patients, but I think what this is telling me. Uh, is that, and this is only relevant to focal therapy, may not be relevant to MRI invisible disease in general, but post treatment. That's a very, very, very low yield, and since the overwhelming majority of our biopsies are negative, this is why I think that we can comfortably without a significant PSA signal, without an MRI signal, we don't have to do all of those uh biopsies in the cohort that have invisible disease because it's gonna be rare to find what would be an actionable uh lesion. So, uh, uh, in our cohort of intermediate risk disease, uh, we actually did a total of 633 MRI's were performed. And so clinically significant prostate cancer was detected in 33. 19 MRIs were performed to detect the case of clinically significant prostate cancer. But if we say actionable cancer, in other words, not the low risk that we didn't do anything, we had to do 30 MRI's were performed to detect a case of uh of, of actionable, uh, clinically significant cancer. So we biopsy uh. Only MRI, so the biopsy of only MRI visible disease would eliminate many of the MRIs and of the 663 MRIs, uh we would only miss 4 cases of uh of, of actionable uh disease. And, and again, this is relevant just to our cohort of patients, uh, where the risk of clinically significant disease is, is very low. Now, let's go back to my focal laser ablation. And, and as Jim sort of highlighted the different energy sources, um, we are because you only have one probe, right? Our infield uh recurrence rates are high. So now let's look at our 2 year data that we uh that we published. So our clinically significant cancer detection rate was 9 of the 22 men who were undergoing a 2-year biopsy. So, 10 of the group, we just couldn't convinced to have uh a a biopsy. Well, that clinically significant prostate cancer detection rate is 41%. It's not what I showed you for uh our uh our, our, our cryo data. So of those that were suspicious, OK, 75%, we found clinically significant cancer. However, in those that were MRI, uh, invisible. Uh, there were 14 men who had, uh, MRI invisible disease who actually underwent a biopsy. We had 10 who were MRI invisible that refused to do a biopsy. Uh, and that rate of clinically significant cancer was 21%. Uh, in that, uh, in that cohort that we actually did a biopsy for invisible disease and if we looked at the overall cancer detection rate, uh, with our focal laser, it was 27%. I think that justifies a protocol to detect these cancers and again, uh, if it was MRI visible, 75% had cancer, clinically significant cancers, MRI invisible, it was 21%. And a biopsy of MRI visible uh lesions uh would miss 33% of the clinically significant cancers while avoiding biopsies uh in 46% of cases, and that's really not a, a, a, a trade-off. So when I look at our data, uh, the clinically significant prostate cancer detection rate justifies the effort to detect clinically significant cancer even at 2 years in. In men, uh, undergoing the laser, there's a definite indication to biopsy visible MRI lesions, and even with that rate of 21%, I would say it probably is reasonable to biopsy that cohort, how many actually were actionable, uh, I, I, I didn't drill down to that point, but the point that I'm making is the our, our recommendations are based on having a very low rate. Uh, of clinically significant cancer. Now you sort of look at some of the high food data, right? It's 40%, uh, in the first two years have significant disease. That's a whole different story of where imaging and biopsy, uh, may be relevant. Now, in terms of early contrast enhancement, uh, Emerton's group, uh, has come up with sort of a grading scale. I'm not gonna go into this, but they have a P1 FA 1, P1 FA 2, P1 FA 3, and they say if you see contrast enhancement, fab 3 is it's 3 millimeters and at the site of the original tumor, there's early focal enhancement is greater than 3 millimeters within the ablated zone edge of the ablation cavity, and they say you should biopsy this with absolutely no data. I mean, I guess if maybe you're Mark Emerton, you can get things published like that, um, but with no data, I mean, it just says here it is, there wasn't one validation of this, right? So we look back because early on, we had 6 cases, right, Jim, of, of contrast enhancement, and we biopsied them all and we only found one patient that had the disease. So that's why we didn't biopsy everybody thereafter, but look, so we had Angela who's an exceptional radiologist. Uh, she went back and looked at those 9 cases that would be fab 3, OK? And again, this was at 6 to 12 months. 4 of them, we underwent a biopsy and we didn't find cancer, right? 5, we did not, did not undergo a biopsy, but watch this, we had all of those 9 guys we followed for 2 years, and in all 9 of the lesions, they disappeared. Well, can't be cancer if it disappeared. And what were those lesions? It was granulation tissue that was because we did these uh MRI's at 6 to 12 months, which is probably too early. And actually, the one guy, uh, who we found cancer in, uh, his lesion had disappeared. So I'm not so sure. Um, if this FA 123 is validated, and that's probably why I can't, I haven't heard for a couple of months about the paper that we put in because they probably sent it to his group and they probably squashed it, right? Um, so how about MRI visible and invisible lesions because I always thought, and maybe correct me if I'm wrong. Is that Mark, uh, who I love. I mean this guy has, has really, I would say has been I think institutionally we have made tremendous contributions to this area of focal therapy. I, I have to say, uh, Mark's the godfather, but my understanding is he doesn't really think much of MRI invisible disease, and I would say. Based on our experience with very high rates of uh of uh of of or low rates of detection, that's probably reasonable. But let's sort of go to the concept of things Samir uh addresses as well. So now you go to UCLA they have 580 men who have a radical. That's a very different cohort than the guys that we're selecting for folk with therapy. And they found 1200 histologic cancers and they did this meticulous sort of co-registration of the MRI with the histology, uh, and they said that the MRI detected only 45% of the cancers. That's a pretty bad sensitivity, right? At 45%, uh, and so if it was clinically significant, it was 65%, if it was a high risk, it was 80%. Uh, and basically, if you had a Pyres 3 to 5, the positive predicted value was uh uh 81%. And I think this was a, this is very instructive. So this is looking at the cohort, remember, this could be guys Gleason 8, Gleason 9, multiple disease, extra capsular extension, seminal vesicle involvement. So, so you could see how in this prostate, there's a lot of MRI invisible disease that, you know, if you look up on top, this is what was detected, this is what was missed, you can see a reasonable number of cancers, uh, that in fact uh was missed. But I think this is important because what they did is up top are the solitary lesions. These are pretty much the guys that we would probably be doing focal therapy on, not the bottom guys. And if you look at the Gleason, um, if you look at the tumor diameter and focusing up above, uh, there really weren't that many tumors, uh, that were above a centimeter that were missed. Uh, there were very few Gleason Gray Group 23, and 4. This is on the bottom, the ones that were, were missed. So I think when you start looking at data about the whole radical prostatectomy and looking at sort of MRI visible and invisible disease, that's not relevant to the cohort that we're dealing with where you pretty much have uh uh uh an index uh uh lesion. And this again to me is probably I think the best paper, not just because it's my paper, uh, because it addresses the issue of OK, here are guys that were candidates sort of that, you know, Gleason grade group 2 to 3 index lesion, negative biopsy or low risk contralateral. And when we took out those prostates, how often did we miss any Gleason pattern uh for. And that was 23% and virtually all of those cases, the uh pattern for was less than a millimeter and I think many of us are comfortable following low risk cancer when we see it on a biopsy, right? And that's the data from the Protect trial that says we we we it's there we biopsy it. These are guys that you can't even biopsy and we take out their prostates and we drill down to this tiny little prostate cancer that we would say, you know, is uh is is significant and so that's why um I believe that not only do we have MRI unlikely that we and all these patients, it was MRI invisible, but you can see why. Now, again, the, the, the, the Pro trial, I won't go over it in detail. The unique feature is when they talk about sensitivity, it's legitimate because it's basically doing what I did, which is I took out the prostates, but they got that information from their saturation biopsy. Uh, and when you look at MRI detected versus MRI, uh, undetected lesions, uh, you could see, um, that in terms of uh the lesions that were MRI, uh, uh, undetected, uh, that the overwhelming number of those, uh, were, uh, uh, were, uh, were low risk uh disease of those that were significant. Uh, it was a, a, a, a, a, a relatively uh small uh component. Now, this is, uh, what they did is, and I, I, this is almost too small to see it and it's just as good because if you can see, it probably wouldn't really understand it too much, uh, but what you do is these are all, they're looking at, you know, MRI visible and MRI, uh, invisible cancers, and they're looking at sort of various predictors of aggressiveness. Uh, and this is a difference, this is, this is another study, uh, that basically shows what we think might be a molecular predictor of, uh, of aggressiveness. It was no difference uh between the visible and invisible cancers, right? And this is another study that was done. Um, it was out of Michigan, and they did 10 radicals, they had 26 cancer foresite, 12 were visible, 14 were invisible, and I think I understand this study, and I've read the paper multiple times, um, is that they had a 9 gene signature, um, and what they were able to show, uh, that they were actually able to predict visible from invisible disease, OK? So then they went ahead and they went to the cohort from uh Cedars, um, and they had 16 patients and they were able to show us a validation that yes, they can predict visible from invisible cancers, um, and then they got the Hopkins data, and they actually determined whether it was visible versus uh uh uh invisible. Uh, by looking at the biochemical uh uh recurrences, but they were actually able to look at the signature, they didn't have the MRIs. But they just had the signatures uh of the tissue in terms of uh determining whether it was visible versus invisible on the basis of their signature which they had validated is able to differentiate um um uh visible from uh invisible tumor and here they didn't show any difference in biochemical recurrence, development of metastases. Uh, between MRI visible, uh, and, uh, uh, invisible, uh, disease. So this is beginning to say, well, you know, um, visible, invisible disease, you know, may a certain and a radical specimen, uh may be uh significant. Uh, and then another study, uh, where this is retrospective, they had, uh, almost 2000 radical prostatectomies. This was again diagnosed by systematic uh biopsies, uh, and it sounds like they then did MRI's, uh, and they had 11% of patients had, uh, uh, invisible disease. Uh, and again, these are candidates who underwent radical. This is not necessarily candidates who we would, uh, uh, recommend focal therapy, but again, what they say is, listen, MRI invisible disease, look at the pathology. You know, you had, uh, almost what, 40% plus, uh, had, uh, um. Uh Gleason gray group 2 to 5, that 13% had extra capsular extension, 9% had positive margins, 0.5 had uh seminal vesicle, uh, and in fact these MRI lesions that were taking out, you know, well may be uh significant and then if they looked at the recurrence rates, you can see a big difference between invisible and invisible disease, uh, in a group of men. Uh, who the decision for radical prostatectomy, uh, was not driven by their MRI visibility or, or invisibility. So, sort of in conclusion, uh, I have to say, That the intensity of your oncologic testing must be influenced by rates of clinically significant prostate cancer, defined by a biopsy protocol which you should embrace when you start doing focal therapy. And if you're like the center up the street and you say you have 40% clinically significance, I, you know, what you would strongly urge them is to uh use imaging. I would tell them to stop doing focal therapy actually. And if you got a 40% rate of recurrence, you've just proven that at least the way you're doing it, what you're selecting patients, your, your confluence of energy, it it just stop, uh, but if you're not gonna stop, you should use imaging to at least identify uh that risk. So you can't take our data and say, OK, you know what, we don't need to uh uh do, uh. Uh, you know, uh, be aggressive in imaging and biopsy until you show that you have a low rate. Well, we routinely measure PSA. I don't know what to tell you, right? Get it, simple, uh, how can you not get a PSA? But again, is there anyone who has a death, you know, everyone here follows patients? Is there anyone here who will tell me. When they biopsy somebody on active surveillance because of the PSA, there's not one guideline. There's how many patients written about surveillance and recurrence. Have you ever seen anything about PSA, any predictor of recurrence, I'm not aware, but get a PSA. Uh, our low rates of clinically significant cancer at 6 and 12 months following protocol biopsy. Questions the aggressive intensity of early oncologic and that's why we've abandoned and if your MRI is negative. You know, we started doing the optimal so what's the optimal timing, I don't know. Jim and I have thought about sort of rethinking this, but we already got the protocol going. I think doing about uh uh an imaging in 6 months is too early, because I think as we showed, when you have the FA 3, it ends up being contrast enhancement that really ends up being, um, uh, I think just granulation tissue. So maybe if you're starting out, I would do the MRI at 1 year, and then 3 years, and then 5 years, uh, and 7.5 and uh and, and 10 years. They are 5-year detection rate of 70% clinically significant prostate cancer, uh following uh our clin our 5-year detection rate of that should be 30%, that's not 7 that those that haven't recurred, but those that the detection rate is, is the 30%. Following focal therapy justifies a thoughtful protocol, uh, designed to detect these cancers, and I would say that that would mean a uh uh a change in the uh in the MRI that's worrisome, uh, or even maybe uh, uh, a progressive change in the, uh, in the PSA of those who have a negative um MRI. We we biopsied everybody and we still continue to do so, but our data at least is showing that if you had low volume contralateral disease, our two year data said that was a predictor for recurrence, um, but at 5 years, that signals dropped out, but I still do an active surveillance biopsy on that group, um, and we biopsy all suspicious areas on the MRI for a conclusion. The clinical significance of MRI visible disease is controversial. I would say the majority of those cases that we find they're not actionable. So if you're gonna find clinically significant cancer and do nothing about it, um, uh, does that justify, uh, getting that, um, uh, biopsy at 5 years? We're gonna continue to follow those patients with PSA and MRI and we'll catch them maybe at 7.5 years or 10 years if that becomes relevant. The rate of clinically significant prostate cancer so MRI invisible disease in our series, uh, is too low to recommend a protocol biopsy looking for MRI invisible slash significant slash actionable disease. The majority of significant cancers associated with MRI invisible biopsy is very low amount of Gleason. 4, and so maybe the indications to biopsy MRI invisible disease is a genetic mutation, a rising PSA or primary demonstration of MRI invisible disease on a, on a prior biopsy. So that's sort of, you know, we've certainly evolved in our, uh, in, in our experience, um, and, uh, I think that I've at least gotten to the point uh that uh if your MRI is negative. And your PSA doesn't show a real progressive change, then, then I feel comfortable. I think the data supports, uh, not doing a biopsy. The invisible disease, I'm not talking about in a radical, I'm talking in this disease, generally the invisible disease that we find, uh, is not actionable, um, and so more to come as this series uh uh matures. So that's that, uh. Published June 26, 2024 Created by Related Presenters Herbert Lepor, MD View full profile
CME Knowledge Hub
