Prostate Cancer Diagnosis: MRI vs. Systematic Biopsy "Discussion"
July 24, 2024
At the CAncer or Not Cancer: Evaluating and Reconsidering GG1 prostate cancer (CANCER-GG1?) Symposium, experts debate the evolving landscape of prostate cancer diagnosis and management. They examine the limitations of traditional biopsy methods and the potential of MRI in improving diagnostic accuracy. The conversation highlights the shift towards imaging-based diagnoses in countries like Australia and the UK, while acknowledging the challenges of implementing this approach universally. The experts conclude by stressing the need for standardized, high-quality MRI access and suggest simplifying diagnostic codes to better guide treatment decisions.
Related Content:
Grade Group 1 Prostate Cancer: A Pathologist's Case for Cancer Classification 'Con' "Presentation" - Gladell Paner
Grade Group 1 Prostate Cancer: A Pathologist's Case for Cancer Classification 'Pro' "Presentation" - Theodorus van der Kwast
Prostate Cancer Biopsy Sampling: Challenges in Grade Group 1 Classification "Discussion" - Lawrence True
EAU 2018: Assessing the external validity of the updated prostate cancer intervention versus observation trial (PIVOT)
CAncer or Not Cancer: Evaluating and Reconsidering GG1 prostate cancer (CANCER-GG1?
Video Presentations: CAncer or Not Cancer: Evaluating and Reconsidering GG1 prostate cancer (CANCER-GG1?)
Cooperberg, M.R. et al. (2024) ‘When is prostate cancer really cancer?’, JNCI: Journal of the National Cancer Institute [Preprint]. doi:10.1093/jnci/djae200.
Grade Group 1 Prostate Cancer: A Pathologist's Case for Cancer Classification 'Con' "Presentation" - Gladell Paner
Grade Group 1 Prostate Cancer: A Pathologist's Case for Cancer Classification 'Pro' "Presentation" - Theodorus van der Kwast
Prostate Cancer Biopsy Sampling: Challenges in Grade Group 1 Classification "Discussion" - Lawrence True
EAU 2018: Assessing the external validity of the updated prostate cancer intervention versus observation trial (PIVOT)
CAncer or Not Cancer: Evaluating and Reconsidering GG1 prostate cancer (CANCER-GG1?
Video Presentations: CAncer or Not Cancer: Evaluating and Reconsidering GG1 prostate cancer (CANCER-GG1?)
Cooperberg, M.R. et al. (2024) ‘When is prostate cancer really cancer?’, JNCI: Journal of the National Cancer Institute [Preprint]. doi:10.1093/jnci/djae200.
Read the Full Video Transcript
Matt Cooperberg: I'm struck by the difference in the biopsy cancer rate compared to the number of biopsies done. Is there screening going on in some of the countries that have lower biopsy yield rates?
Speaker 2: In the secondary study?
Matt Cooperberg: Yeah. No, the first study, the earlier one.
Speaker 2: No. We just went to the pathology, so we had limited clinical data. We just wanted to see how the biopsy was processed and what were the outcomes.
Matt Cooperberg: But some of them, so 13 out of 15, almost everybody, but it's a small number, but they all had cancer as opposed to, say, 10%.
Speaker 2: All of them had cancer. We just took the patients who had cancer, and again, that's the issue of VISION group here.
Matt Cooperberg: Other questions from the colleagues?
Speaker 2: Just the premise of this whole conversation, I think something that we would like to understand is that we don't have, except for some places in Australia, good screening imaging for prostate cancer. When we talk about breast cancer diseases, they have mammograms. You see the MRIs. We don't do that for prostate, the] MRIs that spoke about, and I think until we get there, a lot of what you're saying is really going off 12-core biopsies and I think a lot of what we're describing is what we've learned from treated patients, which is our peak.
But if you're doing a 12-core biopsy, I don't think there's a way to really assert that from that 12-core systematic biopsy, and yet still random, that you have certainty to say that this patient is one way versus another. So let's maybe think about ways to actually lay on our screening better so that when we do something like in Australia where we have another type of imaging, then that allows us to call certainty a lesion not cancer, rather than rely on biopsy because we are speaking about treated patient information. That's what we're talking about now. That's my comment.
Matt Cooperberg: The paradigm is changing because it's not only Australia, but the UK as well. You do not get a biopsy without a lesion and teammates in London are saying we really should be treating based on imaging only, regardless of what the biopsy says.
Speaker 2: Yes.
Matt Cooperberg: I think most in the US are not there mentally, and I don't think our radiology consistency is there in this country, but we're heading that way. But I think the question comes back to what we're talking about at the last Q&A, which is do you now add an imaging test as part of what's required to diagnose or not diagnose cancer? Maybe we do. I mean, they're obviously in RCC and kidney. It's all about the IMDC score. Please, yeah?
Andrew Vickers: Can I just make a comment on this building on what you said? Anybody who quotes data from the trust's systematic random biopsy, myself and Declan and Caroline will just go, "We won't listen to you. All of this over grading and missing T3 disease, these high rates of 20-30% that Tony D'Amico spoke about, this is nonsense], you just don't get it." So if you keep on talking about the pre-biopsy era and using data from that, you really won't get anywhere. You've got a test that is so random, it's so bad.
Two or three generations have been affected in terms of your risk classification. I mean it really is just terrible. You shouldn't be using any data from that era because it's just not helpful.
Matt Cooperberg: But in the United States, there's not universal access to MRI.
Andrew Vickers: It doesn't matter. You need to get universal access, but you cannot use data that's bad to carry on just practicing badly. It really is a major, major problem, a major, major point.
Speaker 4: So actually, I want to push back just a little bit because I do kind of agree with you, so I don't want to push back too hard, but I think it's a little unknown in 2022. Somebody who has a negative MRI and was biopsied and found out Gleason is low volume Gleason six, what the radical prostatectomy shows because I don't think those patients should be operated on. So the N must be really low. You know what I mean?
Matt Cooperberg: But the flip side to that is we know if we did RPS on all those, we'd find a little bit of pattern four, some subset. How much does that matter?
Speaker 4: So I would agree with that statement totally, and I would, but I wonder what the percentage, I think it's a little unknown in the MRI era. I'm not saying it's not zero. I don't know what it is now. You know what I mean? I'd be curious. It's an interesting study if somebody knew what that was. Laurie, you know what that is?
Laurence Klotz: Well, just to make it exclude, there's been a whole slew of studies recently that show image negative cancers are much more indolent cancers not only in terms of natural history, but in terms of their genetic profile. There's a couple of studies that show the gene expression, all the invasion and metas associated genes, by and large, their aberrancy is not present in the image negative cancers independent of great.
Speaker 4: Totally agree with that. I'm just saying that I agree it's not 35% anymore, at least in pattern.
Laurence Klotz: Correct.
Speaker 4: And the sevens might actually be less aggressive than the sevens were prior to the MRI era. But I don't know what the percentage is now. I really don't.
Matt Cooperberg: Andrew's got to stand up. But I do want to take Tim's comment per I think he's still having audio issues. Don't forget men who had low-risk disease in the PIVOT trial, which don't forget, was diagnosed by a sextant biopsy by residents in the VA system in the 1990s. So you can't imagine a worst-case scenario, low-risk disease.
Speaker 6: Well, so Matt, I can't say it's a worst-case scenario. They weren't just residents who were doing that.
Speaker 4: No, I'm excited.
Speaker 6: Thank you. I can hear you. My point I think, which is well taken from others, is that it is very unlikely that men diagnosed today have worse prognosis than the low-risk disease back in the PIVOT era. And they had outstanding long-term survival. With watchful waiting we can argue about how many got crossed over and whether it was by VA surgeons, et cetera. But some of the angst about keep looking for more and more and more higher risk disease is very unlikely to change the good prognosis of currently diagnosed low-risk disease.
Speaker 4: Andrew, go ahead.
Speaker 6: Thank you.
Andrew Vickers: Yeah. Can we focus on why we are here? Why we are here is to reduce the morbidity and mortality from prostate cancer. It's not to find prostate cancer. And a lot of the MRI folk, I mean the MRI, I couldn't quite hear who it was who said, "I'm not even going to look at data from the systematic era because they missed all these Gleason sevens and so on." Well, if you actually look at long-term follow-up data from patients in Denmark in the 1990s because Denmark, someone takes an aspirin and then has a heart attack, you can link those two together. It shows almost unless you had some other red flag like a high PSA, if you had a negative sextant biopsy and this was before the MRI era and no systematic follow-up whatsoever, it was just whatever they happened to be doing in Denmark at the time had extremely low long-term mortality. Way less than 1%, and in fact had about half the mortality of the population at large.
So if you had a negative biopsy, even though you had high PSA, you had about half the chance of dying of prostate cancer than somebody who didn't have a high PSA and had never been to the urologist's office. So this isn't just to have a dig, this idea that we've been doing it all wrong and we need to find more cancers. And what about the Gleason Sevens? I mean, somebody like Adam said, "Oh yeah, what if you did a biopsy and it was Gleason six, but it was really Gleason seven." Let's remember 10 years ago. That's why we weren't doing active surveillance. Laurie, you probably remember this, that people wouldn't, they would say, "Laurie, these are really lovely ideas you have, but what about all those men who you think have Gleason six? And they actually have Gleason seven and the data have shown that those men do absolutely fine."
And in fact, if you look at men with Gleason six who are up-graded on biopsy, they have excellent long-term mortality. So I want to stop talking about what if they're Gleason seven and what if it looked like this? And what are we trying to do here is prevent men from suffering and dying from prostate cancer? And can we focus on that as the endpoint a little bit more.
Speaker 2: To that point, you kind of contradict yourself because if you say you're preventing from dying from prostate cancer, then you've done that by treating. I think we're preventing men from getting treated unnecessarily. Right? So I think you contradict yourself by saying that which is significant because you treat them, they don't die from the disease. But we're trying to prevent that.
Speaker 7: You only said half of it. You also meant to say, we want to prevent men from dying from prostate cancer and we want to stop overtreating them.
Andrew Vickers: The real endpoint is my point.
Laurence Klotz: Last couple of points.
Speaker 8: My understanding, if it comes out of this meeting, for example, if we all think that MRI is an important adjunct to understanding prostate cancer, we all have to champion the idea that not only access to MRI, but the quality of MRI has to be standardized at a minimum standard. The region has to have a minimum standard so people cannot get MRIs that completely ruin the clinical picture. And then I think on the second point, just like to CJ's point, when I was talking about ICD 10 codes and stuff, we don't want to be over nuanced there either. They're never going to be a risk-adjusted payment model. It's just if we decide as a group what's not cancer that should never go to treatment, that should be a different code. And it's just not treated. Prepare to love it. So just small steps is good there.
Matt Cooperberg: I'm struck by the difference in the biopsy cancer rate compared to the number of biopsies done. Is there screening going on in some of the countries that have lower biopsy yield rates?
Speaker 2: In the secondary study?
Matt Cooperberg: Yeah. No, the first study, the earlier one.
Speaker 2: No. We just went to the pathology, so we had limited clinical data. We just wanted to see how the biopsy was processed and what were the outcomes.
Matt Cooperberg: But some of them, so 13 out of 15, almost everybody, but it's a small number, but they all had cancer as opposed to, say, 10%.
Speaker 2: All of them had cancer. We just took the patients who had cancer, and again, that's the issue of VISION group here.
Matt Cooperberg: Other questions from the colleagues?
Speaker 2: Just the premise of this whole conversation, I think something that we would like to understand is that we don't have, except for some places in Australia, good screening imaging for prostate cancer. When we talk about breast cancer diseases, they have mammograms. You see the MRIs. We don't do that for prostate, the] MRIs that spoke about, and I think until we get there, a lot of what you're saying is really going off 12-core biopsies and I think a lot of what we're describing is what we've learned from treated patients, which is our peak.
But if you're doing a 12-core biopsy, I don't think there's a way to really assert that from that 12-core systematic biopsy, and yet still random, that you have certainty to say that this patient is one way versus another. So let's maybe think about ways to actually lay on our screening better so that when we do something like in Australia where we have another type of imaging, then that allows us to call certainty a lesion not cancer, rather than rely on biopsy because we are speaking about treated patient information. That's what we're talking about now. That's my comment.
Matt Cooperberg: The paradigm is changing because it's not only Australia, but the UK as well. You do not get a biopsy without a lesion and teammates in London are saying we really should be treating based on imaging only, regardless of what the biopsy says.
Speaker 2: Yes.
Matt Cooperberg: I think most in the US are not there mentally, and I don't think our radiology consistency is there in this country, but we're heading that way. But I think the question comes back to what we're talking about at the last Q&A, which is do you now add an imaging test as part of what's required to diagnose or not diagnose cancer? Maybe we do. I mean, they're obviously in RCC and kidney. It's all about the IMDC score. Please, yeah?
Andrew Vickers: Can I just make a comment on this building on what you said? Anybody who quotes data from the trust's systematic random biopsy, myself and Declan and Caroline will just go, "We won't listen to you. All of this over grading and missing T3 disease, these high rates of 20-30% that Tony D'Amico spoke about, this is nonsense], you just don't get it." So if you keep on talking about the pre-biopsy era and using data from that, you really won't get anywhere. You've got a test that is so random, it's so bad.
Two or three generations have been affected in terms of your risk classification. I mean it really is just terrible. You shouldn't be using any data from that era because it's just not helpful.
Matt Cooperberg: But in the United States, there's not universal access to MRI.
Andrew Vickers: It doesn't matter. You need to get universal access, but you cannot use data that's bad to carry on just practicing badly. It really is a major, major problem, a major, major point.
Speaker 4: So actually, I want to push back just a little bit because I do kind of agree with you, so I don't want to push back too hard, but I think it's a little unknown in 2022. Somebody who has a negative MRI and was biopsied and found out Gleason is low volume Gleason six, what the radical prostatectomy shows because I don't think those patients should be operated on. So the N must be really low. You know what I mean?
Matt Cooperberg: But the flip side to that is we know if we did RPS on all those, we'd find a little bit of pattern four, some subset. How much does that matter?
Speaker 4: So I would agree with that statement totally, and I would, but I wonder what the percentage, I think it's a little unknown in the MRI era. I'm not saying it's not zero. I don't know what it is now. You know what I mean? I'd be curious. It's an interesting study if somebody knew what that was. Laurie, you know what that is?
Laurence Klotz: Well, just to make it exclude, there's been a whole slew of studies recently that show image negative cancers are much more indolent cancers not only in terms of natural history, but in terms of their genetic profile. There's a couple of studies that show the gene expression, all the invasion and metas associated genes, by and large, their aberrancy is not present in the image negative cancers independent of great.
Speaker 4: Totally agree with that. I'm just saying that I agree it's not 35% anymore, at least in pattern.
Laurence Klotz: Correct.
Speaker 4: And the sevens might actually be less aggressive than the sevens were prior to the MRI era. But I don't know what the percentage is now. I really don't.
Matt Cooperberg: Andrew's got to stand up. But I do want to take Tim's comment per I think he's still having audio issues. Don't forget men who had low-risk disease in the PIVOT trial, which don't forget, was diagnosed by a sextant biopsy by residents in the VA system in the 1990s. So you can't imagine a worst-case scenario, low-risk disease.
Speaker 6: Well, so Matt, I can't say it's a worst-case scenario. They weren't just residents who were doing that.
Speaker 4: No, I'm excited.
Speaker 6: Thank you. I can hear you. My point I think, which is well taken from others, is that it is very unlikely that men diagnosed today have worse prognosis than the low-risk disease back in the PIVOT era. And they had outstanding long-term survival. With watchful waiting we can argue about how many got crossed over and whether it was by VA surgeons, et cetera. But some of the angst about keep looking for more and more and more higher risk disease is very unlikely to change the good prognosis of currently diagnosed low-risk disease.
Speaker 4: Andrew, go ahead.
Speaker 6: Thank you.
Andrew Vickers: Yeah. Can we focus on why we are here? Why we are here is to reduce the morbidity and mortality from prostate cancer. It's not to find prostate cancer. And a lot of the MRI folk, I mean the MRI, I couldn't quite hear who it was who said, "I'm not even going to look at data from the systematic era because they missed all these Gleason sevens and so on." Well, if you actually look at long-term follow-up data from patients in Denmark in the 1990s because Denmark, someone takes an aspirin and then has a heart attack, you can link those two together. It shows almost unless you had some other red flag like a high PSA, if you had a negative sextant biopsy and this was before the MRI era and no systematic follow-up whatsoever, it was just whatever they happened to be doing in Denmark at the time had extremely low long-term mortality. Way less than 1%, and in fact had about half the mortality of the population at large.
So if you had a negative biopsy, even though you had high PSA, you had about half the chance of dying of prostate cancer than somebody who didn't have a high PSA and had never been to the urologist's office. So this isn't just to have a dig, this idea that we've been doing it all wrong and we need to find more cancers. And what about the Gleason Sevens? I mean, somebody like Adam said, "Oh yeah, what if you did a biopsy and it was Gleason six, but it was really Gleason seven." Let's remember 10 years ago. That's why we weren't doing active surveillance. Laurie, you probably remember this, that people wouldn't, they would say, "Laurie, these are really lovely ideas you have, but what about all those men who you think have Gleason six? And they actually have Gleason seven and the data have shown that those men do absolutely fine."
And in fact, if you look at men with Gleason six who are up-graded on biopsy, they have excellent long-term mortality. So I want to stop talking about what if they're Gleason seven and what if it looked like this? And what are we trying to do here is prevent men from suffering and dying from prostate cancer? And can we focus on that as the endpoint a little bit more.
Speaker 2: To that point, you kind of contradict yourself because if you say you're preventing from dying from prostate cancer, then you've done that by treating. I think we're preventing men from getting treated unnecessarily. Right? So I think you contradict yourself by saying that which is significant because you treat them, they don't die from the disease. But we're trying to prevent that.
Speaker 7: You only said half of it. You also meant to say, we want to prevent men from dying from prostate cancer and we want to stop overtreating them.
Andrew Vickers: The real endpoint is my point.
Laurence Klotz: Last couple of points.
Speaker 8: My understanding, if it comes out of this meeting, for example, if we all think that MRI is an important adjunct to understanding prostate cancer, we all have to champion the idea that not only access to MRI, but the quality of MRI has to be standardized at a minimum standard. The region has to have a minimum standard so people cannot get MRIs that completely ruin the clinical picture. And then I think on the second point, just like to CJ's point, when I was talking about ICD 10 codes and stuff, we don't want to be over nuanced there either. They're never going to be a risk-adjusted payment model. It's just if we decide as a group what's not cancer that should never go to treatment, that should be a different code. And it's just not treated. Prepare to love it. So just small steps is good there.