Activity Transcript

Clare Tempany, MB, BCh, BAO, FACR, FISMRM: Hello, everybody. My name is Clare Tempany. I'm the Ferenc Jolesz Professor of Radiology at Harvard Medical School, and the vice chair of Radiology Research at the Brigham Women's Hospital here in Boston. Welcome to this program titled ''Multiparametric MRI for Active Surveillance of Prostate Cancer.''

Joining me today are Masoom Haider, who is a professor of radiology and senior scientist at the University of Toronto, director of MRI research at Sinai Health System, and Caroline Moore, who is professor and head of urology and honorary consultant urologist at the University College of London. We are going to spend some time today talking about the role of magnetic resonance imaging (MRI) in the selection and management of men with prostate cancer using active surveillance (AS)

Caroline, we're going to start with you, and just ask a few questions to get the conversation going. Can you help us all understand what is exactly meant by the term active surveillance, please?

Caroline M. Moore, MD: Sure. So it's for men who've got a biopsy-proven diagnosis of prostate cancer. So it's not for those men who are just having their prostate-specific antigen (PSA) monitored. So they've been diagnosed with prostate cancer, but it's small and not particularly aggressive at the moment, so it can be safely monitored. And importantly, they need to be men who would be fit for radical treatment if they needed it, but who want to either avoid or defer that treatment for a period of time. And some men on surveillance won't ever need treatment, but some men go into surveillance, knowing that they need treatment, but they don't need it just now. So it's a not yet, not a never.

Dr Tempany: You've studied the natural history of men with untreated prostate cancer. And tell us a little bit about what you might have learned, or the takeaways from that.

Dr Moore: Sure. So there's a real difference depending on how men have been diagnosed. So in what I sometimes call the olden days of doing a digital rectal examination and a PSA blood test to determine who gets a biopsy, we would find a lot of low-grade cancer that didn't need treating. And many of those men might never reach the threshold when they're needed to have a prostate cancer treated. In a more modern setting, where we often, particularly in the United Kingdom (UK), use MRI to decide who needs biopsies, we often don't diagnose those men. And we find that men with magnetic resonance (MR)-visible disease, there's about a 1 in 3 chance of things changing in the first 3 years of surveillance, where the MR lesion grows, might have another biopsy, and then move into treatment, either radical treatment with surgery or radiotherapy or perhaps, and in our center, with focal therapy.

Dr Tempany: And there's an old term, and I hope is getting less used, but the so-called watchful waiting terminology. People kind of confuse those. They're very different, right?

Dr Moore: That's right. Yeah. So with watchful waiting, the intention is for palliative treatment if it's needed. So typically, perhaps a man in his eighties, or even nineties, who's not very fit, and he would have hormone treatment, so not curative treatment, but treatment to manage the cancer, if he notices some symptoms, and that might be bone pain, or it might be trouble peeing.

Active surveillance is much more active than that, and it's directed at those men who are fit enough to have treatment, but just don't need it yet. And you're right. People often get mixed up, both patients and clinicians, between the 2 and what it might mean and how it's done.

Dr Tempany: And just to get a little specific on the patient selection side, there's quite a bit of debate about the so-called Gleason grading, or the grade grouping, how much of a 3, how much of a 4. So for those who are sort of informed on this in the audience, that is based, of course, on the pathology. Can you just help us understand where we are today with that? Is it a 3+4, or 4+3?

Dr Moore:. So the grade group 1, or Gleason 4+3, there's a lot of consensus that those men are very suitable for active surveillance. And in the UK, we're so passionate about that, that if somebody in the UK takes out the prostate of somebody with only 3+3, the National Prostate Cancer audit write to them to ask why. So they'll need to explain. And there may be reasons, but it's really not done so often.

Gleason 3+4, or grade group 2, there's much more debate. We very often have people with smaller amounts of Gleason 3+4 that might be visible or not visible on MRI included on surveillance. But in some programs, if you reached the threshold of Gleason 3+4, you are recommended to have treatment. So the range of people that get invited or offered surveillance in different places can vary widely. But certainly, in the UK, we offer men, some men with Gleason 3+4, so more of the pattern 3 than of the pattern 4, active surveillance.

Dr Tempany: And Masoom, can I ask your input here on the approach in Canada and the United States (US), perhaps, in that specific scenario?

Masoom Haider, MD:. So I think the things are viewed on a spectrum. So you have patients with grade group 1, minimal amounts of grade group 2, small volume, maybe larger volume grade group 2 cancers, and then getting into grade group 3 cancers. And depending on the patient's risk tolerance, the age of the patient, a whole variety of factors, actually, that includes, I would say, definitely consultation with the patient, decisions are then made on what to do. I do think that the grade group 3 group is probably very clear that something needs to be done, and that spectrum sits in the low intermediate-risk category is where I think there’s the most controversy.

Dr Tempany: And I think that it’ll be interesting when we’ll shift the conversation out a little bit to Masoom and talk more about the MRI and how it can help as well here. And perhaps you can start off, Masoom, by telling us a little bit about how MRI is currently used in your practice for men considering active surveillance, please.

Dr Haider: So as you noted, we are up in Canada, so I would say that our practice is maybe reflective of a spectrum of viewpoints of the use of MR and active surveillance that include the American and European and British sort of extremes. I think the commonest indication we see is for a worsening risk profile and patients already on active surveillance. So there’s an increased risk assessed, typically, by PSA or PSA kinetics, and the patient gets sent for an MR to see what’s going on with the cancer.

I think in general, it’s a preference to do MR also before the regular interval biopsies that are often mandated in active surveillance. So, in a typical active surveillance protocol, you want to do a systematic biopsy every 2 to 3 years, at least in the classic active surveillance protocols. And that may be even if the PSA is stable. And to avoid the biopsy, again, we may get an MR done. If it's low risk, the patient may not need a biopsy and avoid that uncomfortable procedure.

The follow-up time, actually, how often do you do the MR varies. Again, from urologist to urologist, we are seeing niche cases, for example, BRCA cases, where we're recommending closer follow up in patients. Some urologists send a patient every year, some every 2 years, some every 3 years, so there's variation there. But in general, they are sending them regularly.

And then there's this interesting category that's emerging of the patient, either because of their own aversion to biopsy, or because of a medical condition, anticoagulants or something, just never get a biopsy. And not really active surveillance, because we don't have a Gleason score on them. But we're getting a lot of MRs, and they're being followed, and it's almost like active surveillance without the needle. And that's an interesting group of patients that we're seeing. So that's where we are.

Dr Tempany: You've had the pleasure of working with Laurie Klotz for many years, who's probably, I don't know if I should call him the father or the grandfather of active surveillance, he'd probably prefer father, I'm sure. But it seems like it's been a very long journey, and he's really regarded highly. And you both together have worked so closely. Thank you for all your contributions. I'd love to hear a little bit about the latest analysis of the ASSIST trial, or sort of how you've digested that trial. I know there are different parts to it, the early data, and then the 2-year follow-up data. Maybe you can sort of walk us through that. I that was one of the very first that sort of formally introduced MRI into an AS protocol.

Dr Haider: Yeah. No, the results are actually quite interesting. So the ASSIST trial is a randomized controlled trial. It involved 3 centers in Ontario, and patients were randomized into an MR arm and non-MR arm prior to confirmatory biopsy on active surveillance. So this is sort of at the entrance point of active surveillance patients will typically get a confirmatory biopsy in the first year. So we said, "Well, let's do an MR in half the patients before and see if it helps." Now, the biopsy strategies were to do systematic in the non-MR arm and systematic combined with targeted. So we were very conservative in the MR arm. So if we saw a target, they would get the systematic, plus additional targets. And the primary endpoint was upgrading to grade group greater than or equal to 2.

And interestingly, the study found no difference in the 2 groups in terms of grade group upgrading, which was a real surprise for us, a disappointment, in fact. And it goes to a few issues. I think one of them is that active surveillance is probably one of the toughest groups to find good evidence for MR on entrance, because you're already selecting patients who've had a systematic biopsy and are low risk, based on traditional risk stratification. So the more conservative you are, so let's say an American sort of Hopkins approach in the start of active surveillance where they're very, very conservative in selection of patients for active surveillance, the utility of MR and detecting additional targets may be quite small. And while if you're more open to intermediate-risk profiles, it could be very powerful. So I think this is one thing is that the utility of MR in entrance certainly depend on the risk profile.

Now, interestingly, in our group, I'd say that we're a pretty standard risk profile, PSAs were around 6, the mean PSAs, so not overly conservative, I would say. But I think that brings up the second point is that there were major differences in the performance of the targeted biopsies by center. So when we looked at the 3 centers, there was a 33% target yield at 1 site, which was the best with the majority of patients, but only 8% to 10% at the other 2 sites. So that's almost a 3-fold difference in target yield. And it goes to this issue of quality control in both the MR... Now, the MR was done and interpreted by a single expert radiologist. So I think the issue here was actually the targeted biopsy equipment and how it was used. And it was brand new. We were using brand new fusion equipment. And it goes to how important this pair is together.

So that's a bit of a long response on this one, Clare, but I think it's important to understand, and I could maybe discuss for a minute the 2-year follow-up, which is I think where the real sort of impact shines through. In this trial, we did design a formal 2-year follow-up. So patients in both arms, if they were still on active surveillance, did get a biopsy at 2 years, there was about 130 in each arm, and the active surveillance failure rate was the endpoint there. And in the MR arm, the failure rate was 19%, and it was 35% in the non-MR arm. So MR really did reduce significantly the active surveillance failure rate. And on top of that, at the 2-year biopsy, there was only 10% clinically significant cancer in the MR arm vs 23% in the non-MR arm.

So MR as an intervention, if you look 2 years down the road, does reduce the cancer stem cell (CSC) detection rate and the active surveillance failure rate, even with this variation across the centers. And again, we saw differences in the active surveillance failure rate across the centers. So 4.2% in 1 center, so very low failure in the MR arm, 17% in another, and 27% in another. So the utility of MR as with many things where there's a human interaction, human touchpoint really depends on, I think, quality experience, expertise, as it does in many things in medicine. But this study did show, I think, a real role for MR on entrance, especially when you follow patients longitudinally.

Dr Tempany: Yeah. And I think it was heartening to see the 2-year data, for sure.

And Caroline, would love your opinion on some of the issues that Masoom has hit on here, and obviously, leading us forward into some of the more recent trials. But specifically sort of addressing perhaps a little the biopsy techniques; we've learned a lot about biopsy techniques, I think, since we all started doing a MR-guided biopsies back, whatever it is, 10, 15, 20 years ago now. So perhaps you could shed some light on some of those issues and others, please.

Dr Moore: Yeah, sure. So I think in our surveillance cohort, men would get into it if they've got a concordant MR and biopsy. And we have a big tendency not to biopsy our negative MR men. So we already have a sort of high-risk cohort than some places. In terms of the actual technique, we've been doing transperineal biopsy for probably 15 years now, and we do a lot of cognitive fusion, and sometimes software fusion, depending on which room, which machine, that kind of a thing.

And I think Peter Pinto did a very nice analysis, showing that even with software fusion, experience really makes a difference. So if you do have a new center using fusion that hasn't done it before with a new set of software, then you can expect that there might be some teething problems. And if the things that you're targeting are lower risk, then your yield will be less as well. So that kind of what will you find on surveillance really depends on the cohort of men that you've got going into surveillance, which slightly brings me onto the whole concept of active surveillance failure.

So I would say that having treatment on surveillance isn't necessarily failure. So some of those men knew they were coming into surveillance knowing they'd need treatment, so they get to that point, but it's not really a failure. And then the other is that we also have the concept of graduating from active surveillance. So you've done really well, you're 75, and you still didn't need radical treatment, this is brilliant. We can now relax things a bit and bother you less with the MRIs, that sort of a concept.

Dr Tempany: I think everything is about experience and training, particularly the biopsy. I think the cognitive for radiologists, we do that all the time. And of course, many times now, the biopsies are being done by urologists, or radiologists, and combinations.

Just quickly before we leave it, the transperineal vs the transrectal, I think that we're all worried about infection. Is that becoming a bit more of a trend? I mean, I think the Europeans are adopting that more quickly than we are.

Dr Moore: Yeah. So I would say certainly in the UK, and hastened, if anything, by coronavirus disease (COVID), local anesthetic transperineal is a really common way for biopsies to be done. I think in terms of accuracy, when we started MR-targeted biopsy, our radiologists did it, so Clare Allen and the team. And they did transrectal targeting, and you can still get the same accuracy, apart from in the big prostate where you can't get anteriorly so well. But I think in terms of infection risk, transperineal is the way to go. It just massively reduces the risk of infection.

Dr Tempany: Thank you for that. And maybe I can keep, if you don't mind, change subject a little bit, but for us to understand what you've been doing in some of the bigger groups and the consensus panels, the GAP3, the Movember, bring us up to date on sort... I know you've surveyed, I've been part of this best current practices, things like that. If you could help us understand where we are today?

Dr Moore: So Movember helped us to convene an international panel, not only of healthcare professionals, but of patients, looking to do 2 things. So firstly, to define best current practice, and then secondly, to look at what we should be researching next in active surveillance. So it was quite a detailed process, firstly, of selecting people representing the different specialties in different countries to come together and work out what are we doing and what should we be doing right now.

And the conclusions from that were quite interesting. So for some of us, not a lot of change. But for others, it would be changing clinical practice. So to omit the digital rectal examination, if you're using MRI routinely, sounds really obvious to many people using MRI routinely, but it's still in the guidelines that you should do a digital rectal examination. Also, the men with a stable MRI and stable PSA kinetics don't need to have routine biopsies necessarily. So you can consider omitting the routine biopsies. And again, that's not in other guidelines yet.

We talked about having a change in PSA or digital rectal examination shouldn't lead to immediate treatment, but it should lead to an MRI potentially with another biopsy to assess their risk. And a change in active treatment shouldn't just be based on clinical parameters, but really needed to incorporate a discussion with the patient so that it was a joint decision. And then for those men who were sort of on their MRI and biopsy who might be suitable for surveillance but are anxious, the important thing is to put in strategies to help with the anxiety rather than to use radical radiotherapy or radical prostatectomy to lessen that anxiety. So that was the best current practices.

And then when we looked at what to do in the future, what were the most important research priorities? The one that came out really strongly was looking at dynamic risk-adapted active surveillance. So currently, many centers and many countries have a protocol that's the same for the wide range of risk and ages of men who go into the surveillance program. And Movember have now put out a funding call and selected people to join in a new project, looking at developing that dynamic risk-based active surveillance, with the idea that men at lowest risk would have fewest investigations and those men at higher risk of progression would be monitored more closely so that you don't miss the opportunity for cure.

Dr Tempany: So just to make sure I got it correctly, the dynamic piece, what goes into determining the dynamism or the kinetics there?

Dr Moore: Yeah. Yeah, really good, really good question. So essentially, we did a particular survey on that, and the most important things are the Gleason grade and the MRI findings, followed by PSA, PSA density, potentially cancer core length, and digital rectal examination was always the least important of those, both from the patients and from the healthcare professional point of view. So really focusing on Gleason grade and MRI findings.

Dr Tempany: And that's a perfect segue, because I want to shift us back to talking a little more about the MRI, Masoom. I think we're all pretty solid on what is Prostate Imaging Reporting and Data System (PI-RADS), and what are the minimal technical standards. But what are the current sort of diagnostic challenges that we have, obviously, in both acquiring, interpreting and understanding the MRI as its component contributes to this dynamism, if I can coin the phrase, at this point for risk analysis?

Dr Haider: The understanding is essential that the relationship between histopathology and grade group of various cancer foci and the prostate, and what we see on MR imaging is not 1:1. It's not like renal cell carcinoma, or something, where there's a ball and there's a tumor and it all matches up pretty well. There is this sort of imaging phenotype heterogeneity. And we know on pathology, there's a lot of heterogeneity. So this multifocal and heterogeneity is an issue that creates some noise between the actionable grade group upgrading and how you sample and what you see on MR.

So for example, secondary foci, smaller grade group 2 foci, we can miss a lot of them. I mean, depending on the trial you look at, or the data you look at, we can miss 30%, 40% of secondary grade group 2 foci. So that's sort of the worst way of looking at it. Obviously, the primary focus is often the driving focus. But we could use some help there, especially as focal therapy ascends in importance, and temporizing measures become treatment options for patients, temporizing therapeutic options.

And I think the second one goes to this dynamic risk profiling is we really need to understand the relationship between histologic change or grade group upgrading and MR change. And this one is really vital. So if something looks the same on MR, is that okay if the PSA is rising, or is it safe to wait? And there's mixed data on this, but I would say the data is a little... It's raising a little, some small red flags right now, where we say, "Okay. Yes, a 50-year-old patient can have grade group upgrading on their biopsy, a repeat biopsy, even though their MR findings are stable." And we don't know if that's a sampling issue, or if we're on safe ground, I think this is an issue that needs to be addressed. And I think a lot of the groundwork is being laid by Caroline's work and others in the field in defining systematic ways of assessing this.

So those are examples of some of the issues that we're facing. But the core utility of MR has not changed. I think it's clearly a key important part of the active surveillance paradigm.

Clare Dr Tempany: Maybe this is a good segue to Caroline. Just a little, the Prostate Cancer Radiological Estimation of Change in Sequential Evaluation (PRECISE) score system that we are not using yet over here, perhaps you could give us some insight into how that was established, and then how you're using it.

Dr Moore: Sure. So that was set up as a consensus meeting through the European School of Oncology. And it was really to recognize the fact that much of the work in defining what we should report on MRI had been focused on what you say at the point of diagnosis, how likely is it that there's cancer, and how bad might the cancer be, how big it is. For men on active surveillance, the question is different. We know that they have a cancer, and the question really is, "Has it changed?"

So the PRECISE score is a sort of sliding scale, a Likert score 1 to 5 where 3 is stable, 5 is stage progression, and 4 is progression of a lesion, and 1 and 2 can be regression. So we know that sometimes you get a little bit of cancer in an inflammatory lesion on MRI, and that inflammatory lesion might settle down, but you've still got the cancer.

The other key things were to report each lesion separately, and to... We're working on a second version of PRECISE, or rather thinking about working on a second version of PRECISE, where we will go into more detail. And it may be that those grades change at some point, because there's not too much difference between a 1 and a 2 at the moment, and we don't differentiate very well between a new lesion appearing and an established lesion getting bigger. And it might be that we should be taking more note of that. So we'll see what the future holds, but it's probably a year or 2 away.

Dr Tempany: Yeah. And that's something that I think Masoom and I and the PI-RADS, and with your help as well, and the PI-RADS Committee have wrestled a little bit with how we quantify MRI and how we use, say, the apparent diffusion coefficient (ADC) value, or the pharmacokinetics of dynamic contrast-enhanced (DCE). And I think these are things that we're looking for to the future, really, to try to be able to do that. And obviously, I think this also calls to the vendors for standardization and stability of the systems that we use to acquire these data points before we can convert to quantification. Many of my fellows still ask, "Why aren't we using ADC in all our reporting?" And I'm like, "We just are unable to really rely on it very well." And Masoom, you may want to chime in here and sort of give me any updates or thoughts that you're thinking on that before we wrap up, but I'd love to hear a little bit more about quantification.

Dr Haider: Well, I think you said it well, Clare. I mean, the manufacturers and the larger community that is working on multiple different vendor platforms has not really come together to standardize this measurement. So it really requires a broad-scale effort to do this. I think there are efforts underway to create methodologies to do this, but I would say we're still in the technical validation stage of this type of research. So I do think it's a tremendous opportunity to really get to real quantification, which is what we get from fluidics. So when you do PSA, or 4K score, or Genomic score, you're sort of taking out this sort of interpretation in the assay itself, the assay itself is sort of agnostic to that. And I think if we can do that with MR, that would be a real breakthrough. So I'm optimistic that we'll get there, stay positive, but there's a lot of work still to be done.

Dr Tempany: There really is. And I think that's really important. And it's a good place to start to wrap up and give you some key take home points. But certainly, I think artificial intelligence is beyond the scope of our conversation today, but clearly is in the mix there in the quantification and standardization, without any doubt, and hopefully will help us a lot.

This has been a really fun conversation for me, and I hope it's been educational for our audience. I just want to summarize the key takeaway points that I've got from this are that we are all in agreement that MRI plays a significant role in selection and management of men on active surveillance protocols. There are issues that we as a community and in the academic research world will wrestle with, but training, quality, and measurement of interval change are 3 big buckets that we're really focused on right now. And I think, as you can see from our panel today, this has to always be a multidisciplinary team effort. Our urology colleagues, our radiology colleagues, pathology colleagues, perhaps as well, even a very, very important key component to this, and would advise everyone in their institutions to continue to work as teams.

And I would like to conclude by thanking you both very much indeed for this great conversation. And I would like to thank you, the audience, for participating in this activity, and please continue on to answer the questions that follow and complete the evaluation. Thank you very much, and goodbye.

This transcript has not been copyedited.