68Ga-PSMA PET: A Novel Imaging Technique Prior to Radical Prostatectomy in Men with Intermediate or High Risk Prostate Cancer - Thomas Hope

May 28, 2020

In conversation with Charles Ryan, Thomas Hope details the prospective, multicenter single-arm open-label phase 3 imaging trial evaluating the accuracy of 68Ga-PSMA-11 for pelvic nodal metastasis detection prior to radical prostatectomy or definitive treatment and pelvic lymph node dissection. Between UCLA and UCSF  277 patients went through prospective imaging.  When comparing these patients to nodal pathology 75 patients were positive at the time of pathology and imaging with 68Ga-PSMA-11 detected pelvic nodal metastases in 30 of these patients or a detection rate of 0.40 and a specificity of nodal metastases 0.95 before any primary intervention.  

Biographies:

Thomas Hope, MD, Associate Professor, Department of Radiology and Biomedical Imaging, University of California, San Francisco

Charles J. Ryan, MD, The B.J. Kennedy Chair in Clinical Medical Oncology at the University of Minnesota and Director of the Division of Hematology, Oncology, and Transplantation.


Read the Full Video Transcript

Charles Ryan: Hello, I'm joined by Tom hope, Associate Professor of Radiology and Biomedical Imaging at the University of California, San Francisco, and a former colleague of mine from when I was out there. And Tom has a very exciting presentation at ASCO this year, virtual ASCO, an oral presentation, looking at Gallium-68 PSMA PET, which is really an important new novel imaging technique that will be really important, I think for all of us treating prostate cancer in the years to come. Tom, thanks for joining us and tell us about your study.

Thomas Hope: Great. Thanks for the introduction, Chuck. Yeah, so our study is a culmination of a long, long project that we imaged our first patient back in 2015 of imaging patients with intermediate- to high-risk prostate cancer prior to definitive therapy. The primary endpoint of the study was a correlation with pelvic nodal pathology at the time of prostatectomy. So the general idea here was to try to get an understanding of the sensitivity, specificity, and positive predictive value of PSMA PET for the detection of pelvic nodes and patients prior to undergoing radical prostatectomy.

Charles Ryan: So we've heard a lot in the last few years, at least those of us in oncology about the sensitivity and specificity of PSMA PET in patients who have already had surgery and it's become an important emerging tool for the study of recurrence. How do the results that you're going to present differ from what we know about the postsurgical setup setting?

Thomas Hope: That's a good question, in terms of how the different from the postsurgical setting. So in the postsurgical setting, we typically think of sensitivity, specificity, positive predictive value, more so in the setting of a comparison to the PSA at which the patient presents. So the patients, if you stratify them from a low PSA 0.2 or 0.5 or something to a high PSA in the 10 to 20 range, the sensitivities change obviously dramatically across that spectrum. And a patient at initial presentation is a little different. You're looking at patients based on Gleason score, or maybe a PSA based on the D'Amico classification system is being either intermediate- to high-risk. And you're trying to determine whether or not there are small nodes at that time at the imaging. So it's a little hard to compare across those, but even within the biochemical recurrence setting, there's a broad spectrum of types of patients.

In general, in the BCR setting, particularly if you get to a PSA about two or something, your sensitivity, or I should actually really call it the detection rate. The percent of patients who have disease that are imaged with the PSMA PET gets above 85 to 90%, so it's pretty high. When you're talking about a patient who was undergoing prostatectomy with a PSMA PET, the sensitivities are going to be lower. Because if you think about that, those are the patients with really low PSAs, they haven't even recurred yet. So they're going to be the patients with PSAs that would be 0.05 at the time, post-op, something like that. So it's a little bit of a harder population to image in. So you would anticipate there to be a slightly lower detection rate or sensitivity for nodal metastasis in that population.

Charles Ryan: And is that indeed what you've shown?

Thomas Hope: Yeah. So in our population, we ended up imaging 277 patients between UCSF and the University of California, Los Angeles, UCLA. And in those patients, when we compare to nodal pathology of those 277, 75 ended up being positive at time of pathology. And we found disease in 30% of them preoperatively. So actually a sensitivity of 40%. The specificity was quite high of 95%, so if you saw it, it was pretty much going to be tumor, but the sensitivity was only 40%. So it was much lower than you would have anticipated. And actually some of the literature out there already was somewhat higher. So there was a study out of a Technical University in Munich with the detection sensors, 65 to 67%, whereas this is more in line with some of the more prospective studies that have been done with PyL, DCFPyL on other studies. So there's a wide spectrum of these sensitivities. And we can go into it a little bit, if you want to talk about why there might be a difference between these studies, but it was lower than what we had actually anticipated when we originally designed the project.

Charles Ryan: So this is lower than what you anticipated, but I think let's just dive right to the clinical implications of it. Is this a finding that you would expect would change surgical approaches, surgical outcomes, even though it's lower than you expected? Is it better than what we had before? And is it something that clinicians should be expecting to change their approach?

Thomas Hope: Yeah, so that's a very good question. Is it better than before? Yes. Before, you're talking about a CT scan or a bone scan, which obviously has a much lower detection sensitivity. So I think compared to conventional imaging or whatever we would call it, it outperforms that. And your second question, or embedded in there the second question of how does it impact the management of a patient? And I think that's a much more complicated question. In our study, we actually looked at the patients who were true positives, patients who had disease on pathology at prostatectomy that had it on PSMA PET, and nearly all of those patients, over 85% of them had biochemical recurrence after surgery. Now, that's not that different than the surgical cohorts. If you look at patients who undergo prostatectomy and if they have nodes on a surgery, they also very high percent chance of recurrence. But this is different because you know that now before they go to surgery. You know that those nodes are there before they've had the surgery. It's not like you retrospectively look back and see if they have nodes at time of surgery.

So the question then becomes, well, if you know ahead of time that this patient is likely to have nodes, very likely, and then that patient also is likely to immediately biochemically recur, now what's the best treatment of that patient? I think that there is a question we don't actually know the answer to. You probably have a better insight into that than I do, but I think that's the big question of now that we have a way to see these nodes, and albeit, not all patients, the sensitivity's lower, but let's say something like 40 to 50% of patients will see disease. What do you do with that now that we know that you have this population of patients that's likely to occur after radical prostatectomy?

Charles Ryan: Yeah. I mean, that's a question that I wouldn't say it's beyond the scope of this conversation, but if a person is N1 by PSMA PET, how is our treatment approach going to differ from somebody who is, for example, N1 in the historic studies, in which those patients were treated, for example, with radiation and hormonal therapy, et cetera, et cetera. There are studies that are being planned that are looking at this, of course, but it's beyond the scope of what we can determine at this time. Do you see nodes that are emerging on these scans that are outside of the standard template of the lymph node dissection, such that maybe it's not going to address the question of should a urologist do a lymph node dissection, but should they alter the pattern of their lymph node dissection?

Thomas Hope: Yeah, that's a really good question. So definitely there are nodes, a lot of nodes actually more than you might have anticipated outside of a standard template for a pelvic lymphadenectomy. And that obviously changes the surgeon's approach, but we didn't prospectively acquire that information about the changing of surgical management. But one thing we did do is we looked at what I would call the false positive, the patients who were positive on PSMA PET but were negative on surgery. And we actually found out a lot of those patients ended up being what you would retrospectively consider true positive.

So they immediately biochemically recurred, nearly all of them. And those nodes that were positive were left behind because they're outside of places that are what I would describe as easy to get at. They might be hiding behind a ureter or perirectal nodes that you're not normally used to getting at. And so it shows nodes that are in hard places to get into, oftentimes even when you know they're there, you oftentimes miss them, which is why a lot of work in Europe has gone into trying to create PSMA targeted agents that you can image during the operation to help you find a these ... these nodes are pretty small, some of them can get down to a couple of millimeters, so they can be very, very difficult for a surgeon to remove.

Charles Ryan: So are these cases where you had, if you will, the false positives, these were cases where they were visualized. Was the surgeon not able to get to them or opted out of going for them? And I realize we're in the early phase of how decisions are made based on these findings. Or they couldn't find them as you might allude to, it's one of those three options, I suppose, right?

Thomas Hope: Yeah. I mean, as I said, we didn't prospectively acquire it. I didn't go back and ask our surgeons and say, "Did you purposely leave this node behind?" No. Obviously the surgeons went and tried their best to get them. Presumably, they were in places that were difficult to get at. And when you look at some of these, they're very hard to find some of these nodes.

Charles Ryan: Right. But the extension could be that more extended dissections that happen in response to PSMA PET could lead to a greater number of positive nodes and improve the clinical outcomes, but also with the risk of some increased morbidity, perhaps.

Thomas Hope: I think the other way to flip that a little bit is we always talk about the tip of the iceberg thing, right? Is a single PSMA positive node in pre-prostatectomy patient mean that he has one positive node and if you take that out, you cure him? Or is it indicating he probably has 50 little nodes and you're only seeing the one? And I think that's probably is a little bit more along the tip of the iceberg thing. Which is, although you might take out the one positive node, you're probably leaving behind more than you know. And so that's where we don't really know how ... I mean, there are clearly some patients where if you take out the node, you cure them, but there are many, many ... and that's why I pointed out that in the true positive patients, nearly all of them recurred. Even if you took out the ones that were positive on PSMA PET.

Charles Ryan: Yeah. And I'm still fascinated by the cases we see of people who have PSMA PETs who already had experienced a serologic relapse, you do a PSMA patent and it's negative and suggests that they either have the tiny nodes that you can't see, or they have a systemic burden of disease. I like to refer to it sometimes as a liquid form of the disease, and it's not something that's packed into a node. So moving forward, most sites in the country do not have PSMA PET available to them, Axumin® PET is widely used and other imaging modalities. Where are we in the evolution of this technology and how quickly will it be available and should it be available nationwide, and should it replace Axumin® PET?

Thomas Hope: So should it be available? Yes, it definitely should. So where are we in the process of getting it available? So we submitted an NDA in collaboration with UCLA and our producer dates this fall. So I'm hoping not that far from now, it'll get FDA approved. Now, this is a little bit unusual, very rarely are there academically sponsored NDAs and never before there have been a two center academic NDA before. But what that will mean is upon approval, it's actually unfortunately only available at our two institutions. We're not drug manufacturers, we're not shipping it, but we also aren't making it proprietary. We didn't pay to register it or anything, which means that immediately after approval, any other site can submit an ANDA or obtain local approval up on top of ours, which also includes radiopharmaceutical distributors who can do that and kit manufacturers. So it'll take some time for all this paperwork to get through, but at some point in time, maybe a year to two years afterward, then everywhere in the country we'll have much more access to this.

So obviously our intention at UCSF and at UCLA is not to provide PSMA PET to the world. Which also sort of intones, it's also important to get other agents, as I'm sure you're aware at this same conference immediately proceeding me is a presentation on DCFPyL, which is a foreign native PSMA target agent, which is a very good imaging agent as well, and works very well. And that's being developed by a company, Progenics, and hopefully that will obtain approval someday. I don't think they've yet submitted an NDA, but hopefully soon they will get their NDA submitted because that drug is more easily distributed than Gallium. Gallium has a much shorter half-life, only 68 minutes, and they can't be made as in as much volume or activity, so you can't do a single-center and distribute it out in a region like you can with a fluorinated compound.

Charles Ryan: Yeah. Well, congratulations on all this work and it's really, as I think you went through it quickly, but it is quite a feat for an academic center, two academic centers to basically put an I&D together and get a drug approved. So congratulations on that and we'll look forward to the PDUFA date. And you said it's in September?

Thomas Hope: I hope so. You never know, though, the issue right now is the FDA can't inspect you because of COVID. So they can't physically do a manufacturing inspection. So without a manufacturing inspection, we might be in some hot water, but other than that, we're looking good.

Charles Ryan: Excellent, excellent. Well, I'm sorry that I won't be able to see you in person at ASCO this year, but I'm looking forward to the presentation and congratulations again, always a pleasure.

Thomas Hope: Well, thank you, Dr. Ryan, have a good afternoon.

Charles Ryan: Thanks.