The Accidental Journey of Shepherding PSMA PET Imaging in Prostate Cancer to the Clinic in the United States - Thomas Hope

January 15, 2021

Thomas Hope, MD highlights how he came to study somatostatin receptor-based molecular Imaging for neuroendocrine tumors (NET).  He takes us back to 2013, to the World Theranostics Conference at Hopkins, focused on neuroendocrine tumors because at that time there were no prostate theranostics.  For this work, they put together a gallium dotatate program, which required the same synthesis module required for PSMA, the same generators needed for Gallium PSMA, and the same chemistry.  This is the start of the journey that leads to the FDA approval of 68Ga-PSMA-11 PET, on December 1, 2020.  This new “PSMA PET” scan can detect prostate cancer metastases much earlier when they are much smaller, which leads to improved treatment decision making in patients with prostate cancer.  Dr. Hope shares critical support that will revolutionize the staging of prostate cancer.  


Tom Hope, MD, Associate Professor, Director of Molecular Therapy, Radiology and Biomedical Imaging, 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

Read the Full Video Transcript

Matthew Cooperberg: Hi, I'm Matt Cooperberg, and welcome to another installment of our interview series on the UroToday Localized Prostate Cancer Center of Excellence. Today, it is a great pleasure to be joined by my friend and colleague Tom Hope, who is an Associate Professor of Radiology at UCSF, and here at San Francisco VA, where he is actually sitting just a couple of doors down.

Tom really deserves a huge amount of credit for really bringing PSMA-PET to what is almost real-time community-based clinical practice in the United States. We're going to have a conversation today about what that journey has been like and the future of PSMA-PET, and other molecular imaging for prostate cancer. So Tom, welcome.

Thomas Hope: Great. Thank you for the introduction, Matt, looking forward to it.

Matthew Cooperberg: So just to kick-off, as I just mentioned, that you really have been instrumental in this interesting and convoluted journey that PSMA-PET has had in the US. So, start us off. Just tell us a little bit about what that journey has been for you. How did this become your personal quest, first of all? How has that quest been, progress, and what's the journey been like?

Thomas Hope: Yeah. Well, first of all, I can certainly not take the most credit or any significant amount of credit. It takes a village to get any of this done. And not only just at UCSF, but the number of people at numerous institutions across the country who've been involved in studying, developing, and putting forth PSMA-PET has been pretty enormous. It's really been a remarkable effort from a lot of people.

But my journey here, I would say sort of started in two places. One place is actually you. I don't know if you remember this, but back in 2013, I was a grinning, young faculty member at the VA here in San Francisco. And I came to you and I wanted to do research projects, and I asked you what I should work on. And I was like, "Oh, I want to do..." I don't remember what it was, HIFU and whole-body diffusion.

And you were like, "No, no, no, no, no. What you should do..." And you and Kristen Green said it together, "You should do PSMA-PET." And I was like, "What is PSMA-PET? What are you talking about?". So then I started reading about it and got interested in it because of your and Dr. Green's suggestion. So that's one half. And the other half is, when I first came to UCSF, I got involved in neuroendocrine tumors of all things. And neuroendocrine tumors at that time was when somatostatin receptor PET had just begun. That uses a compound, not similar, but uses the same radionuclide, gallium. And so we put together a gallium dotatate program, which required the same synthesis module you needed for PSMA, the same generators you needed for gallium PSMA, the same chemistry, et cetera.

So by building that up for the neuroendocrine tumor side of things, accidentally made us prepare to do gallium PSMA imaging at UCSF. And so from there, then we just sort of rolled over from neuroendocrine tumor to prostate cancer and moved forward. It's been surprisingly, consistently productive. Every year, it kept moving forward. Usually, these things fall into big holes, but this one's actually been very successful.

Matthew Cooperberg: And what about your training with the FDA? The effort to get this to prime time for us?

Thomas Hope: Yeah. So I would say the FDA has been the most supportive group of people to academia that I could ever ask them to be. This actually again starts in neuroendocrine tumors. Back in 2013, we had a meeting, it was the World Theranostics Conference at Hopkins, and it was really actually focused on neuroendocrine tumors because at that time there was no prostate theranostics.

And the organizers invited the Director of CDER, Lou Marzella, to give a talk. And one of the patient advocates of neuroendocrine tumor, this guy, Josh Mailman, who's one of the most amazing patient advocates ever, brought 60 neuroendocrine tumor patients. He had them walk up one after another and ask, "Why do we not have somatostatin receptor PET in America?" While they have it just like you think elsewhere. "Why do we not have it?".

And Dr. Marzella turned around and said, "Send me an NDA. I cannot approve a drug unless you give me something to review. My role as the FDA is to review your drugs." He would say, "We have not rejected anything." And he pretty much said, "All you have to do is send me an NDA and we'll approve it." That's not always true. Obviously, you need data and you need to have chemistry and manufacturing all that put together, but that definitely supported the ability for people to do this.

Since then, there have been two somatostatin receptor PET drugs approved by the FDA. And when we've met with them, they have always obviously required data, they've required trials, they've required the same level of thoroughness that you would need from a company in order to succeed. But they've also been incredibly supportive in trying to help us. We've met with them a number of times to provide protocols, our initial staging, and biochemical recurrence protocols, we discussed and developed with the FDA with continuous input. And those protocols have actually gone on.

And if you look at what companies are running now, they're actually the protocols that we developed with the FDA. And they provided really helpful, positive feedback throughout the process. And I think without Dr. Marzella really telling us, "This is something we want you to do. And we think you, someone who doesn't know anything about regulations, can succeed at it," really, really helped us believe that this can be successful.

Matthew Cooperberg: And has it taken longer? I mean, this used to be called the Heidelberg compound, and it seems like we've been reading studies that are first Germany, then Australia, then more and more countries where this is... You hardly get a CAT scan before you get a PSMA these days in Australia and other parts of the world. It feels like we're behind the curve here a little bit. I mean, do you think this is just the regulatory process in the US or is there anything specific about this particular situation that is setting us back?

Thomas Hope: Yeah. Let me disabuse you of the notion that we're behind.

Matthew Cooperberg: Please do.

Thomas Hope: You think Germany, right? Germany, "Oh, they're so far ahead of us in Germany." Germany does not have an approved PSMA-PET radiopharmaceutical. There is no regulatory approval for any of these compounds in Germany. So we, hopefully, at UCSF and UCLA will get PSMA-11 approved within the next two months, probably less than that. (FDA approved in December 2020). And we will be way ahead of Germany at that time with an actual approval.

So not all hospitals in Germany have PSMA-PET available. It's only certain hospitals, and it can only be performed under compassionate use, which is this weird loophole in certain countries and regions within Germany, that these studies can be performed. It's the idea that we're behind in a way that is not true. It's just the difference with PET imaging and radiopharmaceutical therapies is that you can make this in-house.

So if this were like abiraterone, right, you can't make abiraterone and use it at the Technical University of Munich, right? Like it's just not going to happen. You need a pharmaceutical company to manufacture this drug and distribute it and everything. The pharmaceutical firm can control access to their drug. But with PSMA-11 PET, you can just buy this precursor, label it in-house, and anyone can do that.

And so it gets used all over the place and places particularly where regulations are more lax. But to say that we're behind... Take PSMA-11; we finished multiple Phase III trials. We have an NDA under review. And they're nowhere near that in Germany, actually, when it comes to getting insurance reimbursement in Germany and everything. So we're actually soon to be ahead of most places in the world. We're doing just fine here.

Matthew Cooperberg: Tell me about your personal thoughts on that process actually. Because I know you and I had a conversation about what they're now doing, I think it's in Munich, somewhere in Germany, labeling the same targeting compound with technetium for intraoperative localization now. Once they know what the compound is and everybody knows what technetium is, it's okay to mix and match A and B or A and C or Z and X. Is that a better approach or is it more safe and correct to think about technetium PSMA as a completely new pharmaceutical from gallium PSMA, and start from scratch with a new NDA? That's not meant to be a leading question. I'm actually really curious about what you think.

Thomas Hope: It's very complicated. So the way in Germany that it's done, it's done under something called compassionate use, right? So they have a regulation that allows the use of these radiopharmaceuticals to be used to image and treat patients when they have no other alternative available. So there's these requirements that you have to do now. That in general is pretty reasonable.

Once PSMA-11 is clearly beneficial, I think compassionate use is a reasonable approach. The US sort of has this through the expanded access program. It's just a lot more lifting to do, and you actually have to consent patients, et cetera, and follow all the regulatory guidelines in the US which becomes somewhat burdensome. I think where, in Europe, it gets a little misused, is when things haven't even been in humans before. And the first in human of a compound is done under compassionate use.

I have some ethical concerns, particularly when it comes to radiopharmaceutical therapies which have significant toxicities. I think you actually probably should be required to go through a full research protocol with ethical review and consent forms and everything. Once those therapies are actually proven to be effective, then you can start doing it under compassionate use, but this is where the line is very gray in other countries is, what is research and what's compassionate use?

So I think the US way is great and works very well. It does not work well when there's no funding. The German environment and parts of Europe environment worked very well when no one was interested in this and no one was investing any money in this. And it allowed this stuff to sort of develop and multiple compounds were being evaluated.

And that's how we got PSMA-11, PSA-1007, PSMA-617, PSMA-610. All of these compounds came out because of this environment with very low regulatory walls. Now that there are many companies jumping into this, there are compounds all over the place and we actually don't have that issue anymore. And in the US, the US requirements are what lead to all of these great trials and data. Now there are multiple Phase III trials being run in the United States with five different compounds. It's really that regulatory structure that gives you data that actually tells you how to use these agents. Essentially, there are pluses and minuses of both. I couldn't argue that one was better than the other. I think together they work really well, actually.

Matthew Cooperberg: Give us a brief highlights reel of the study you had out with Jeremie Calais, you mentioned, at UCLA. What in your mind are the major advances in the last 12 months here in terms of findings?

Thomas Hope: We've done a lot of collaboration with UCLA which has been, in all honesty, maybe the best part of my academic career, is the collaboration with the group down there. They've been incredibly supportive and collaborative, which has been wonderful. A lot of our work has been recapitulating what people have found elsewhere. So our biochemical recurrence trials, our initial staging trials with PSMA-PET.

Although they're important for getting the drug approved and they form the basis of our new drug application, which hopefully, as I said, will get approved very soon, the data themselves aren't particularly insightful or different from what other people have published previously. The UCLA group has done, I would say, two things, in addition, that would be more important. They did a very nice head-to-head trial of fluciclovine to PSMA in the same patients, which I think was... It's expensive to do and they just paid for it out of pocket, back-to-back patients within one week of each other. Without question, PSMA out-performs fluciclovine quite dramatically. It had a two-fold increase in detection rate, more lesions were seen, et cetera.

And the second study that they have run is a therapeutic study where they randomized patients after radical prostatectomy who have a rising PSA, of a PSA less than two, who would standardly get salvage radiation therapy, to either receiving a PSMA-PET or not. And then seeing obviously A, as the radiation therapy plans change, and then B, more importantly, do you have an improved biochemical recurrence-free survival? And I think those are the types of studies that we really need to see.

Matthew Cooperberg: Do you think there's any role for fluciclovine going forward once PSMA is approved? Are there situations where you think it has better sensitivity or specificity?

Thomas Hope: Not in prostate cancer, no. The thing that fluciclovine potentially could have value in is showing you metabolic disease. So it can show you a metabolic response, if it goes up or down after treatment, whereas PSMA is impacted by androgen receptor activity, et cetera. So it might not be as good as a response biomarker, but if you're going to do that, you might as well just use FDG, which costs a hundred bucks a dose and not $3,000. You're not gaining anything using the fluciclovine.

The fluciclovine might have roles in other diseases as we start using it. So for example, it might work really well in GBM and brain tumors, or in lobular breast cancer or things like that. But I think in prostate cancer, once PSMA becomes widely available, sadly for the company that owns that, it will not be a very useful drug.

Matthew Cooperberg: Well, they have other things in the pipeline. Speaking of which, what are your thoughts on, you mentioned the growing set of PSMA tracers that are out there in various stages of development. Where does PSMA-11 sit in the spectrum? Are there other tracers that you're even more excited about? Cost aside and all that sort of thing.

Thomas Hope: These last couple of months have been sort of exciting. There was, there were two NDAs outside of our NDA submitted. So a DCFPyL submitted their NDA a month ago, which is really exciting. That compound is a wonderful compound. And the thing about PyL is it's fluorinated, so you can make it and ship it all over the place, and it makes it widely available.

The other thing that's exciting is someone submitted an NDA also on PSMA-11, referencing our literature. And so they submitted a kit NDA, which is fine. That's the whole point of our NDA. Our NDA is non-proprietary academics. So we want people to just submit NDAs on this drug and get it approved. We can't distribute it from USCF. So we need companies like this to make kits to distribute them more widely.

So, a year from now, there will be, in essence, PSMA-11 approved as a kit, hopefully. Our NDA, which will be not widely available, and then DCFPyL. Now, should PSMA-11 is one bet, but on the other, no, I don't really think there is a significant clinical difference between the two. You know, I think the end game is going to be availability. PyL will probably be more easily available than PSMA-11. So it'll take more of the market share, but not because it's particularly a better or worse compound, just because of ease of... If you're a random community hospital, are you going to get a galleon generator? No. You're just going to order from the local radiopharmacy and use the drug, which is great.

These compounds, to me, they're a class in general. There are some that are worse. I haven't seen any real significant data that suggests A is better than B, between PyL or PSMA-11. There's PSMA-1007, rhPSMA-7, PSMA-R2, et cetera. So there's a whole load of these compounds. It'll be fun watching companies try to advertise why theirs are better than the others. But in the end, it's really about reimbursement, cost, availability more so than anything else.

What we really need is therapeutic trials showing us what to do with the information, right? Like how do you actually use your management of the patient, when you see the patient who has a left internal iliac pelvic node initial staging, do they get radical prostatectomy? Should they get radiation? What do you do with ADT? How do we know? Does Decipher® play a role in that decision or? Which information is more important? Et cetera.

Matthew Cooperberg: So that was my next question. As you know, as much as I'm super enthusiastic about PSMA, I keep getting asked to give these con talks. I think I'm doing one for a local retreat, and then maybe at the SUO at the end of this year. And it's not because I don't believe it. You'll remember that one of the first cases at UCSF was a patient of mine who went for PSMA and showed nothing in the pelvis and a node in the neck. And I'm sure I made some snarky remark about false positives, and we biopsied it and it was positive, meaning we didn't really understand prostate cancer biology as well as we do now, knowing that it skips much more commonly than we thought.

But I still wrestle with what we do with the information, as you just said. Admittedly, this is more our problem in urology and radon than it is radiology's problem. But you know the space really well. What would your ideal trial be? What studies do we still need to do going forward to figure out, from your perspective, where the clinical benefit really lies?

Thomas Hope: Yeah. Well, I would say, what is "the study"? There's no "the study".

Matthew Cooperberg: Yeah, sure.

Thomas Hope: The key thing here is getting wide availability of Medicare coverage. Because once you have wide availability Medicare coverage, then you can include PSMA-PET in cooperative group trials, company-sponsored trials, without having to budget $3 million to pay for all these PSMA-PETs. It makes a huge difference once it becomes a standard of care. So that's sort of my first comment.

I always like to throw these sorts of thought experiments up there. You take the CHAARTED trial and you read the CHAARTED trial, and if you have more than three bone mets or four bone mets, then you get chemotherapy. Well, what the hell do you do in setting a PSMA when everyone has three or four bone mets? What is low and high volume disease?

What are you going to do? But you can't go back and rerun the CHAARTED trial. It just becomes sort of a messy field. And I think right now, working through the cooperative groups, there's a number of studies that are starting to come through, GU Steering Committee is starting to be understood. For a while, it was, "Well, PSMA-PET's not yet available, how do you run a trial in a PSMA-PET world when you can't get it?".

And so now, the timelines are such that the PSMA-PET will be available in the lifespan of these trials, and when these trials start to open. So right now is a perfect time to start designing studies that really use PSMA-PET. As I always say, within a couple of years, I don't want to see another bone scan ever done in a patient with prostate cancer, right? There's no point. It's a worthless study. And so very soon we can just sort of drop them doing that and move towards just a nice, simple imaging study to save these patients.

Matthew Cooperberg: Sodium Fluoride PET, no?

Thomas Hope: Don't get me down.

Matthew Cooperberg: But it does raise... I mean, I have seen the scenario not infrequently where a patient gets PSMA scan first, it shows eight lesions. And then it's usually an oncologist actually orders a bone scan after the fact to see how many lesions that were on the bone scan, to figure out if this patient should be considered oligometastatic per CHAARTED, per STAMPEDE, radiation arm.

And it creates these very odd conversations with the patients that I don't know what to do with in practice personally. You take somebody who's high risk, clinically localized by conventional criteria. And now, like you said a few minutes ago, now we find one rib spot on a PSMA or even a couple of spots. Do you say you have systematic disease, systemic disease, incurable, just go on hormones? Do we extrapolate from oligometastatic disease or from high risk, quote-unquote M0 trials? I feel like we're launching into this unknown that we don't really know what to do with this data. It's going to take a while before the trials mature to the point where we will be able to finish the [crosstalk].

Thomas Hope: Yeah. CHAARTED and STAMPEDE, you're never going to get that data again. Right?

Matthew Cooperberg: No.

Thomas Hope: Like you're not going to go run it.

Matthew Cooperberg: Exactly.

Thomas Hope: Someone someday will retrospectively go back and look at bone scans, in a cohort of patients who had bone scans and PSMA tests within a month, to see what your PSMA-PET had to look like in order to get a bone scan that looked like this or something like that. But in general, I feel like the field is moving towards, you have really high volume, high, high volume, which means like gross mets, not just four lesions and needing docetaxel.

When it's smaller volume, positive bone scan, then you might be doing the abiraterone, when it's much lower down that you're high risk, then you might. It's sort of playing itself out in its own way, but I agree, nothing's ever perfect and it's going to take time to convert me, but bone scan has been around for 50 years, 40 years, right?

Matthew Cooperberg: With no trial.

Thomas Hope: So it's going to take a while to recreate some of that data.

Matthew Cooperberg: Well, and hopefully, some of those decisions become guided by molecular as well as by imaging or by tissue or liquid diagnostics.

Thomas Hope: Just imaging.

Matthew Cooperberg: What are your final thoughts on where prostate cancer is evolving next in terms of molecular imaging? So, looking at the next few years, PSMA is standard of care. What are the next challenges and where do we go from here in terms of research?

Thomas Hope: Yeah. Well, I think that there's a couple of things. Let's take your world, the initial staging world, right? The patient coming in for a prostatectomy. At some point in time, you get down to the limitation of PET imaging, right? Like you're just not going to see microscopic disease, no matter what the radiotracer. You might make marginal improvements here or there, or, technology-wise, better PET scanners, and we'll learn a little bit to sort of improving detection sensitivity. But at some point in time... It is what it is and you can't change it.

Now, I think for you guys, what will be really interesting and you guys being urologists, is interoperative probes. Now you had mentioned technetium-labeled PSMA compounds. I think that it's a hard technology intraoperatively to deal with because you have to have your gamma camera pointed at the node and see it. And it's very hard. You don't actually see it, per se. So these fluorescent-labeled compounds are very interesting in trying to use interoperative fluorescent compounds to see residual disease in the prostate bed, make sure you have clean margins, maybe see those nodes you missed, trying to figure out how to incorporate the imaging data that you see so you don't leave behind that disease. I think that's really where it gets particularly interesting.

I'm not exactly interested... And I am sure there will be, but it will take a while for there to be the next PSMA in the initial staging of prostate cancer. I think it's just that there's nothing particular on the horizon that is markedly... In PSMA negative disease, and there's 5% of patients at initial staging, and you never know who they are, because you can't see their nodal mets, but maybe there's a role for something.

But yeah. So I think it's more about what you're doing in the OR, how we use the data rather than new radiotracers. Now, on the other end of the spectrum, that's a totally different story, right? When you're talking about metastatic castration-resistant prostate cancer, and you're talking about maybe the... We're learning all about PSMA negative disease, neuroendocrine small cell atypical cancers, post-chemotherapy, what are we doing with those patients? How do we image them? How do we treat them? Are we going to be using non-PSMA radioligand therapies, et cetera? And I think in that area, there's a lot. It's just an open field. We haven't even defined it yet, which I feel like we still haven't defined this small cell variant, let alone figured out how to image it. But there's a ton of work out on that spectrum with different agents in what we're doing.

Matthew Cooperberg: Excellent. Listen, always a pleasure. Thanks so much for spending your time.

Thomas Hope: No problem. Have a good afternoon. Thanks.

Matthew Cooperberg: Thanks, Tom.