Role of Radium-223 in the Setting of PSMA "Presentation" - Oliver Sartor
February 15, 2024
At the 2024 UCSF-UCLA PSMA Conference, Oliver Sartor discusses radium-223's pioneering role as the first FDA-approved alpha emitter in medicine, sparking interest in radioligand therapies, and discusses its limitations, including the inability to target soft tissue. Dr. Sartor suggests that improved patient selection through PSMA PET imaging and ongoing trials exploring treatment combinations could redefine radium-223's utility in the evolving landscape of metastatic castrate-resistant prostate cancer treatment.
![UCLAUCSF video logo stroke](/images/UCLAUCSF_video_logo_stroke.png)
Biographies:
A. Oliver Sartor, MD, Professor of Medicine, Urology, and Radiology, Director Radiopharmaceutical Trials, Mayo Clinic, Rochester, MN
![UCLAUCSF video logo stroke](/images/UCLAUCSF_video_logo_stroke.png)
Biographies:
A. Oliver Sartor, MD, Professor of Medicine, Urology, and Radiology, Director Radiopharmaceutical Trials, Mayo Clinic, Rochester, MN
Read the Full Video Transcript
Oliver Sartor: Well, I'm sorry to say I'm your last speaker. So it's been a lovely conference, and in some ways, I wish it could go on a little further. So it's really been fun. I'm going to talk about radium, a little bit forgotten perhaps, but there's a lot to say about radium. The first thing I'll say is it is, in fact, the first alpha emitter in medicine and the first and only FDA-approved alpha emitter. It had, actually, repercussions, I think, beyond simply a positive trial. It really invigorated a lot of interest in the radioligand space, and it's not a radioligand in the usual sense of the word because it's not bound. It's a free isotope, but it created a buzz that was, in part, due to, I think, a $2.9 billion buyout from Bayer for Algeta. It invigorated the field, I think, in a way that was almost disproportionate.
I have a little hats off to Roy Larsen and Øyvind Bruland. I don't know if you know those guys. They're the Norwegians who put it together. I think if we had to give any one particular credit, it'd be to Roy, who's an incredibly bright guy, and he continues to do good work. One of the things that some people may not recognize, but I wanted to point this out, it's slightly relevant, and that is the design elements that are shared between ALSYMPCA and VISION. It turns out that we looked, in ALSYMPCA, at patients with metastatic CRPC, most of whom were beyond all known therapies.
We had a positive biomarker. In this case, it was bone scan positivity as opposed to PSMA PET positivity in VISION. We had a negative biomarker selection. We took visceral mets and excluded them because we obviously weren't going to radiate those with radium. We took big lymph nodes and excluded them, as well. We also had an exclusion in VISION. We had a standard of care, plus or minus design. That was not an accident, and that was the design that was envisioned, but not in PSMAfore, not in ECLIPSE, not in the third trial, SPLASH trial. Chemotherapy was excluded in the control arm because of the concomitant issues.
The treatment was fixed, six cycles q 4 weeks, similarly fixed in the VISION trial, and the endpoint was overall survival. By the way, for rPFS, that was added a little later in VISION. That was not part of the original design. rPFS was thrown in. I just wanted to point that out because a lot of people may not have recognized it. Overall survival was present, relatively modest. But if we look at metastatic castrate-resistant prostate cancer, everything we've done has been modest, and I think we have to acknowledge that.
Nevertheless, the positive trial had a lot of implications. Here's the problem. So this is the landscape at the beginning of ALSYMPCA in 2008, and if we look at what we had, there was one therapy, docetaxel. If we look at the landscape today, how shall we say, it's different, very, very, very different. So we have to imagine that the therapies that were designed in 2008 or started in 2008 are going to be relevant today, but we just really don't know. That's really one of the huge problems with radium, is, how do we interpret the positive ALSYMPCA trial in the context of an immensely changed therapeutic landscape?
I'm going to say 2008 and 2024 are just light years apart in this disease. So when we talk about giving radium today, how much confidence do we have in the post-abiraterone, post-enzalutamide space when we never really tested it in that space? That's a problem. Now, the problem number two with radium is good. We have a disease, prostate cancer, that's remarkably osteoblastic, remarkably bone-centric. That's really good, and you can take that osteoblastic lesion and target radium pretty effectively to it. That's good.
But the problem is that it doesn't go to soft tissue, and we know that there's a lot of soft tissue disease. Particularly when you do PSMA PET, you appreciate the degree of soft tissue disease in a manner that we didn't even know before. Now, maybe we can do a little better on positive selection. Bone scans actually are not too bad, and by the way, the hydroxyapatite that is found with the bone scan is identical to the binding site that we are going to be using for radium. So it is a biomarker. It is a real biomarker.
Now, the sodium fluoride PET can bind to the same hydroxyapatite, and it's good. But of course, there can be some false positives along the way. Not everything that binds to hydroxyapatite is a tumor. I think we all know that. Now, there's a little bit of data about the negative selection. So CAT scans, I don't need to tell this audience, are just so-so, and PSMA PET is better. If we turn out, if we use the PSMA PET to be able to exclude visceral and soft tissue lesions, we can probably come out better, and we can.
There's a nice study, again, citing the Germans who seem to be, over and over, present in this particular space doing great work, and here's some really good work. It's showing that the selection, in terms of PSA waterfall plots, for radium treatment with bone scan and CT scan as compared to PSMA PET, you actually do have PSA declines if you choose your patients more carefully. So the exclusion of the soft tissue results in a better outcome for radium, at least as measured by PSA. Nobody's ever done this in a large trial, of course.
Now, the other thing that is a little bit interesting is assessing the response to radium has really been problematic because you certainly don't want to do a bone scan. That doesn't work. If we want to look at PSMA PET improvement, we can see this in a subset of patients, but it's only a small subset. So imaging in the bone with PSMA PET can actually be used to monitor the response with radium-223, but there's also been interesting work done with diffusion-weighted MRI, looking at this heterogeneity and quantitating this with ADC mapping. It turns out this is not bad.
Now, not everybody responds. This is from an annual meeting abstract, but you can go patient by patient and see some of these patients have a substantial increase in their ADC. This seems to correlate quite well with declines in PSA, and some of these patients actually go on for a period of time without additional therapies and, perhaps, show a response. So monitoring the response to radium can occur, perhaps, through better imaging.
Now, the RALU study, I think, deserves a mention. The RALU study is a real-world study of individuals who received radium and then went on to receive PSMA lutetium later. This has now been published in the Journal of Nuclear Medicine, and it turns out that this is a real-world type of experience. But what we did is gather, as best as possible, some of the AEs. We had pretty good AEs, and then we had survival. It turned out to be about 12.6 months from lutetium. Now, that's not too far away from the 15.3 that we saw in VISION.
Now, what's not weird, but notable is these are not your typical radium patients treated in ALSYMPCA. The overall survival from the first radium injection was 31 months. That means these patients lived a long, long, long, long time after the radium. But the bottom line is you probably can safely give, and there's additional data, by the way, you can safely give lutetium post radium. So that's, perhaps, noteworthy.
Now, some new Phase IIIs, and one of these is getting fairly close to reporting. It's called PEACE III. This was put together by the EORTC. Bertrand Tombal, Silke Gillessen have been able to look at enzalutamide and radium. We have Louise talking to us about ENZA-p. Maybe we can call this one ENZA-r, but nevertheless, they called it PEACE III. So one of the things that was found out relatively early on, and I'm not going to go into all the very problematic issues associated with radium and fracture that were found in the ERA 223 trial... This was from Bertrand's presentation at 2019 ASCO, and the bottom line is if you had bone-protective agents, you dramatically decreased the probability of a fracture.
But if you did not give the bone-protective agents, then the fracture rate was extraordinarily high, let's talk about 50%, and that's rather alarming. That was the same problem, by the way, seen in the ERA 223 trial, and it really killed the trial. I'll simply say that was a huge setback for the radium folks. Now, DORA, sorry, Mike, I did not get your... This is Mike's trial, and he sent me a scheme. I just didn't have enough time to put it in. So basically, this is going to be docetaxel versus the combination of radium and docetaxel.
Now, given in combination with radium, there's a lower dose of the docetaxel. The trial's ongoing. I'm going to give Mike a real shout-out because he's been pretty heroic in getting this thing underway. This started in COVID and had all sorts of issues, but he's kept hammering and hammering. Mike's going to bring it over the finish line. So look forward to seeing what that might show.
There are a couple of earlier phase studies I thought I might mention. This is the AlphaBet study. Okay? We just talked about the combinations of alphas and betas, and here we go. One of the things that makes this a little bit interesting to me is that if you look at the probability of failure, if you're giving lutetium beta, you're going to be looking at failure in the liver. We've emphasized that. We had a great tumor board. Thanks, Tom, for bringing out that liver issue on more than one case. But if we look at the lymph nodes versus the bone, most of the failures are going to occur in bone. So excluding the liver, which most patients do not have liver metastases, bone is more susceptible to progression post-lutetium. Maybe what you could do is hit the bone with a combination of radium and lutetium, and then the lymph nodes can be taken care of with lutetium. Maybe something interesting could happen here.
Now, this is the one that I think is the most surprising. This is BAT-RAD. Okay. Now, BAT is bipolar androgen therapy. That's high-dose testosterone, 400 milligrams of testosterone cypionate. We've been crazy enough or stupid enough to be able to look at this in a little bit of detail. We actually published a case report about, I don't know, four or five years ago, and there's a crazy hypothesis that we don't know if it's really accurate or not. But now, it's being looked at in a prospective trial. The interesting thing about this prospective trial is I'll just say it's interesting. I'm getting reports that it's interesting. So we'll just watch it closely as we go forward.
So in summary, radium-223 suffers from the lack of new data. We have a prospective randomized trial published in 2013, begun in 2008. It's the only positive Phase III trial. It's from a different era. It's almost like an archeological dig right now. Now, the various combinations of radium in current clinical trials are interesting, with enzalutamide or docetaxel. We're going to likely hear from PEACE III this year, and we'll have the DORA trial, which is still accruing.
I'll say some interesting combination studies are underway, but we need more work, more data to make radium-223 relevant today. I think in the tumor board that Tom presented, I don't think anybody was jumping to give radium to any of those patients, and I think that's sort of the real-world realities. So with that, that's your last lecture of the day. We'll turn it over, I think, to Jeremie. Okay.
Oliver Sartor: Well, I'm sorry to say I'm your last speaker. So it's been a lovely conference, and in some ways, I wish it could go on a little further. So it's really been fun. I'm going to talk about radium, a little bit forgotten perhaps, but there's a lot to say about radium. The first thing I'll say is it is, in fact, the first alpha emitter in medicine and the first and only FDA-approved alpha emitter. It had, actually, repercussions, I think, beyond simply a positive trial. It really invigorated a lot of interest in the radioligand space, and it's not a radioligand in the usual sense of the word because it's not bound. It's a free isotope, but it created a buzz that was, in part, due to, I think, a $2.9 billion buyout from Bayer for Algeta. It invigorated the field, I think, in a way that was almost disproportionate.
I have a little hats off to Roy Larsen and Øyvind Bruland. I don't know if you know those guys. They're the Norwegians who put it together. I think if we had to give any one particular credit, it'd be to Roy, who's an incredibly bright guy, and he continues to do good work. One of the things that some people may not recognize, but I wanted to point this out, it's slightly relevant, and that is the design elements that are shared between ALSYMPCA and VISION. It turns out that we looked, in ALSYMPCA, at patients with metastatic CRPC, most of whom were beyond all known therapies.
We had a positive biomarker. In this case, it was bone scan positivity as opposed to PSMA PET positivity in VISION. We had a negative biomarker selection. We took visceral mets and excluded them because we obviously weren't going to radiate those with radium. We took big lymph nodes and excluded them, as well. We also had an exclusion in VISION. We had a standard of care, plus or minus design. That was not an accident, and that was the design that was envisioned, but not in PSMAfore, not in ECLIPSE, not in the third trial, SPLASH trial. Chemotherapy was excluded in the control arm because of the concomitant issues.
The treatment was fixed, six cycles q 4 weeks, similarly fixed in the VISION trial, and the endpoint was overall survival. By the way, for rPFS, that was added a little later in VISION. That was not part of the original design. rPFS was thrown in. I just wanted to point that out because a lot of people may not have recognized it. Overall survival was present, relatively modest. But if we look at metastatic castrate-resistant prostate cancer, everything we've done has been modest, and I think we have to acknowledge that.
Nevertheless, the positive trial had a lot of implications. Here's the problem. So this is the landscape at the beginning of ALSYMPCA in 2008, and if we look at what we had, there was one therapy, docetaxel. If we look at the landscape today, how shall we say, it's different, very, very, very different. So we have to imagine that the therapies that were designed in 2008 or started in 2008 are going to be relevant today, but we just really don't know. That's really one of the huge problems with radium, is, how do we interpret the positive ALSYMPCA trial in the context of an immensely changed therapeutic landscape?
I'm going to say 2008 and 2024 are just light years apart in this disease. So when we talk about giving radium today, how much confidence do we have in the post-abiraterone, post-enzalutamide space when we never really tested it in that space? That's a problem. Now, the problem number two with radium is good. We have a disease, prostate cancer, that's remarkably osteoblastic, remarkably bone-centric. That's really good, and you can take that osteoblastic lesion and target radium pretty effectively to it. That's good.
But the problem is that it doesn't go to soft tissue, and we know that there's a lot of soft tissue disease. Particularly when you do PSMA PET, you appreciate the degree of soft tissue disease in a manner that we didn't even know before. Now, maybe we can do a little better on positive selection. Bone scans actually are not too bad, and by the way, the hydroxyapatite that is found with the bone scan is identical to the binding site that we are going to be using for radium. So it is a biomarker. It is a real biomarker.
Now, the sodium fluoride PET can bind to the same hydroxyapatite, and it's good. But of course, there can be some false positives along the way. Not everything that binds to hydroxyapatite is a tumor. I think we all know that. Now, there's a little bit of data about the negative selection. So CAT scans, I don't need to tell this audience, are just so-so, and PSMA PET is better. If we turn out, if we use the PSMA PET to be able to exclude visceral and soft tissue lesions, we can probably come out better, and we can.
There's a nice study, again, citing the Germans who seem to be, over and over, present in this particular space doing great work, and here's some really good work. It's showing that the selection, in terms of PSA waterfall plots, for radium treatment with bone scan and CT scan as compared to PSMA PET, you actually do have PSA declines if you choose your patients more carefully. So the exclusion of the soft tissue results in a better outcome for radium, at least as measured by PSA. Nobody's ever done this in a large trial, of course.
Now, the other thing that is a little bit interesting is assessing the response to radium has really been problematic because you certainly don't want to do a bone scan. That doesn't work. If we want to look at PSMA PET improvement, we can see this in a subset of patients, but it's only a small subset. So imaging in the bone with PSMA PET can actually be used to monitor the response with radium-223, but there's also been interesting work done with diffusion-weighted MRI, looking at this heterogeneity and quantitating this with ADC mapping. It turns out this is not bad.
Now, not everybody responds. This is from an annual meeting abstract, but you can go patient by patient and see some of these patients have a substantial increase in their ADC. This seems to correlate quite well with declines in PSA, and some of these patients actually go on for a period of time without additional therapies and, perhaps, show a response. So monitoring the response to radium can occur, perhaps, through better imaging.
Now, the RALU study, I think, deserves a mention. The RALU study is a real-world study of individuals who received radium and then went on to receive PSMA lutetium later. This has now been published in the Journal of Nuclear Medicine, and it turns out that this is a real-world type of experience. But what we did is gather, as best as possible, some of the AEs. We had pretty good AEs, and then we had survival. It turned out to be about 12.6 months from lutetium. Now, that's not too far away from the 15.3 that we saw in VISION.
Now, what's not weird, but notable is these are not your typical radium patients treated in ALSYMPCA. The overall survival from the first radium injection was 31 months. That means these patients lived a long, long, long, long time after the radium. But the bottom line is you probably can safely give, and there's additional data, by the way, you can safely give lutetium post radium. So that's, perhaps, noteworthy.
Now, some new Phase IIIs, and one of these is getting fairly close to reporting. It's called PEACE III. This was put together by the EORTC. Bertrand Tombal, Silke Gillessen have been able to look at enzalutamide and radium. We have Louise talking to us about ENZA-p. Maybe we can call this one ENZA-r, but nevertheless, they called it PEACE III. So one of the things that was found out relatively early on, and I'm not going to go into all the very problematic issues associated with radium and fracture that were found in the ERA 223 trial... This was from Bertrand's presentation at 2019 ASCO, and the bottom line is if you had bone-protective agents, you dramatically decreased the probability of a fracture.
But if you did not give the bone-protective agents, then the fracture rate was extraordinarily high, let's talk about 50%, and that's rather alarming. That was the same problem, by the way, seen in the ERA 223 trial, and it really killed the trial. I'll simply say that was a huge setback for the radium folks. Now, DORA, sorry, Mike, I did not get your... This is Mike's trial, and he sent me a scheme. I just didn't have enough time to put it in. So basically, this is going to be docetaxel versus the combination of radium and docetaxel.
Now, given in combination with radium, there's a lower dose of the docetaxel. The trial's ongoing. I'm going to give Mike a real shout-out because he's been pretty heroic in getting this thing underway. This started in COVID and had all sorts of issues, but he's kept hammering and hammering. Mike's going to bring it over the finish line. So look forward to seeing what that might show.
There are a couple of earlier phase studies I thought I might mention. This is the AlphaBet study. Okay? We just talked about the combinations of alphas and betas, and here we go. One of the things that makes this a little bit interesting to me is that if you look at the probability of failure, if you're giving lutetium beta, you're going to be looking at failure in the liver. We've emphasized that. We had a great tumor board. Thanks, Tom, for bringing out that liver issue on more than one case. But if we look at the lymph nodes versus the bone, most of the failures are going to occur in bone. So excluding the liver, which most patients do not have liver metastases, bone is more susceptible to progression post-lutetium. Maybe what you could do is hit the bone with a combination of radium and lutetium, and then the lymph nodes can be taken care of with lutetium. Maybe something interesting could happen here.
Now, this is the one that I think is the most surprising. This is BAT-RAD. Okay. Now, BAT is bipolar androgen therapy. That's high-dose testosterone, 400 milligrams of testosterone cypionate. We've been crazy enough or stupid enough to be able to look at this in a little bit of detail. We actually published a case report about, I don't know, four or five years ago, and there's a crazy hypothesis that we don't know if it's really accurate or not. But now, it's being looked at in a prospective trial. The interesting thing about this prospective trial is I'll just say it's interesting. I'm getting reports that it's interesting. So we'll just watch it closely as we go forward.
So in summary, radium-223 suffers from the lack of new data. We have a prospective randomized trial published in 2013, begun in 2008. It's the only positive Phase III trial. It's from a different era. It's almost like an archeological dig right now. Now, the various combinations of radium in current clinical trials are interesting, with enzalutamide or docetaxel. We're going to likely hear from PEACE III this year, and we'll have the DORA trial, which is still accruing.
I'll say some interesting combination studies are underway, but we need more work, more data to make radium-223 relevant today. I think in the tumor board that Tom presented, I don't think anybody was jumping to give radium to any of those patients, and I think that's sort of the real-world realities. So with that, that's your last lecture of the day. We'll turn it over, I think, to Jeremie. Okay.