Results from the Phase 3 SPOTLIGHT Trial of 18F-rhPSMA-7.3 in Recurrent Prostate Cancer - David Schuster
April 14, 2022
David M. Schuster, MD, Division of Nuclear Medicine and Molecular Imaging, Emory University, Winship Cancer Institute, Atlanta, GA
Phillip J. Koo, MD, FACS, Chief of Diagnostic Imaging and Oncology Physician Executive, Banner MD Anderson Cancer Center
ASCO GU 2022: Detection Rate of 18F-rhPSMA-7.3 PET in Patients with Suspected Prostate Cancer Recurrence: Results from a Phase 3, Prospective, Multicenter Study (SPOTLIGHT)
Blue Earth Diagnostics Announces Key Results from Phase 3 SPOTLIGHT Study of 18F-rhPSMA-7.3, an Investigational PET Imaging Agent, in Biochemical Recurrence of Prostate Cancer
Phillip Koo: Hello. My name is Philip Koo and welcome to URO Today, for a very special presentation which is hot off the presses from GU ASCO 2022. We have with us Dr. David Schuster, who is Professor and Director of the Division of Nuclear Medicine at Emory, and also a Georgia Research Alliance distinguished scientist. And today he's going to be presenting on the late breaking data with regard to the phase 3 SPOTLIGHT Study, looking at rhPSMA in patients with suspected prostate cancer recurrence. So thank you very much, Dr. Schuster for joining us.
David Schuster: Thanks, Phil. It's a pleasure to be here. I guess the first question is, what is 18F-rhPSMA-7.3? Well, it is a radio hybrid platform and it is a high affinity PET radiopharmaceutical with the potential for labeling with alpha or beta emitting radiometals for systemic radiation therapy. So the molecule that was tested had a hot fluorine and a cold gallium, but what you could also do is instead of this gallium, you could put in a hot lutetium 177 or actinium and a cold fluorine, making this a true theranostic. There's excellent image quality and there's also the potential for low urinary excretion, which as we know, is very important in prostate cancer imaging.
So going over the inclusion criteria of the SPOTLIGHT Study, basically inclusion criteria boiled down to standard definitions of biochemical recurrence, post prostatectomy, or non-prostatectomy, and the radiotracer was injected 8 mCi and the patients were imaged 50 to 70 minutes later on PET CT. And the scans were evaluated by three blinded central readers with the majority read predominating. Now a composite standard of truth was utilized. That is histopathology was preferred when possible, but we know it's not always possible or mostly possible to get a biopsy, but conventional imaging was also folded in, and any biopsies had to be performed within 60 days. And if confirmatory imaging was used, it had to be within 90 days. And this could include serial follow up.
Now the co-primary endpoints were patient level correct detection rate at a pre-specified lower statistic threshold of 36.5%. And a region level positive predictive value at that pre-specified lower threshold of 62.5%. So for everyone out there, you should realize that the correct or the confirmed detection rate is not overall detection rate or a raw detection rate. It is the percent of all patient scans who had at least one true positive finding. So the confirmed detection rate is always lower than detection rate or the raw detection rate, since it is a confirmed detection rate. Now, region level PPV is all PET positive regions from all PET positive patients combined into the PPV calculation, so that the denominator of the PPV will always then be greater than the number of positive patients. Since each patient can have up to three positive regions, that is the prostate slash bed pelvis, and what we call extrapelvic, which we know is systemic or nodes outside the pelvis, viscera, bone, et cetera.
Now this combined region level PPV endpoint is really new for the PSMA space, and it was recommended by the FDA. It has not been included to date in other phase 3 registry trials. And I think most importantly, in analyzing the data, a strict imputation method in both co-primary endpoints was utilized. That is if a PET positive lesion was present in it, was not biopsied. And then it was not "proven on conventional imaging," and we know how suboptimal conventional imaging can be for prostate cancer. Then it was considered false positive by default.
So this binary determination either true positive or false positive is quite the high bar. 420 patients were consented. 391 received the radiotracer. 389 had a PET scan done afterwards. That's called the EPSP or the evaluable PET scan population. 366 out of those had sufficient data to determine the composite standard or truth, that's called the EAP or the efficacy analysis population. Roughly 20% of those patients out of the EAP, had histopathology as your standard of truth. And the remainder had the composite, which included imaging. Here we could see a demographics table, for example, under the EAP, the median PSA was 1.27. Most patients originally were a grade group three et cetera. And 79% of these 391 patients were post radical prostatectomy so that for the majority of patients, and we have to also remember that the EAP includes some patients with only a single time point imaging for the standard of truth assessment even if they would've been better served with serial imaging, but the serial imaging cannot be obtained for whatever reason. So let's look at our overall, and this overall detection rate.
83% of valuable PET scans were positive. And we've divided the patients into orange, which are post prostatectomy plus or minus radiotherapy, or gray, which is post radiotherapy only. And we could see here, for example, at a PSA of less than 0.5, we had a 64% detection rate, and that's actually the largest bucket, that went up to 93% with a PSA between one and two. And then, as we get to our higher PSAs, we have a 100%. For the post radiotherapy only, as we would expect, we would have higher PSA levels. And these were a 100% at those higher PSA levels. Looking at the Patient Level Correct Detection Rate, when we look at the majority read, it was 57% using that composite standard of truth, which exceeded the pre-specified threshold. But when we looked at the subset with histopathology only standard of truth, which is the more rigorous standard of truth, the correct detection rate is 81%.
Next the combined region level PPV. Looking at the majority read, it was 60%, which almost, but did not quite meet that pre-specified threshold. And again, that was the composite standard of truth. But when we look at the more rigorous histopathology only standard truth, we could see that, that region level PPV, combined region goes up to 72%. So we have some interesting cases I'd like to show. First of all, we also have some very interesting post prostatectomy at very low PSA levels, but this was a short presentation. So I just picked out two interesting ones. And both of these were non-prostatectomies, but in this patient who had biochemical recurrence following HDR monotherapy with a nadir PSA of 2, trigger PSA of 4.
A prostate biopsy had been done two weeks before this PET, and this correlated quite nicely as we could see here with positive histopathology and uptake in the right apex and the mid prostate, and the rest of the exam was negative. So following SBRT in six months, ADT androgen deprivation, the PSA was so below nadir at 0.11, six months post ADT completion with testosterone recovery. That is really biochemical proof that the rest of the exam was true negative. The second spectacular case is a patient with biochemical recurrence following LDR brachytherapy with a nadir PSA of 0.38. The trigger PSA was 2.6. On the rhPSMA PET, there were multiple pelvic lymph nodes, positive of various sizes. I'm just going to show one here, 9 mm node, very high target to background ratio with a high SUV max. But this patient also had a 5 mm subtle left ischiorectal fossa focus, which had less activity than that node, but to our eyes, just did not belong.
So we did an MR, and that ischiorectal fossa lesion correlated to the PET very nicely. And our body imagers did a very nice job doing a biopsy under MR guidance. And indeed that was malignant prostate cancer metastasis. And this changed the stage of the patient. Even so, the patient underwent salvage proton beam to the pelvic nodes with 70 gray to the radiographic nodes, they boosted where they saw it, 50 gray to the standard field. And then the ischiorectal fossa nodule also got 35 gray, plus six months of androgen deprivation. The PSA is still below nadir at 0.02 nine months post ADT completion with testosterone recovery. A very nice result for the patient. As far as adverse events, Phil, you know that radio tracers are very safe for patients, but, we monitor these adverse events since it is regulated as a drug. Only 4.1% to had at least one treatment emerging adverse event that was considered possibly related or related to the radio tracer. Most of these were very self limiting and no serious adverse events were attributable to the radio tracer.
So in conclusion, the overall detection rates with 18F-rhPSMA-7.3 are high, even at low PSA values. This radio tracer demonstrates a clinically meaningful correct detection rate that meets the pre-specified statistical threshold. So the combined region level PPV did not quite meet the pre-specified threshold, but keep in mind that the data from the cohort with a composite standard of truth are limited by the high proportion of scans, including those in the prostate slash bed that relied on conventional imaging, which we know is a suboptimal standard of truth. Even for nodes, you wouldn't call a 5-mm node positive on a CT scan, but it could light up like a light bulb on PET, and the patients where we had histopathology available as a standard of truth, a higher region level PPV was observed. That was at the level of the expected performance that we would consider excellent. No significant safety signals were identified. So together, these data support the clinical utility of 18F-rhPSMA-7.3 PET and men with recurrent prostate cancer across the wide PSA range. And this acknowledges our whole spotlight study group. This was a truly multicenter international effort.
Phillip Koo: Wonderful, thank you very much, David. That was an amazing presentation reviewing the data from an important study, SPOTLIGHT. I think it really highlights the challenges with regards to designing diagnostic trials-
David Schuster: Mm-hmm (affirmative).
Phillip Koo: But it's great to see that level of robustness with regards to standard of truth and how it creates a lot of challenges. So it really requires a little bit more critical thinking. A lot of PSMA agents are now out there. We have a couple that are FDA approved, but you did mention how rhPSMA was a little unique in the sense that it's a true radiohybrid. So can you expand upon that and how you think this might distinguish this PSMA agent from the others?
David Schuster: Sure. I can't really compare it to the other PSMA agents because this trial wasn't designed to do that, but it is a very interesting molecule in that you have your PSMA scaffold. And it's like having a car and changing the tires depending on what you want to do with it, like in the winter, or go off road or whatever. So what you could do is unlike for example, right now. Again, this is something that needs to be studied, but when we do a theranostic, we don't use exactly the same molecule, just switching out to the radionuclide. Even I think we dotatate that, that really doesn't happen, right? So it's a little different, the biodistribution could be a little different, and of course, sodium iodide is the perfect theranostic and we've had for what 70, 80 years now.
So this has a potential to be that where you could actually take the same molecule and just flip out the radionuclide or the effective radionuclide, and then make therapy, radiotherapy out of it, versus a diagnostic. I think the other potential, and this needs, I think, a little more study to determine how to capitalize on this. There is a potential for lower urinary excretion. And again, this is a very high affinity radiopharmaceutical. We know that what's been and published In Vitro doesn't always equate exactly what happens when you put into a human cells islet. Everyone out there, looked up the studies themselves and compare this to other radiotracers. But that's another, I think, potential where if we could capitalize on the low urinary excretion, then this could be an important radiopharmaceutical for looking, especially at the areas around the bladder, such as the vesicourethral anastomosis.
Phillip Koo: Great. So, David, you've done pioneering work with regards to prostate cancer imaging for years and years. And we really appreciate all the contributions you've made with [inaudible] acumen and now rhPSMA. A sort of stepping back and given the wealth of experience you have, and seeing that case, case number two, where you're seeing metastasis in places that you weren't used to seeing-
David Schuster: Right.
Phillip Koo: What advice do you have for people who are starting to use this right now and are finding it challenging, especially from clinical perspective, with regards to how to manage patients?
David Schuster: Yeah. So this is the million dollar question because tuning into ASCO GU, you had oncology... Med oncs out there and rad oncs and urologists, and it's a little bit of what do we do now? I take it like a similar to, when we started having better and better CT scans, right? And then we started seeing these tiny little nodules in the lungs and we did not know what to do with them. So I think that this needs to be studied more. One could say, well, the standard of care was based on conventional imaging, and therefore, you should at least get standard of care. And that's how our algorithms came about. But a good counterargument is, well, maybe those algorithms were based on poor data because the standard of conventional imaging did not show you really everything you were dealing with.
So I think this is where having really well designed outcome studies are important. And you mentioned some of the pioneer work, my rad-onc partner, [Ashish Sahni 00:15:56] and I have done work on EMPIRE-1 and now EMPIRE-2 is underway. And I think we're going to need studies like that, which actually show not just a change in management, but does it really make a difference that we're seeing these little things, or will the biology predominate in the end? I think that, biology will always predominate, right? Because we're humans, but the question is how much and what degree. So I think this is an area that's right for study, and we need to do the studies. We need to find out. Will outcomes change. And that's a million dollar question right now.
Phillip Koo: You're absolutely right. And thank you very much. We're very excited to see more of that data, which Emory and your group has always been leading with regards to a lot of those studies looking at outcomes. I remember some of your trainees presenting at the [inaudible] meeting a couple years ago, and they were just really, really thoughtful presentation. So thank you again.
David Schuster: Great.
Phillip Koo: We appreciate the time. And again, we look forward to having you back.
David Schuster: Thank you for having me.