Appropriate Use Criteria for PSMA PET Imaging - Hossein Jadvar

March 10, 2022

Phillip Koo is joined by Hossein Jadvar to discuss the Appropriate Use Criteria (AUC) for PSMA PET in prostate cancer. Dr. Jadvar, the first author of the publication, discusses the process of developing the AUC for PSMA PET in prostate cancer and reviews 11 clinical scenarios that may be encountered by ordering professionals when they want to consult the AUC for ordering their imaging.

To guide referring and imaging physicians to use PSMA PET imaging agents appropriately, The Society of Nuclear Medicine and Molecular Imaging (SNMMI), American College of Nuclear Medicine (ACNM), American Urological Association (AUA), Australia and New Zealand Society of Nuclear Medicine (ANZSNM), American Society of Clinical Oncology (ASCO), European Association of Nuclear Medicine (EANM), and American College of Physicians (ACP) worked collaboratively to develop criteria for the appropriate use of this new imaging technology.

Biographies:

Hossein Jadvar, MD, Ph.D., MPH, MBA, FACNM, FSNMMI, Professor of Radiology, Urology, and Biomedical Engineering (Tenured), Keck School of Medicine, University of Southern California, Los Angeles, CA

Phillip J. Koo, MD, FACS, Chief of Diagnostic Imaging and Oncology Physician Executive, Banner MD Anderson Cancer Center


Read the Full Video Transcript

Phillip Koo: Hello. My name is Phillip Koo. Welcome to UroToday. Today, we are going to talk about a very important paper that was recently published, focusing on the appropriate use criteria for PSMA PET in prostate cancer. And to talk about this, we have none other than the first author on that paper, Dr. Hossein Jadvar, who's known to many of us in the prostate cancer community for a variety of great work that he's done over the years. Dr. Jadvar is a tenured professor of radiology at the Keck School of Medicine at University of Southern California. And he also has joint appointments in urology and also biomedical engineering, which I think gives him a very broad perspective on a lot of topics. So we're very fortunate to have him with us, and I'll turn over the mic to Dr. Jadvar.

Hossein Jadvar: Well, thank you very much, Phillip, for that very kind introduction. It's my pleasure to be joining you in this discussion. So my task is to talk about the appropriate use criteria for PSMA PET in prostate cancer that was recently published in the Journal of Nuclear Medicine. Before we dive into the AUC, I just wanted to show you this slide regarding the natural history of prostate cancer, because the AUC directly represents different phases of the disease in prostate cancer, which are generally divided into three main phases. One is initial staging, the middle phases, biochemical recurrence after a definitive primary therapy, which can be either radical prostatectomy or radiation therapy, and finally, the metastatic disease part of the prostate cancer.

As you know, about 30% to 40% of patients with biochemical recurrence will eventually develop a castrate-resistant prostate cancer, which is at the right-hand corner of this slide. And there are a number of treatments that are available in this space, including excitingly most recently the Lutetium-177-PSMA-11, with very favorable results on the VISION trial, which I'm sure you have talked with other experts regarding that trial.

Also, before I move on, I just want to talk about oligometastatic disease, which is also another area of major interest. And we will talk about that a little bit in the AUC. Oligometastatic disease is considered to be a limited metastatic disease, variably defined as 1 to 5 metastatic lesions identified on imaging. Of course, the more sensitive to imaging, like PSMA PET-CT, the more chances that you have to detect the oligometastatic phase of the disease. And those are generally divided into three sections, the synchronous oligometastatic disease, which is at the time of initial staging with the primary tumor detection. Then the second one is metachronous oligometastatic disease, when we image the patients for biochemical recurrence. And then finally there is induced oligometastatic disease, when the patient has oligometastatic disease, undergoes treatment, and then after the treatment is successful at many sites, there may be a few metastases that are left behind. Why is this important? Because you can use metastatic directive therapy in this patient.

So with that introduction, let's talk about PSMA, which is the focus in this talk. Now we have come a long way with regard to development of PSMA, when the very first tracer became FDA-approved back in the late 80s, which was known as processing. Of course, that was not very useful and is not available anymore, but after that development, there were many other strides in understanding what PSMA is with regard to its genetics, expression in the cells across the cancer. And then there were a number of developments in new generation radio traces based on PSMA that we will talk about. And today in 2022, we now have two traces that are FDA-approved for use, and we have AUC based upon these traces.

Here are the two approved traces in the United States, the Gallium 68 PSMA-11, that was approved for use at UCSF and UCLA in December of 2020. More recently, in May of 2021, the F-18-DCFPyL was approved, which is with a commercial name of Pylarify. But there are other radio tracers out there that we may see in the future, including another interesting one called F-18-PSMA-1007 that was developed in Germany, which has a little different biodistribution, more of a hepatobilary clearance rather than renal clearance for the two approved traces that we talked about.

So with that, let's talk about what AUC is. Now, AUC history goes back to 2014, when the United States Congress passed a law, the Protecting Access to Medicare Act of 2014. And basically through that act, the advanced diagnostic imaging services, the so-called ADIS, was tied... The reimbursement for them was tied to development of these appropriate use criteria. The advanced diagnostic imaging services were considered to be magnetic resonance imaging or MRI, CT, and all of nuclear medicine procedures, including PET-CT. The idea here is that the physicians who want to have one of these imaging studies done on their patients need to consult these documents, these appropriate use criteria, and this will be available through a clinical decision support mechanism, which is basically software. And after that consultation, they will see if that imaging is appropriate for the particular clinical condition of their patient. And if that is appropriate, then they can order that test.

Now, who develops these AUCs? These are created or endorsed by national medical specialty societies or the so-called provider-led entities. Society of Nuclear Medicine and Molecular Imaging is one of these approved provider-led entities. The law basically excludes insurance companies or patient advocacy groups to develop these documents. The implementation at this time for the AUCs and use of the decision support mechanism has been earmarked for January 1 of next year, 2023. So at that time, physicians are expected to be consulting these AUCs.

Now, how do we actually develop one of these AUCs? There's a rather involved procedure that needs to be followed, including finding the evidence, which is usually in the form of systematic reviews and then rating the evidence according to a core scheme. And this is done through a multidisciplinary expert panel, who will actually rate these documents or references. And then after that rating, an evidence table is provided and discussed among the experts. Now, it is possible also for the experts to provide expert opinions, if the evidence is lacking in some areas. And eventually a decision flow chart is implemented based on all this information and scored with regard to various types of clinical scenarios that are pertinent to the clinical condition that is on the table.

Eventually, there is a writing process. Specific language needs to be used. Then this document is endorsed by all the organizations that were involved from the beginning to develop these AUCs. And we also send it for peer review to external reviewers who were not involved with the development of the AUC. And at the end, this will be sent for publication to a peer review journal. In our case, it is Journal of Nuclear Medicine.

So the document that we are going to focus on was published in Journal of Nuclear Medicine in January of 2022, entitled Appropriate Use Criteria for Process-Specific Membrane Antigen PET imaging. The organizations that were involved included SNMMI; ACNM, which is American College of Nuclear Medicine; American Society of Clinical Oncology or ASCO; American Urological Association, AUA; European Association of Nuclear Medicine, EANM; [inaudible] College of Physicians, which is ACP; and also the Australian and New Zealand Society of Nuclear Medicine.

The panel came up with 11 clinical scenarios that may be encountered by ordering professionals when they want to consult the AUC for ordering their imaging. These clinical scenarios are listed in here in this table, and I'm going to go to them one by one to give you more detail.

So the 11 clinical scenarios were divided into 5 related to initial staging, 3 that are related to a biochemical recurrence phase of the disease, and 3, which are related to the metastatic cancer disease and prostate cancer phase of the disease, again, going back to the initial slide that I showed.

So let's go over these scenarios. This is scenario number 1 on their initial staging. These are patients with suspected prostate cancer, and they are being evaluated for targeted biopsy and detection of intraprostatic tumor. For this particular scenario, the panel gave it a score of 3, which is rarely appropriate. I should mention that the scoring is between 1 and 9. One to 3 is considered rarely appropriate. Four, 5, and 6 are considered to be maybe appropriate. And then 7, 8, and 9 are considered appropriate. So in this case, this falls into rarely appropriate. Why? Because there was limited evidence that was collected with regard to this particular clinical scenario. Also, we know that PSMA expression is heterogeneous. And, in fact, 5% to 10% of primary tumors and metastases may not express the PSMA. It is, however, thought that it may be useful in some specific cases as an accompaniment to the multiparametric MRI, which is typically performed in patients who are going to be having biopsy for possible or suspected prostate cancer.

And in fact, in that regard, there is a clinical trial that is ongoing with the name of primary. It will be a multicentered clinical trial to provide evidence to see if there is any added value of PSMA PET to multiparametric MRI for detecting clinically significant prostate cancer. So until we get the results on this particular trial or other trials, this particular scenario was considered a score of 3 or rarely appropriate.

In scenario 2, under initial staging, these are patients with very low, low, and favorable intermediate risk prostate cancer. This was given a score of 2 or rarely appropriate. The risk stratification here is based on the NCCN guidelines. And also noted in the NCCN guidelines, imaging is usually not indicated in this group of patients anyway. And initial management tends to be observation or active surveillance. And also there is really not much evidence in this clinical space. Therefore, because of that, this score was 2 for scenario number 2.

Scenario number 3, these are again on their initial staging. These are patients who are newly diagnosed. They already have the diagnosis of prostate cancer, and they have unfavorable intermediate, high, or very high risk prostate cancer. This was scored as 8 or appropriate. Again, the risk stratification is based on NCCN guidelines here. And in this particular clinical space, there was actually pretty good supportive evidence that shows PSMA tests are more informative than conventional imaging.

One of these pivotal trials was ProPSMA from Australia that basically randomized the patients receiving either conventional imaging or Gallium 68 PSMA-11 for a staging of high-risk prostate cancer. And it was shown that PSMA is more accurate and also has a major impact in staging and management of these patients. And there were fewer equivocal findings with PSMA testing in comparison to conventional imaging. Even the radiation exposure was less with PSMA, and overall, the cost also turned out to be less with PSMA.

There was another clinical trial called OSPREY, which was with the DCFPyL as the tracer. This is F-18 labeled. And in their cohort A, they had patients with high risk primary prostate cancer. And again, this particular trial showed that there was near perfect specificity with PSMA. Sensitivity was moderate, around 40%, because this was compared to pathology. And, of course, you can have micrometastases into the small lymph node, which may not show significant or sufficient signal on PSMA PET. Again, this was scored 8 as appropriate.

Under initial staging scenario number 4, these are patients with newly diagnosed unfavorable, intermediate high, or very high risk prostate cancer, but they already had conventional imaging, which was negative or equivocal. Or conventional imaging showed oligometastatic disease, which we talked about when I was talking about the first slide. Again, this was the NCCN guidelines risk stratification, and the panel recognized that at least for some time, even after the approval of the PSMA traces, some clinicians may still use conventional imaging, because, perhaps, the PSMA PET may not be readily available to them. But in this case, it was recognized that based on evidence that is available, patients who have negative or equivocal conventional imaging may in fact have metastatic disease, as we talked about in the last slide, for example, with appropriate PSMA trials. Therefore, PSMA was considered appropriate in this case, if there is access to PSMA PET imaging.

Also, patients who have oligometastatic disease on conventional imaging, these are, again, variably defined as 1 to 5, limited metastasis. Studies have shown that if PSMA PET is done in these patients, in fact, these patients are not oligometastatic disease but have polymetastatic disease. So again, if there is access to PSMA PET, this was considered to be an appropriate use of this imaging modality. So, therefore, their score was at 8.

Finally, under initial staging scenario 5, these are newly diagnosed prostate cancer patients who already have widespread metastatic disease on conventional imaging. In this case, the panel decided on a score of 4, which is maybe appropriate. Now, at this time, there is little evidence that PSMA PET will add any additional value when you already know that the patient has widespread metastatic disease on conventional imaging. However, the score was given in kind of a midrange of 4 with the reasoning that we anticipate that PSMA radioligand therapy, for example, with Lutetium-177-PSMA-617 may become FDA-approved very soon. And when that happens, then PSMA PET imaging would be necessary to document that the metastases of these patients are expressing sufficient amount of PSMA before they undergo radioligand therapy. So at this time, it's called a 4, but this scoring may be revised in the future after approval of the PSMA radioligand therapy.

Now moving on under biochemical recurrence, this is scenario number 6. These are patients with PSA persistence or PSA rise. It's a higher score of 9, which means it's appropriate. There's a very good amount of supportive evidence in this clinical space. I've listed some of them in here, for example, a paper in Lancet Oncology that was published in 2019, which compared fluciclovine to Gallium 68 PSMA-11; another paper in JAMA Oncology, which focused on Gallium 68 PSMA-11; and then a couple of trials, including CONDOR trial. This was a phase 3 trial with patients who had uninformative conventional imaging. And the tracer in this case was the DCFPyL, showing that a good number of these patients, almost 50% of these patients, have detectable disease with PSMA PET. Also, the OSPREY that we mentioned before, the cohort B of that trial, using the PyL, showed very good use of PSMA PET in this clinical space.

There has been also trials that have shown that this new information provided by PSMA PET will impact management. Of course, there are new studies that are needed to see if this changing management will actually result in improving of patient outcome. And there are some trials underway with that, including PSMA SRT, which is spearheaded at UCLA. This is a randomized phase 3 trial, which compares PSMA versus conventional imaging-guided salvage radiotherapy for recurrent prostate cancer. Of note, I should mention that the panel decided that there should be no PSA threshold to define when using PSMA. It has been shown that PSMA PET can be positive, even if the PSA is below what is the current criteria based on AUA for patients post prostatectomy or in patients who have received radiation therapy based on the ASTROPhoenix criteria and also the fact that there are other risk factors other than just the level of TSA that should be considered. And, therefore, the panel decided that no PSA threshold should be defined in this case.

And then we move on to scenario number 7. This is very similar to scenario number 6, except that this patient has PSA rise above the nadir, which is the ASTROPhoenix criteria for biochemical recurrence in patients who received definitive radiotherapy. Very similar to scenario number 6, the panel decided that a score should be high, and, therefore, this is appropriate. There is supporting evidence in this case. So again, very similar to our discussion with regard to post prostatectomy patients.

In scenario number 8, under biochemical recurrences, these are patients who have PSA rise after focal therapy of primary tumor. And this was scored in the midrange of maybe appropriate, mostly because there is very little data in this area with patients who receive focal therapy, for example, with high intensity ultrasound or fulguration or cryoablation or methodology like that for their primary tumor. The other problem was that there is really no good definition of what biochemical recurrence is in these patients who have received focal therapy. So at this time, the panel decided that in some cases, this may be appropriate, but not in general.

And now we move on to the last phase of the disease. This is the castration-resistant prostate cancer scenario number 9. These are patients with nonmetastatic castration-resistant prostate cancer, the so-called M0 disease, on conventional imaging. This was given a score of 7 or appropriate. Again, because of a rather good amount of evidence that is available in this scenario, showing that what we call M0 or nonmetastatic unconventional imaging is actually often positive on PSMA PET as we already discussed. And also the importance of oligometastatic disease, there's a lot of interest in this space at this time.

The number of clinical trials that have been done or are ongoing, some of them are mentioned here, such as STOMP trial, which was actually based on choline, POPSTAR, which was based on sodium fluoride, but there are other trials, including the ORIOLE trial, which is based on PSMA PET. And it shows that if you detect these patients with what's considered to be relatively indolent metastatic disease, limited metastatic disease, who may be candidates for metastatic-directed therapy, if they're detected by more sensitive imaging studies, such as PSMA PET, the intervention can be done earlier. And in fact, it turns out that, for example, in the ORIOLE study, the progression-free survival at 6 months was much better with patients who had PSMA compared to conventional imaging. So because that is the available data or evidence, this was given a score of 7.

Scenario number 10, these are patients with post-treatment PSA rise in an already-known metastatic castration-resistent prostate cancer setting. This was given a score of 6 in the middle range or maybe appropriate. Again, just similar to what we talked about before with another scenario, it was unclear immediately of why PSMA PET should be done in this case, because we already know the patients have castration-resistent prostate cancer. And how would the PSMA PET information impact management at this time? However, again, if the radioligand therapy with PSMA becomes approved, for example, based on the VISION trial that was recently published in the New England Journal of Medicine or a trial before that, which was the TheraP trial, comparing cabazitaxel chemotherapy with PSMA radioligand therapy out of Australia, when the PSMA radioligand therapy becomes approved, then there may be a reason to have this patient imaged with PSMA PET, again, to document that there is sufficient PSMA expression before this patient undergoes PSMA radioligand therapy. And again, all of these AUC documents are monitored and revised as new developments occur, at least once a year. So this will be revised when the PSMA SRT becomes approved.

And the last scenario that we'll talk about is scenario number 11, again, under metastatic disease phase of the disease. This was for evaluation of response to treatment. This was given a score of 5 or maybe appropriate. Again, there is not sufficient data in this area with regard to use of PSMA PET in evaluating response. Part of it is that in some cases we don't know exactly how different treatments, including the androgen axis targeting drugs or chemotherapy drugs or other novel drugs, immunotherapy drugs, for example, will affect the PSMA expression. We need more data on that.

We know that for example, androgen deprivation therapy can have valuable effects on PSMA expression, depending upon if you are looking at the short-term ADT or long-term ADT and also the phase of the disease. Is it castration-sensitive or castration-resistant? So because of this kind of lack of knowledge or sufficient evidence in this space, it was decided that at this time, it should be given a score of 5. Of course, we know that the PSMA PET can be useful in assessing response to PSMA radioligand therapy. And that's the reason that the score was in the midrange, not in the rarely appropriate range. But again, this will be revised as we learn more about the impact of different treatments on PSMA expression.

And with that, I end my talk. And thank you very much again for the invitation. Thank you.

Phillip Koo: Great. Thank you very much, Hossein. Clearly, a lot of thought put into pretty much all the clinical scenarios in which PSMA would be considered and a very comprehensive summary of that. We'll have a link to the formal AUC on this site. Just a question that I had. AUCs and this sort of multidisciplinary group of people that came together to develop this, I think, is wonderful. And clearly moving forward, we need to maintain that multidisciplinary approach just because the perspectives are so different. And together, I think we can come up with better solutions. Can you give us a little glimpse with regards to the discussions? And what were the areas of controversy that you noticed with regards to appropriateness?

Hossein Jadvar: Yes. Very good question. Yes, there were some. For example, I mentioned one of them regarding the PSA threshold. There were a lot of questions. You get that question yourself from your colleague, when should I send a patient, for example, for imaging with PSMA? What PSA threshold should I use? And we, in fact, maybe spent an hour just talking on that with colleagues, different experts. And some people said that we should probably use the PSA threshold because that will guide the clinicians. And some said that we shouldn't do that because we know for a fact, based on the literature, that there are patients who can have a positive disease, a substantial number of patients, maybe 10% or 20% of them, who can have substantial disease detected on PSMA PET, even below what is considered to be biochemical recurrence based on the current data. And then there was a discussion about, should we actually change those criteria? Should AUA or ASTROPhoenix criteria be revised? So we even discussed that.

But this was actually a very interesting discussion. I really enjoyed it because we discussed many other things, like PSA genetics, how that may play a role, like doubling time and other things. We even talked about genetics. We talked about liquid biopsy results, which may be playing a role in this. Again, it was a very interesting discussion among these colleagues. And there were many others, but this was one of the major ones that I can recall. And the final decision was we won't use the PSA threshold, and it should be based upon a totality of factors when the PSMA is ordered.

Phillip Koo: Yeah. I think that's a great decision. Oftentimes, those cutoffs or thresholds are often too limiting. And you're right, you need to take a totality of factors into consideration. I also love the fact that you guys address the limitations with the Phoenix criteria of biochemical recurrence. Clearly, nadir plus 2 is something that won't work moving forward, given the improvement in technologies. So just any last comments you have with regards to the appropriate use criteria that you want the readers or the listeners to walk away with.

Hossein Jadvar: Yes. So, as I mentioned, these AUCs are dynamic documents, so we will be revising this. In fact, it's already happening. We are discussing with an anticipation of the approval of the Lutetium-177-PSMA-617 by the FDA. We are already working on this so that we can have these AUCs revised almost immediately upon the announcement of the FDA approval. And then just look out for other AUCs. The SNMMI is working hard to provide these very useful documents to the community. And again, these are documents that will be revised at least on an annual basis as the data is accumulated in different areas.

Phillip Koo: Wonderful. Well, thank you very much, Dr. Jadvar, for joining us. We look forward to future iterations of this AUC. And just thank you for your service and work for all of us.

Hossein Jadvar: Thank you so very much. Take care.

Phillip Koo: Bye-bye.

Hossein Jadvar: Bye-bye.