Role of PSMA PET in CRPC Patients "Presentation" - Wolfgang Fendler

February 12, 2024

At the 2024 UCSF-UCLA PSMA Conference, Wolfgang Fendler emphasizes the transformative role of PSMA-PET imaging in managing castration-resistant prostate cancer (CRPC), showcasing its ability to reveal a more extensive disease burden and detect distant metastases in patients previously considered non-metastatic. Dr. Fendler highlights a multicenter trial that demonstrates PSMA-PET's significant impact on upstaging nmCRPC patients, linking PSMA-PET findings to overall survival, and introduces a nomogram that uses PSMA-PET parameters for better prognostication and treatment planning, advocating for its broader integration into clinical practice.

UCLAUCSF video logo stroke


Wolfgang Fendler, MD, Department of Nuclear Medicine, University of Duisburg-Essen and German Cancer Consortium (DKTK), University Hospital Essen, Essen, Germany

Read the Full Video Transcript

Wolfgang Fendler: So moving along. Hello, everyone again. My talk will focus on the broad group of patients with castration-resistant prostate cancer and the role of PSMA-PET in these patients. Again, disclosures, which you have seen already before. We are now moving in the lines of progression of prostate cancer from the early lines of local disease, biochemical recurrence, to the more advanced stage diseases. This is characterized by waves of progression, and one of the interim progression waves is the so-called nmCRPC cohort. Currently described by patients that have a rising PSA of more than two, but conventional imaging is not showing any disease. These would be characterized as castration-resistant but non-metastatic to conventional imaging, and then of course, the metastatic CRPC patient cohort and the more advanced disease patient cohort. And we will talk about these more advanced stage disease patients starting from the nmCRPC.

All these have in common, again, similar to the initial staging to the earlier stages of prostate cancer, that the addition of PSMA-PET will show a much higher volume and much higher load of the disease. So we see much more than what is here compared to the iceberg which is under the surface, not only the peak of disease, peak of the iceberg. Of course, we do not see the entire disease burden yet, sensitivity is not optimal in all stages, but we see much more. This leads to an upstaging and, of course, to a different viewpoint, to a different assessment of patients with castration-resistant prostate cancer.

And we tried to assess this in a trial that compared conventional imaging to PSMA-PET focused on the so-called CRPC staging groups that are proposed in the prostate cancer working group 3 criteria, which is immensely important to defining your CRPC population for clinical trials. And we compared patients that received both imaging modalities, conventional imaging, which would be CT plus bone scan, or whole body MRI, to head-to-head results from the PSMA-11 PET. In total, 67 patients have been enrolled, and we looked at these categories by prostate cancer working group, which is nmCRPC, then CRPC local disease, nodal disease only, bone disease, or patients with also visceral disease.

And if you go through these different categories, you see that patients that are outstaged by conventional imaging CRPC, they undergo major upstaging. So basically, the conventional nmCRPC does not exist by the addition of PSMA-11 PET anymore. You have 67% upstaging in these patients to nodal or bone disease. Patients with local disease, there have been none in the trial. Nodal disease, very few patients have been enrolled. Patients with conventional imaging bone only disease, they did not undergo a major change. There was only 8% upstaging and 8% downstaging by the addition of PSMA-PET, so conventional imaging seems to be fairly accurate in these patients. And then interestingly, in patients that had, by conventional imaging, visceral metastasis, they have been downstaged in nearly every second patient. So there's a lot of correction of potential false positive findings in patients with conventional imaging lesions in the liver, for example, or other organs, that by PSMA-PET would not be visceral disease.

So major changes, and as a summary, major changes by upstaging in patients with the so-called nmCRPC group. We see major downstaging in patients with conventional organ, potential organ involvement, that are actually not confirmed on PSMA-PET, and we see a high consistency in patients with bone disease of PSMA-PET as compared to conventional imaging. So looking more closely into this interesting intermediate stage patients of the so-called conventional non-metastatic castration-resistant prostate cancer. This is a multicenter trial where we retrospectively try to find out what is the impact of PSMA-PET if this is being applied in patients with nmCRPC and high-risk features. So either Gleason score 8 or higher, or doubling time which was lower than 10.

You can see that the addition of PSMA-PET led to disease detection in nearly all patients, and more strikingly, PSMA-PET led to the detection of distant metastasis in more than every second patient. 55% have been considered metastatic, so a major upstaging in 55% of patients, and in total, a change in how we perceive and how we evaluate this disease population. The other half, 44%, had local or pelvic only disease. And then in a recent update, we wanted to find out if there is really a clinical impact relevance of our PET findings. So we followed these 200 patients by their overall survival and assessed here by logistic regression, is there an association between PET findings and clinical findings and overall survival?

And these are the Kaplan-Meier curves of positive findings. So polymetastatic disease by PSMA-PET, meaning five lesions or more, five distant lesions or more, was significantly associated with shorter overall survival. Then a higher PSMA-PET avidity, so an SUV max of 8.4 or higher, was significantly associated with overall survival, probably associated just with the disease load, the disease volume of patients. And then also a clinical factor was associated, which is the pathologic N1 stage at initial diagnosis, was associated with shorter overall survival in these patients. So there is a clinical relevance of our findings, but of course, it's not just detection of any distant disease, it needs to be a fair amount of distant disease here defined by five lesions or more detected by PSMA-PET. And these findings have just been accepted in European Urology as of this morning for publication, so these should be online within the next weeks.

So the summary for the impact of PSMA-PET on the so-called nmCRPC group is that PSMA-PET detects disease in almost all of these patients, detects distant disease in 55% of these patients, and that we also know that PSMA-PET detection by polymetastatic disease or high PSMA SUV are prognostic; they are associated with overall survival in these patients. And to put it simply, in a scheme, if you image patients with nmCRPC with PSMA-PET, these will be separated into about half of the patients that are then metastatic. These patients can have a poor outcome depending on your pattern of findings, but they are still candidates for the current nmCRPC treatment agent landscape because all of these treatment agents have been evaluated based on conventional imaging. That's important to keep in mind. And then half of the population will have local regional disease, and of course, the question is, can we in the future maybe achieve any delay, disease delay, or any delay in further agents by the addition of local therapy? Again, this is an open question for research.

Then I would like to talk about the volume of disease assessments and assessment of disease load. Of course, high versus low volume of disease has been defined by the CHAARTED criteria in the CHAARTED trial for the so-called metastatic hormone-sensitive prostate cancer population, but I still like to talk about this assessment because I think it's also important for any patient with metastatic disease or any patient in the advanced stage of disease. In this trial, we tried to find out, what is the impact, the perceived high versus low volume disease, in patients by conventional imaging as compared to PSMA-PET?

And we used CT only, not CT plus bone scan, but CT only, to separate these patients into a CHAARTED low volume disease and CHAARTED high volume disease, and then we added onto this assessment by PSMA-PET head-to-head in the same patients. You can see that all the patients with unifocal disease and all the patients with oligometastatic disease also had low volume disease by CT criteria. All the patients with visceral metastasis also had high volume disease by the CT criteria. But there's a large spectrum of patients with disseminated disease on PSMA-PET that could either be in the CHAARTED low volume category or in the CHAARTED high volume disease category. It's a 22% misalignment overall. So we used the total tumor burden, the total disease volume, to be more accurate, and found out that a 40 milliliter disease cutoff by the total tumor volume of PSMA-PET can be used to sub-stratify those patients with disseminated disease into either disseminated compatible with low volume and disseminated compatible with high volume disease to then achieve a 13% misalignment. And we used this 40 milliliter cutoff to define PET volume of disease criteria.

And this is the final definition of these criteria. Unifocal disease, oligometastatic disease, or disseminated disease with a volume of 40 milliliters or lower would be considered low volume disease by PET, and high volume disease is any disease with more than 40 milliliters or visceral organ metastasis. If we apply these criteria to our mHSPC population, you can see there's high consistency in the assessment of low volume disease, but there is some upstaging by PET in terms of upstaging from low volume disease to high volume disease in some patients with a really high tumor load by total tumor volume.

I'd like to show you one example where there was an upstaging by these novel proposed PET criteria: a patient with a Gleason score of 8 and initial PSA of 18. The whole body MRI showed three lesions, including one lymph node metastasis and positive bone lesions. The CT scan in this patient showed only one sclerotic lesion in the pelvis and then the enlarged lymph node. This patient also underwent PSMA-PET, and we saw much more extensive disease, including mediastinal lymph nodes, cervical lymph nodes, and retroperitoneal lymph nodes, which were highly PSMA positive. The tumor load was more than 40 milliliters, so by the PSMA-PET criteria and what we proposed, this would be considered high volume disease by PET, and low volume disease by conventional criteria. Again, this is just a descriptive finding; I cannot prove that this assessment is more accurate, but I still believe that looking at the entire tumor load, this patient should have a much higher risk than the low volume group.

In summary, we can say that PSMA-PET, for volume of disease assessment, leads to a significant upshift in stage into volume of disease. If we add PSMA-PET to conventional imaging in patients with mHSPC, PSMA-PET has huge advantages in detecting unifocal oligometastatic disease but also in differentiating this from disseminated disease in patients with metastatic prostate cancer. Of course, we can measure and quantify the total PSMA-PET positive tumor volume, and we believe that by combining PSMA-PET staging and quantitative tumor volume, we will be able, in the future, to define disease volume criteria that are compatible with what was currently used in the CHAARTED and STAMPEDE trials, but also criteria that might be more accurate in the future, which will require prospective evaluation.

Going on one step further to the mCRPC population and Lutetium PSMA therapy group, I wanted to highlight this trial, a multicenter study that we also participated in from UCLA, UCSF, Essen, but also Australian sites, that assessed the value of PSMA-PET, the prognostic value of PSMA-PET, baseline PET for the outcome under Lutetium PSMA therapy. This trial assessed, in total, more than 20 baseline characteristics of patients and strikingly found that eight characteristics significantly predicted overall survival. These were combined into one nomogram, and of these characteristics, there were four PET characteristics.

SUV mean at baseline, number of lesions, bone metastasis, and presence of liver metastasis have all been independently and with a high level associated with the overall survival under Lutetium PSMA therapy. Thus, PSMA-PET is a predictive or prognostic biomarker at baseline for these patients undergoing PSMA therapy. This nomogram has been proposed, and in my viewpoint, this nomogram, but also findings from the vision trial, are very important for the future, for clinical trial design, but also to support clinical decisions for patients whether to undergo PSMA therapy or for tight monitoring under PSMA therapy. You can see these patients are being separated by the nomogram into low-risk and high-risk patients, with a huge difference in overall survival. The median is 25 months for the low-risk group and 7.4 months for the high-risk group. So, there's a high prognostic value of these PSMA-PET parameters in the nomogram in general.

Of course, we can do more after initiation of PSMA therapy. PSMA-PET can also be used to monitor response under therapy, and this is one of the proposals, also from Gafita, that was derived from the multicenter cohort. The so-called RECIP criteria, Response Evaluation Criteria in PSMA-PET imaging, combine the appearance of new lesions and the change in total disease volume to come up with these three categories: either PSMA-PET partial response, progressive disease, or stable disease. If these two parts, the PSMA volume change and the appearance of a new lesion, are combined into the RECIP categories, you can see that the RECIP framework accurately stratifies those patients who respond well and have a longer overall survival in the partial response group with 22 months overall survival, and those patients with progressive disease and only eight months overall survival. So it's a very good biomarker of response also that might be used in the future to guide treatment decisions and subsequent treatment choice.

So in summary, PSMA-PET is a baseline prognostic biomarker and can also be used to monitor response under Lutetium PSMA therapy. This is also potentially true for other treatments as was shown for the ARPI therapy in patients with mCRPC. There are novel criteria for response assessment, the so-called RECIP criteria, that have been developed in a multicenter cohort and are accurate for response monitoring under Lutetium PSMA therapy. Of course, there's more prospective validation needed, and inclusion of these criteria in prospective large trials for further validation.

So I'd like to summarize. PSMA-PET for advanced prostate cancer. In total, we can see that the PSMA-PET detects more disease, and especially in patients with the so-called conventional nmCRPC, detects distant metastasis in more than every second patient. We have to acknowledge that the current therapy labels, especially for systemic treatment, should not be affected by PSMA-PET in general. Of course, you can alter your treatment decision, but all these agents have been evaluated on conventional imaging and should thus stay as treatment options. PSMA-PET is prognostic if we see more than five lesions.

In addition, PSMA-PET leads to considerable upshift of stage and disease volume in patients with metastatic or the so-called HSPC stage. And in patients with metastatic castration-resistant prostate cancer, PSMA-PET is a gatekeeper of PSMA radioligand therapy, but we can derive more information from PSMA-PET, so the baseline SUV, the baseline disease extent, can be used to predict overall survival, but we can also use PSMA-PET interim staging and for response assessment using novel criteria, such as the RECIP criteria, to guide treatment in the future. And with this, I thank you very much for your attention, and I'm looking forward to our panel discussion.