Chris Wallis: Hello, and thank you for joining us for this UroToday Journal Club discussion. Today, we're talking about a publication entitled Predictors of 18 Fluoride-DCFPyL PET/CT Positivity in Patients with Biochemical Recurrence of Prostate Cancer After Local Therapy. I'm Chris Wallis, Assistant Professor in the Division of Urology at the University of Toronto. With me today is Zach Klaassen, Assistant Professor in the Division of Urology at the Medical College of Georgia. You can see here the citation of this recent publication published in JNM. Biochemical recurrence is relatively common for patients who have local therapy of prostate cancer. Depending on initial disease risk, it may affect 20 to 40% of men in the 10 years following their local treatment. For men with biochemical recurrence, understanding both the specifics of relapse and the patterns of recurrence may improve an understanding of their disease process and allow targeted tailored treatment approaches. Conventional imaging in general performs quite poorly for allowing us to identify sites of relapse in biochemical recurrence, particularly when patients have PSAs less than two. Novel PET radiotracers have improved sensitivity and specificity for the detection of disease and disease setting.

We see here that DCFPyL is a small molecule that binds to the extracellular domain of a PSMA. In both CONDOR and OSPREY trials that led to its FDA approval, DCFPyL has demonstrated a good performance in patients with biochemical recurrence in initial staging in high risk disease and in detecting suspected recurrent lesions that are seen on conventional imaging. And so there's a wide variety of clinical scenarios where this approach may be beneficial. In this study, the authors use 2 institutional trials incorporating 147 patients from the National Cancer Institute and 98 patients from Johns Hopkins. They included patients who had biochemical recurrence defined either as a PSA greater than 0.2 following a radical prostatectomy or using the Phoenix definition of a nadir +2 for those who had undergone radiotherapy. They could also have evidence of clinical failure in the absence of this biochemical definition but had to have negative conventional imaging. Patients were excluded if they were currently receiving Androgen deprivation therapy, were unable to tolerate PET/CT, or had a creatinine more than two times the upper limit of normal.

A standard 18 Fluoride-DCFPyL PET/CT was performed. The median dose was 296 MBq or 8 millicuries. A low dose CT was performed and PET imaging was performed one to two hours following the injection of the radiotracer. Images were interpreted by two board certified nuclear medicine physicians and any disagreements were resolved by consensus. They required a "clear foci" of disease to be considered positive and indeterminate lesions were considered negative. Detection rates were presented as a function of PSA levels at the time of imaging. This was stratified as PSA less than 0.5, 0.5 to less than 1, 1 to less than 2, 2 to less than 5, and greater than, equal to 5 nanograms per milliliter. Sites of recurrence were tested for the correlation between PSA, PSA doubling time, and the time from local therapy. They further performed subgroup analyses according to the treatment type, dividing those patients who'd received surgery from those who'd received radiotherapy. They further categorize patients according to their initial Gleason score and their volume of disease, including oligometastatic defined as one to five lesions versus polymetastatic greater than five lesions.

The authors assessed the association between scan results and they characterized this as negative scans, positive for intrapelvic disease, either local recurrence or pelvic nodal disease, and positive for extrapelvic disease, whether extrapelvic nodal disease, boney disease, or other sites of visceral disease. And they assessed that the associations between these results and clinical factors including Gleason score, PSA level, PSA doubling time, the year since treatment, and treatment type. Variable selection was performed using the AIC, the model fit was assessed with calibration plots, the combined predictive tool was then assessed for its diagnostic accuracy in the area under a receiver operating curve. Now, going to hand over to Zach to walk us through the results of this study.

Zach Klaassen: Thanks, Chris. So this is the patient characteristics table for this study. There was 245 patients included, 195 patients had radical prostatectomy, plus or minus radiotherapy, plus or minus ADT, and 50 patients had radiotherapy plus or minus ADT. The mean age was 66 years. As you can see, the primary Gleason score was most commonly 7, including 120 patients, and the median PSA for the study was 1.6 with a range of 0.2 to 35.5, and the median PSA doubling time was 6.8 months with a range of 0.9 to 75.2 months. This figure looks at DCFPyL overall, intrapelvic, and extrapelvic detection rates by PSA, and we'll walk through this sequentially. So starting with the overall detection rate, which is in the dark blue, we see that this goes from 48.2% at a PSA of 0.2 to 0.5 all the way up to over 90% for patients with a PSA of greater than 2.

In patients with intrapelvic disease, which is light blue bars, we can see that this is even at PSA 0.2 to 0.5, 32.1%, up to 63.6% for a PSA of 1 to 2, and greater than 50% for a PSA greater than 2, and 38.8% for a PSA greater than 5. In terms of extrapelvic disease, which is the gray bars, even at a PSA of 0.2 to 0.5, there's 16.1% detection of extrapelvic disease all the way up to 53.1% in patients with a PSA of greater than or equal to 5. This figure looks at the DCFPyL positive sites by location and PSA, and we'll also go through this figure sequentially, looking first at the negative PSMA PET/CT scan. So at a PSA of 0.2 to less than 5, we see that approximately 50% of these studies were negative, which decreases to less than 10% for a PSA greater than or equal to 5. Looking at local recurrence, which is in the light blue bar, we see that this is roughly 10% in patients with a low PSA and essentially moves up to roughly 35% with a PSA of 1 to 2.

With regards to pelvic lymph nodes, we see that roughly 25% of these patients at all these PSA levels have pelvic lymph nodes detected, and roughly 10 to 20% of patients with retroperitoneal or distant lymph nodes are positive regardless of PSA. Where it gets interesting as we see that bone metastatic disease, even at a very low detectable PSA of 0.2 to 0.5, roughly 8 to 10% of patients have positive bone metastases upwards to 20 to 25% in PSAs over 2. And with visceral disease, again, not common in patients with a PSA less than 2, but roughly 5 to 12% in patients with a PSA greater than 2. This looks at the disease location versus time from prostatectomy and radiation. So we'll look at prostatectomy first, which is at the top of this slide, and you can see here that within the first 8 to 10 years, the majority of these patients will have detectable disease, whether localized or all the way up to visceral metastases. Certainly, even after 10 years, we see that some patients with localized and pelvic nodal disease will have detectable disease on PSMA PET scan.

With regards to radiotherapy, again, a smaller sample size of 50 patients, we see a similar pattern with the majority of these patients having detectable disease on PSMA PET within 10 years of treatment. This figure looks at patients on the recurrence versus time from therapy and PSA doubling time. On the left, we look at time after radical prostatectomy in figure A and time after radiotherapy in figure C. Again, very similar to the last figure where we see that the majority of patients will have their disease detected within 10 years, with a few having disease detected, particularly localized in nodal disease, after 10 years. With regards to PSA doubling time, no surprise that within roughly 10 patients with a PSA doubling time of 10 months, the majority of metastatic nodal and localized disease will be detected whether patients had a prostatectomy or radiotherapy. The next couple slides, we'll look at some patient examples from this study. This is the first patient. This is a 70-year-old man with biochemical recurrence after radiotherapy in 2 years of ADT. His primary tumor was T4N1, Gleason 4+5.

His time from treatment was 2.5 years and his pre scan PSA was 2.41 with the fast PSA doubling time of 4.7 months. As you can see in the images on the left, the DCFPyL PET/CT scan showed sub-centimeter pelvic nodes as well, as a foci at T9, as well as a anterior iliac bone lesion, which did demonstrate on biopsy to confirm metastases in the iliac crest. The second patients is a 63-year-old man with biochemical recurrence who did reach an undetectable PSA after prostatectomy and 2 years of ADT. His primary tumor was T3BN1, Gleason 4+5. His PSA started to rise 5 years after initial treatment to a PSA of 0.3 and a PSA doubling time of 3 months. As you can see on the left here, his DCFPyL PET/CT showed focal uptake in a 0.5 centimeter left common iliac node. However, a biopsy in this patient was not performed. It's important to note on this patient that even at a PSA of 0.3, there's a positive signal for DCFPyL in the common iliac lymph nodes for this patient. These are the nomograms predicting likelihood of positive imaging based on patients that had surgery or radiation.

On the left is the nomogram for the surgery cohort, which included Gleason score, PSA, and PSA doubling time, and the AUC for scan positivity overall with 78% and for extrapelvic disease was 76%. On the right, this is the nomogram for the radiation cohort. This nomogram includes PSA and PSA doubling time, and the AUC for extrapelvic disease in this nomogram was 85%. Several discussion points from this study. This study demonstrated that 18F-DCFPyL PET/CT detects lesions in most biochemical recurrent patients. Patients with prostatectomy-bed-only recurrence is associated with the longest duration, with a mean of 7.4 years from treatment, representing the least aggressive disease trajectory. Pelvic nodal involvement is associated with a shorter duration, 3.1 years from treatment, suggesting a more aggressive disease trajectory. And finally, bone and visceral involvement manifests even earlier, roughly 2.3 years after prostatectomy, with similar patterns in the post-radiation cohort. The likelihood of having a scan with extrapelvic lesions was determined in surgical and post-radiation patients using nomograms, as we saw on the previous slide, with the most relevant predictors for both the surgery and the radiation nomogram being PSA and PSA doubling time.

PSA in the post-treatment setting is a reliable marker of disease volume and PSMA PET/CT demonstrates a relationship between PSA and extensive disease with ascending percentages of positive scans with higher PSAs. The overall DCFPyL PET/CT positivity rate in this study was 79.2% with a detection rate of 48.2% at a PSA of less than 0.5. Finally, DCFPyL positive pelvic lesions were present in 48% of patients whereas extrapelvic disease was found in 30% and almost exclusively when patients had a PSA of more than 1. In conclusion, 18F-DCFPyL PET/CT offers a high detection rate in biochemical recurrence among prostate cancer patients. PSA and PSA doubling time are able to predict scan positivity and disease location. The presence of bone or visceral lesions is associated with shorter intervals from treatment than are prostate-bed-only recurrences. And finally, and importantly, these tools may guide clinicians to select the most suitable candidates for 18F-DCFPyL PET/CT imaging. We thank you very much for your attention and we hope you enjoyed this URO Today Journal Club discussion.