PSMA-PET identifies PCWG3 target populations with high concordance however superior reproducibility when compared to conventional imaging

1260 Objectives: Most clinical trials on castration resistant prostate cancer (CRPC) rely on baseline staging by conventional imaging rather than novel positron-emission-tomography (PET). We investigated the impact of PSMA-PET on PCWG3 clinical subtype classification, frequently used to define clinical trial target populations. Methods: Databases at three participating PET centers were retrospectively screened for patients with prostate cancer, for whom PSMA-PET was performed between 2014 and 2019, and who had: (a) documented CRPC, (b) PSA values ≥1 ng/mL and (c) conventional imaging (CI) i.e. bone scan and CT (78%) or whole-body MRI (22% of patients) within 6 months of the PSMA-PET without changes of therapy between the staging modalities. Clinical PCWG3 subtype, i.e. nonmetastatic (nmCRPC), locally recurrent, nodal spread, bone disease and visceral disease, as well as PROMISE stage were determined for PET versus CI by three independent blinded readers. Inter-reader agreement was assessed by Fleiss’ Kappa. Results: 67 patients were included with a median PSA level of 53.2 ng/mL (IQR 5.8-334.6 ng/mL). Median time between PSMA-PET and CI was 1 month (IQR 0-2). 37/67 patients (55%) underwent ADT plus at least one line of a second line treatment for CRPC (docetaxel, cabazitaxel, abiraterone, enzalutamide or Radium-223). Overall CI was positive in 87% (58/67) of patients whereas PSMA-PET in 92% (62/67). PCWG3 clinical subtypes identified by PSMA-PET vs. CI were: 7% (5/67) vs. 13% (9/67) for nmCRPC, 4% (3/67) vs. 0% (0/67) for locally recurrent CRPC, 10% (7/67) vs. 6% (4/67) for nodal spread, 60% (40/67) vs. 58% (39/67) for bone disease and 18% (12/67) vs. 22% (15/67) for visceral disease, respectively. Overall, PSMA-PET and CI were concordant in 70% (47/67) of patients. PSMA-PET led to up-staging in 15% (10/67) and down-staging in 15% (10/67) of patients; down-staging occurred most often for visceral disease (7/67, 10%) (Table 1). PSA at the time of PET, D’Amico risk group and number of different systemic therapies were not significantly associated with a PSMA-PET up-staging compared to CI. There was a statistically significant association between higher D’Amico risk group and a PSMA-PET down-staging compared to CI (p=0.003). PET vs. CI interobserver agreement was 0.81 vs. 0.51, 0.74 vs. 0.47, 0.95 vs. 0.72, or 0.58 vs. 0.65 for locally recurrent, nodal spread, bone disease, or visceral disease, respectively. Conclusions: In our multicenter, retrospective study, PSMA-PET demonstrates high level of concordance with CI for per-patient PCWG3 clinical subtype. PSMA-PET delivers higher reproducibility and more accurate assessment of nmCRPC as well as visceral disease and should be implemented in future enrollment/screening procedures. $$table_{5171EA87-920E-4F20-902F-8532986F59A4}$$