The Impact of PSMA PET–Based Eligibility Criteria Used in the Prospective Phase II TheraP Trial in Metastatic Castration-Resistant Prostate Cancer Patients Undergoing Prostate-Specific Membrane Antigen–Targeted Radioligand Therapy

Prostate-specific membrane antigen (PSMA) radioligand therapy (RLT) has shown encouraging results for treatment of metastatic castration-resistant prostate cancer (mCRPC) in the prospective, multicenter, randomized phase II TheraP study. The inclusion criteria for that study comprised a pretherapeutic 68Ga-PSMA-11 PET scan showing sufficient tumor uptake using a predefined threshold and the absence of 18F-FDG–positive, PSMA ligand–negative tumor lesions. However, the prognostic value of these PET-based inclusion criteria remains unclear. Therefore, we evaluated the outcome of mCRPC patients treated with PSMA RLT using TheraP as well as other TheraP-based PET inclusion criteria. Methods: First, patients were dichotomized into 2 groups whose PSMA PET scans did (TheraP contrast-enhanced PSMA cePSMA PET–positive) or did not (TheraP cePSMA PET–negative) fulfill the inclusion criteria of TheraP. Notably, unlike in TheraP, 18F-FDG PET was not performed on our patients. Prostate-specific antigen (PSA) response (PSA decline ≥ 50% from baseline), PSA progression-free survival, and overall survival (OS) were compared. Additionally, patients were further dichotomized according to predefined SUVmax thresholds different from those used in TheraP to analyze their potential impact on outcome as well. Results: In total, 107 mCRPC patients were included in this analysis (TheraP cePSMA PET–positive, n = 77; TheraP cePSMA PET–negative, n = 30). PSA response rates were higher in TheraP cePSMA PET–positive patients than in TheraP cePSMA PET–negative patients (54.5% vs. 20%, respectively; P = 0.0012). The median PSA progression-free survival (P = 0.007) and OS (P = 0.0007) of patients were significantly longer in the TheraP cePSMA PET–positive group than in the TheraP cePSMA PET–negative group. Moreover, being in the TheraP cePSMA PET–positive group was identified as a significant prognosticator of longer OS (P = 0.003). The application of different SUVmax thresholds for a single hottest lesion demonstrated no influence on outcome in patients eligible for PSMA RLT. Conclusion: Patient selection for PSMA RLT according to the inclusion criteria of TheraP led to a better treatment response and outcome in our preselected patient cohort. However, a relevant number of patients not fulfilling these criteria also showed substantial rates of response.