A Multicenter, Prospective Study on the Prognostic Value of PSMA PET in Patients With Newly Diagnosed, Treatment-naïve Prostate Cancer

Background Prostate-specific membrane antigen (PSMA) positron emission tomography (PET) has fundamentally transformed the staging paradigm for prostate cancer, demonstrating superior sensitivity and specificity compared to conventional imaging for the detection of metastatic disease . The Prostate Cancer Molecular Imaging Standardized Evaluation (PROMISE) criteria were established to standardize PSMA PET reporting, enabling reproducible and anatomically consistent disease characterization . Recent large-scale, international multi-centre studies have validated the robust prognostic value of PSMA-PET PROMISE (PPP) nomograms. These tools have shown superior accuracy in stratifying risk for overall survival across all disease stages when compared to established clinical risk tools such as the NCCN and EAU risk scores . The quantitative and visual PPP nomograms, incorporating parameters such as distant metastatic burden (M1a, M1b, M1c), total tumour volume, and PSMA expression score, have achieved C-indices of 0.77-0.80 for predicting overall survival, confirming the powerful prognostic information embedded within PSMA PET imaging . Need While the prognostic utility of PSMA PET is established, previous studies, including the foundational PPP nomogram development, have largely enrolled cohorts comprising patients with mixed disease states-including those at initial staging, biochemical recurrence, and metastatic castration-resistant prostate cancer . This heterogeneity represents a significant confounder. The inclusion of patients with varying prior treatment exposures (e.g., post-radical prostatectomy, post-radiotherapy, post-systemic therapy) and at disparate points in their disease trajectory may obscure the true baseline prognostic power of the initial staging scan . A critical evidence gap remains regarding the specific prognostic value of the first, pre-treatment PSMA PET scan in a uniformly treatment-naïve population undergoing initial staging. Existing data on the direct correlation between baseline quantitative PSMA PET parameters and long-term outcomes like progression-free survival (PFS) in this specific, homogeneous cohort are lacking. As noted in recent literature, while PSMA PET provides unprecedented risk stratification, its direct impact on patient outcomes and the optimal integration of its quantitative metrics into prognostic tools for untreated patients require further prospective validation . Aim / Objective Primary Outcomes: The primary aim of this study is to prospectively investigate the prognostic value of baseline staging PSMA PET for predicting progression-free survival (PFS) in patients with newly diagnosed, histologically confirmed, treatment-naïve prostate cancer. We will develop and validate a dedicated prognostic tool (nomogram or risk model) incorporating baseline PSMA PET parameters (e.g., SUVmax, lesion count, metastatic stage \[miTNM\]) alongside standard clinico-pathological variables to stratify patients by risk of disease progression. Secondary Outcomes: 1. Comparison with Established Prognostic Tools: To compare the prognostic performance of the PSMA PET-based model (PSMA-VISION score) for PFS against established risk stratification systems, including NCCN risk categories and the PSMA PET Prognosis (PPP) nomogram. 2. Correlation with Clinicopathological Variables: To assess the association between PSMA PET parameters (e.g., miTNM stage, total tumor volume, PSMA-VISION score) and standard clinical variables (Gleason grade group, baseline PSA, ISUP risk category). 3. Prediction of Early Progression (Exploratory): To evaluate the ability of baseline PSMA PET parameters to predict early disease progression, defined as PFS event occurring within 12 months of diagnosis. 4. Exploratory Analysis of Overall Survival: To explore the association between baseline PSMA PET parameters and overall survival (OS), acknowledging the limited 2-year follow-up duration. Inclusion: 1. Patients with pathologically confirmed prostate cancer by prostate biopsy. 2. Undergone PSMA PET examination as initial staging before any treatment. 3. Have at least 2 years of follow-up data for progression-free survival and overall survival. Exclusion Criteria: 1. Patients undergone any prior prostate cancer treatment before undergoing PSMA PET imaging. 2. Patients with metastasized or disseminated malignancy other than prostate cancer 3. Patients with neuroendocrine prostate cancer. Statistical considerations: This is an observational cohort study leveraging prospectively collected data from various centers. The sample size is determined by the availability of eligible patients meeting the inclusion criteria during the study period, with a planned enrollment of approximately 1000 patients. Given the exploratory nature of this study and the absence of pre-specified effect sizes, a formal sample size calculation was not performed. However, with a minimum follow-up of 2 years for progression-free survival (PFS), we anticipate a sufficient number of PFS events to enable multivariable regression analyses. Based on the 10 events per variable (EPV) rule of thumb , this sample size will support the inclusion of 10 key predictor variables in the final model. The prognostic value of baseline PSMA PET parameters (including PSMA-VISION score, miTNM stage, and SUV metrics) for PFS will be assessed using Cox proportional hazards regression models. Hazard ratios (HR) with 95% confidence intervals (CI) will be reported. The proportional hazards assumption will be tested using Schoenfeld residuals. The predictive performance of the PSMA-VISION score will be evaluated using Harrell's C-index for time-to-event data. The C-index quantifies the model's ability to discriminate between patients with different PFS outcomes. To compare the prognostic performance of the PSMA-VISION score against established risk tools (NCCN risk categories, PPP nomogram), we will calculate the difference in C-indices with 95% CIs, using bootstrap resampling (1000 replicates) to account for over-optimism. For the secondary endpoint comparing PSMA-VISION with other prognostic tools, time-dependent receiver operating characteristic (ROC) curve analysis will be performed, and the area under the curve (AUC) at 12 and 24 months will be calculated and compared using the method of DeLong et al. As this is an open cohort study, continuous data accrual will improve the precision of prognostic estimates. All statistical tests will be two-sided, with p \< 0.05 considered statistically significant. Analyses will be performed using R software (version 4.x) with relevant packages (e.g., survival, timeROC, compareC). Central Database: Data will be stored centrally in a RedCap Database with 3-step authentication at the sponsor site. Recurring Data Entry: Data entry will be conducted repeatedly at about 3 to 6 month intervals.