Introduction: The prostate-specific membrane antigen positron-emission-tomography (PSMA-PET) scan is increasingly being utilized for staging prostate cancer (PCa). Here, we sought to evaluate the accuracy of the PSMA-PET scan with the aim of identifying patients who could be spared a pelvic lymph node dissection (PLND) at the time of radical prostatectomy (RP) based on the results of the PSMA-PET scan. Methods: We retrospectively reviewed records of patients who underwent a PSMA-PET scan followed by an RP with extended PLND (external iliac, obturator and common iliac nodes) at our tertiary cancer care centre from Aug 2015 - September 2022. Patients were classified as clinically node negative or positive (cN0 or cN+) based on the PSMA PET scan. The nodal status in the PSMA-PET scan was compared with the final histopathology of the PLND (pN0 or pN+). Results: A total of 205 patients were included and were stratified as follows based on MRI and PSMA-PET scans: 6 (2.9%) low-risk, 49 (23.9%) intermediate-risk, 46 (22.4%) high-risk localized, 89 (43.4%) locally-advanced node-negative and 15 (7.3%) node-positive patients. The sensitivity, specificity, positive predictive value, and negative predictive value of the PSMA PET scan for nodal status were 37.5%, 98.2%, 80% and 89.4% respectively. The NPV was 89.4%, 93.8%, 87% and 86.6% for overall, intermediate-risk, high-risk localized and locally-advanced node-negative patients respectively. The median SUVmax of the primary prostate lesion in patients who were true negatives for nodes, as identified by the PSMA-PET scan, was significantly lower (12.8 vs 26.4, p<0.001) than that in patients who were false negative (cN0 on PSMA but pN+ in final pathology). Of the 190 cN0 patients, 20 had a false negative PSMA PET scan. Three of the 49 intermediate-risk patients had a false negative PSMA PET scan for nodes; the SUVmax scores of their prostate lesions were 26.59, 40.06 and 54.55 which were much higher than the median SUVmax score (12.8) in patients with a true negative scan. For the 135 high-risk cN0 patients, the 17 false negative patients had a median SUVmax of 24.37 compared to those with a true negative scan (13.57), p<0.001. Conclusions: The high NPV of PSMA-PET scan coupled with the SUVmax of the primary prostatic lesion can help avoid a PLND in intermediate and high-risk patients with a ‘negative’ PSMA-PET scan. The results can be considered as hypothesis generating and indicate the use of PSMA as a biomarker for predicting nodal disease. Further larger studies can help determine the ideal cut-off values for the SUVmax of the primary lesion. SOURCE OF Funding: No
