Introduction: The EAU prostate cancer (PCa) guidelines introduced risk groups to stratify men with biochemical recurrence (BCR) after surgery. However, they do not account for PSMA-PET, which is recommended in this setting. We aimed to assess the rates of positive PET after BCR and to evaluate the prognostic role of BCR risk groups. Methods: Overall, we identified 299 men restaged with PSMA-PET at PSA persistence or BCR after radical prostatectomy (RP) between 2016 and 2022 with complete BCR risk groups information. Patients were restaged during follow-up according to the judgement of the treating physician. Clinical recurrence (CR) was defined as any new metastases after PET. Kaplan-Meier analyses assessed the impact of the risk groups vs. PET (negative vs. local recurrence, prostatic fossa and pelvis, vs. distant recurrence) on CR-free survival. The accuracy of multivariable models including risk groups vs. PET was assessed using the c-index after accounting for salvage therapies. Decision-curve analyses assessed the net benefit associated with models including EAU BCR risk groups vs. PET. Results: Overall, 58 (19.4%) vs. 241 (80.6%) patients were low- vs. high-risk according to the BCR risk groups and 199 (66.6%) patients had a positive PET. Median PSA at PET was 0.7 ng/ml (0.5 vs. 0.7 ng/ml for low vs. high-risk). Median PSA doubling time was 8 months. Overall, 57 vs. 21% patients had a negative vs. positive PET, 24 vs. 12% received salvage radiotherapy, 0 vs. 43% received MDT and 19 vs. 23% received systemic therapies in low and high-risk groups. Overall, 67.2% vs. 66.4% patients had a positive PSMA-PET in low vs. high-risk (p=0.5). The rate of local and distant spots at PET was 29.3 and 37.9% vs. 22.2 and 43.6% in low vs. high-risk patients (p=0.5). Median follow-up was 16 months and 87 patients had CR. No differences were detected in 2-year CR-free survival according to risk groups (58 vs. 65%, p=0.6). When patients were stratified according to PET, the 2-year CR-free survival rates were 75.2 vs. 54.3% (p < 0.01). A positive PET (HR: 2.3; p=0.01) but not the BCR risk groups (p=0.8) was associated with CR. The inclusion of PET improved the discrimination of models predicting CR (53 vs. 65%) with a higher net benefit. Conclusions: In patients staged with PSMA-PET, the EAU risk groups lose their prognostic importance. More than 60% of men with low risk BCR had positive PET which was significantly associated with higher risk of CR. When imaging is available, PET overrules the prognostic impact of clinical risk stratification of BCR. SOURCE OF Funding: None
