Enhancing post-HIFU surveillance: Diagnostic performance of 18F-PSMA PET/CT and mpMRI for local prostate cancer recurrence.

Abstract

313 Background: Focal therapy (FT), including high-intensity focused ultrasound (HIFU), has emerged as a minimally invasive alternative to radical prostatectomy or radiotherapy for localized prostate cancer. However, post-treatment surveillance remains challenging, as prostate-specific antigen (PSA) kinetics and multiparametric MRI (mpMRI) interpretation are often confounded by residual tissue and post-ablation changes. Although imaging performance following HIFU remains poorly characterized, prostate-specific membrane antigen (PSMA) PET/CT has demonstrated high sensitivity for detecting prostate cancer recurrence. This study aims to perform a head-to-head comparison of PSMA PET/CT and mpMRI in detecting locally recurrent prostate cancer after HIFU. Methods: This single-institution retrospective study included 30 men with biopsy-proven prostate cancer who underwent HIFU followed by both mpMRI and ^18F-piflufolasat PSMA PET/CT prior to confirmatory prostate biopsy. A board-certified abdominal radiologist reviewed imaging blinded to pathology results. Diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated using biopsy as the reference standard. Proportions were compared using McNemar’s test, and Spearman rank correlation was applied to assess associations between SUVmax, PSA, and ISUP grade groups. Results: Local recurrence was pathologically confirmed in 24 of 30 patients (80%). PSMA PET/CT and mpMRI each detected 18 of 24 biopsy-proven recurrences, corresponding to equivalent sensitivities of 76% (95% CI, 0.60–0.91). Specificity was higher for PSMA PET/CT (83%; 95% CI, 0.44 -- 0.97) than for mpMRI (67%; 95% CI, 0.30 – 0.90). When both modalities were combined, sensitivity increased to 96% (95% CI, 0.80 – 0.99), specificity to 100% (95% CI, 0.61–1.00), and overall accuracy to 96.7% (95% CI, 0.79 –0.98), significantly outperforming either modality alone (p < 0.05). SUVmax demonstrated weak correlation with PSA (ρ = 0.27, p = 0.18) and ISUP grade (ρ = 0.34, p = 0.093). Conclusions: Despite PSMA PET/CT exhibiting higher specificity than mpMRI, both exhibited comparable sensitivity for detecting recurrent prostate cancer following HIFU. Combined imaging significantly improved diagnostic performance, which ultimately minimized false positives and negatives. Dual-modality evaluation may enhance post-treatment surveillance accuracy and guide biopsy decision-making, warranting further validation in larger, prospective cohorts.