Abstract
BACKGROUND: Assessing the extent of disease in newly diagnosed prostate cancer (PC) patients is crucial for tailoring an appropriate treatment approach. Prostate-specific membrane antigen (PSMA)-targeted positron emission tomography/computed tomography (PET/CT) reportedly has greater accuracy than conventional imaging for staging PC. As with any imaging modality, pitfalls and nonspecific findings do occur. The PSMA reporting and data system (PSMA-RADS) version 1.0 offers structured interpretation of PSMA-targeted studies and classifies lesions by likelihood of clinical significance. The aim of this retrospective study was to evaluate the clinical significance of equivocal bone findings on staging PSMA-targeted imaging, as defined by PSMA-RADS version 1.0, in the preoperative setting. Fifteen of 406 consecutive patients staged by PET/CT prior to radical prostatectomy had equivocal bone lesions. The scans were retrospectively scored with the PSMA-RADS version 1.0 system, blinded to disease course and follow-up data. Postoperative persistence of prostate-specific antigen levels supported by imaging and histological findings was used as the reference standard for the true significance of equivocal imaging findings. RESULTS: Thirteen of the 15 patients had an overall PSMA-RADS score of 3B, of whom only two had true metastatic disease. The remaining patients had scores of 4 (n = 1) or 5 (n = 1), all confirmed as true positive prostate-related malignant lesions. A per-lesion analysis identified 29 bone lesions, of which 27 were scored PSMA-RADS 3B, and only three of them were true metastases. Thus, debatable lesions proved to have no clinical significance in 84.6% of cases, and only 11% of equivocal PSMA-RADS 3B bone lesions were true positive. CONCLUSIONS: In intermediate and high-risk patients staged prior to radical prostatectomy, the majority of PSMA-RADS 3B lesions are of no clinical relevance. Bone lesions judged as being highly suspicious for metastases (PSMA-RADS 4/5) were all validated as true positives.