Prospective comparison of 68 Ga-PSMA PET/CT, 18 F-sodium fluoride PET/CT and diffusion weighted-MRI at for the detection of bone metastases in biochemically recurrent prostate cancer

Purpose To prospectively compare diagnostic accuracies for detection of bone metastases by 68 Ga-PSMA PET/CT, 18 F-NaF PET/CT and diffusion-weighted MRI (DW 600 -MRI) in prostate cancer (PCa) patients with biochemical recurrence (BCR). Methods Sixty-eight PCa patients with BCR participated in this prospective study. The patients underwent 68 Ga-PSMA PET/CT, a 18 F-NaF PET/CT and a DW 600 -MRI (performed in accordance with European Society of Urogenital Radiology guidelines, with b values of 0 and 600 s/mm 2 ). Bone lesions were categorized using a three-point scale (benign, malignant or equivocal for metastases) and a dichotomous scale (benign or metastatic) for each imaging modality by at least two experienced observers. A best valuable comparator was defined for each patient based on study-specific imaging, at least 12 months of clinical follow-up and any imaging prior to the study and during follow-up. Diagnostic performance was assessed using a sensitivity analysis where equivocal lesions were handled as non-metastatic and then as metastatic. Results Ten of the 68 patients were diagnosed with bone metastases. On a patient level, sensitivity, specificity and the area under the curve (AUC) by receiver operating characteristic analysis were, respectively, 0.80, 0.98–1.00 and 0.89–0.90 for 68 Ga-PSMA PET/CT ( n = 68 patients); 0.90, 0.90–0.98 and 0.90–0.94 for 18 NaF PET/CT ( n = 67 patients); and 0.25–0.38, 0.87–0.92 and 0.59–0.62 for DW 600 -MRI ( n = 60 patients). The diagnostic performance of DW 600 -MRI was significantly lower than that of 68 Ga-PSMA PET/CT and 18 NaF PET/CT for diagnosing bone metastases ( p < 0.01), and no significant difference in the AUC was seen between 68 Ga-PSMA PET/CT and 18 NaF PET/CT ( p = 0.65). Conclusion 68 Ga-PSMA PET/CT and 18 F-NaF PET/CT showed comparable and high diagnostic accuracies for detecting bone metastases in PCa patients with BCR. Both methods performed significantly better than DW 600 -MRI, which was inadequate for diagnosing bone metastases when conducted in accordance with European Society of Urogenital Radiology guidelines.