Abstract
This retrospective analysis aimed to evaluate the diagnostic performance of 18F-flotufolastat PET/MRI for the detection of clinically significant prostate cancer (csPCa). The primary objective of the study was to assess the proportion of patients for whom a prostate biopsy could be safely circumvented by combining the Prostate Imaging Reporting and Data System (PI-RADS) from multiparametric MRI and quantitative data from PET. Methods: This study included 79 patients with suspicion of csPCa based on elevated prostate-specific antigen (>4 ng/mL) who underwent 18F-flotufolastat PET/MRI before biopsy or surgical resection of the prostate. Analytic methods included receiver operating characteristic analysis, logistic regression, and threshold-based evaluation of PI-RADS and of SUVmax Imaging findings were compared against histopathology. Results: csPCa (International Society of Urological Pathology grade group ≥ 2) was confirmed in 42 patients, whereas histopathology excluded csPCa in 37 patients. The combination of PI-RADS and SUVmax resulted in an area under the curve (AUC) of 87.1%, outperforming PI-RADS alone (AUC of 75.2%) and SUVmax alone (AUC of 80.7); however, differences were not statistically significant. A PI-RADS score of at least 3 and 18F-flotufolastat PET showed comparable sensitivity (0.90 vs. 0.91) and specificity (0.42 vs. 0.32) in the detection of csPCa, when applying a low SUVmax threshold (>4.5). Using a high SUVmax threshold (>10.0) increased the specificity of 18F-flotufolastat PET to 100%. Furthermore, combined PET/MRI increased the number of patients for whom csPCa can be reliably diagnosed to 28%, as well as increasing to 29% the number of patients for whom csPCa can be ruled out with a very high degree of certainty. Conclusion: This retrospective analysis suggests that prostate-specific membrane antigen PET/MRI can stratify patients with suspected prostate cancer into 3 groups: one with a very high likelihood of csPCa who may undergo definitive therapy without biopsy, one with a very low risk for csPCa who may not require biopsy, and an indeterminate group who still need biopsy. Overall, in our cohort, prostate biopsy would have been avoided in 57% of patients. Therefore, there is an urgent need for prospective studies on prostate-specific membrane antigen PET/MRI to reduce the number of unnecessary biopsies in men with suspected prostate cancer.