In-Bore Versus Fusion MRI-Targeted Biopsy of PI-RADS Category 4 and 5 Lesions: A Retrospective Comparative Analysis Using Propensity Score Weighting.

Abstract

BACKGROUND. Few published studies have compared in-bore and fusion MRI-targeted prostate biopsy, and the available studies have had conflicting results. OBJECTIVE. The purpose of this study was to compare the target-specific cancer detection rate of in-bore prostate biopsy with that of fusion MRI-targeted biopsy. METHODS. The records of men who underwent in-bore or fusion MRI-targeted biopsy of PI-RADS category 4 or 5 lesions between August 2013 and September 2019 were retrospectively identified. PI-RADS version 2.1 assessment category, size, and location of each target were established by retrospective review by a single experienced radiologist. Patient history and target biopsy results were obtained by electronic medical record review. Only the first MRI-targeted biopsy of the dominant lesion was included for patients with repeated biopsies or multiple targets. In-bore and fusion biopsy were compared by propensity score weights and multivariable regression to adjust for imbalances in patient and target characteristics between biopsy techniques. The primary endpoint was target-specific prostate cancer detection rate. Secondary endpoints were detection rate after application of propensity score weighting for cancers in International Society of Urological Pathology (ISUP) grade group 2 (GG2) or higher and detection rate with the use of off-target systematic sampling results. RESULTS. The study sample included 286 men (in-bore biopsy, 191; fusion biopsy, 95). Compared with fusion biopsy, in-bore biopsy was associated with significantly greater likelihood of detection of any cancer (odds ratio, 2.28 [95% CI, 1.04-4.98]; p = .04) and nonsignificantly greater likelihood of detection of ISUP GG2 or higher cancer (odds ratio, 1.57 [95% CI, 0.88-2.79]; p = .12) in a target. When off-target sampling was included, in-bore biopsy and combined fusion and systematic biopsy were not different for detection of any cancer (odds ratio, 1.16 [95% CI, 0.54-2.45]; p = .71) or ISUP GG2 and higher cancer (odds ratio, 1.15 [95% CI, 0.66-2.01]; p = .62). CONCLUSION. In this retrospective study in which propensity score weighting was used, in-bore MRI-targeted prostate biopsy had a higher target-specific cancer detection rate than did fusion biopsy. CLINICAL IMPACT. Pending a larger prospective randomized multicenter comparison between in-bore and fusion biopsy, in-bore may be the preferred approach should performing only biopsy of a suspicious target, without concurrent systematic biopsy, be considered clinically appropriate.