Diagnostic Accuracy of Multiparametric Magnetic Resonance Imaging to Detect Residual Prostate Cancer Following Irreversible Electroporation—A Multicenter Validation Study

Abstract

BackgroundAccurate monitoring following focal treatment of prostate cancer (PCa) is paramount for timely salvage treatment or retreatment. ObjectiveTo evaluate the diagnostic accuracy of multiparametric magnetic resonance imaging (mpMRI) to detect residual PCa in the short-term follow-up of focal treatment with irreversible electroporation (IRE) using transperineal or transrectal template ± targeted biopsies. Design, setting, and participantsA retrospective international multicenter study of men with biopsy-proven PCa, treated with focal IRE, and followed by mpMRI (index-test) and template biopsies (reference-test) between February 2013 and January 2021, was conducted. Outcome measurements and statistical analysisSensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of mpMRI were calculated for in- and outfield residual disease based on two definitions of significant PCa: University College London (UCL) 1—International Society of Urological Pathology (ISUP) ≥3 or ISUP ≥1 with maximum cancer core length (MCCL) ≥6 mm, and UCL2—ISUP ≥2 or ISUP ≥1 with MCCL ≥4 mm. Results and limitationsA total of 303 patients from five focal therapy centers were treated with primary IRE. The final analysis was performed on 217 men (median age 67, median prostate-specific antigen 6.2, 81% ISUP 2/3) who underwent both mpMRI and template biopsies. Multiparametric MRI missed 38/57 (67%) positive biopsy locations (UCL1) in 22 patients. Sensitivity, specificity, PPV, and NPV of mpMRI to detect whole gland residual disease (UCL1) were 43.6% (95% confidence interval [CI]: 28–59), 80.9% (95% CI: 75–86), 33.3% (95% CI: 21–47), and 86.7% (95% CI: 81–91), respectively. Based on UCL2, sensitivity, specificity, PPV, and NPV were 35.8% (95% CI: 25–48), 82.0% (95% CI: 75–88), 47.1% (95% CI: 34–61), and 74.1% (95% CI: 67–80), respectively. Limitations are the retrospective nature and short follow-up. ConclusionsThe diagnostic accuracy of mpMRI to detect residual clinically significant PCa following IRE was low. Follow-up template biopsies should be performed, regardless of mpMRI results. Patient summaryWe investigated the accuracy of magnetic resonance imaging (MRI) to detect residual prostate cancer after treatment with irreversible electroporation. The accuracy of MRI is insufficient, and we emphasize the importance of confirmatory prostate biopsies.