Leveraging MRI to reduce overtreatment in low‑grade prostate cancer: a retrospective single‑centre cohort study
Katarzyna Skrobisz, Kevin Miszewski, Bartłomiej Marczak, Laura Miszewska, Jakub Kulski, Marcin MatuszewskiBackground:
Prostate cancer is one of the most common malignancies in men worldwide. Radical prostatectomy is an effective curative treatment, but a noticeable number of men are ultimately found to have clinically insignificant disease, such as a Gleason score of 3 plus 3, on final histopathology. This suggests that some patients may undergo unnecessary surgery. The widespread adoption of multiparametric MRI has improved tumour detection and targeted biopsy accuracy, yet its potential role in preventing overtreatment is not fully understood. Because some low-risk tumours demonstrate low conspicuity on MRI, we hypothesised that specific MRI features, particularly low lesion conspicuity and PI-RADS 3 or lower, may be associated with final Grade Group 1 pathology in men undergoing radical prostatectomy.
Objectives:
The purpose of this study was to assess whether pre-operative multiparametric MRI can help identify men with potentially indolent prostate cancer and reduce unnecessary radical prostatectomies. We focused on MRI features such as PI-RADS 3 or lower, low lesion conspicuity and overall imaging quality to see how strongly they relate to final Grade Group 1 in men who were selected for surgery.
Design:
Retrospective single-centre cohort study.
Methods:
We reviewed 400 consecutive men who underwent robot-assisted radical prostatectomy for presumed clinically significant prostate cancer. Final pathology identified 13 men with Grade Group 1 disease. Pre-operative mpMRI examinations were reassessed for PI-RADS category, lesion location, lesion conspicuity and overall image quality to determine whether imaging features were associated with low-risk pathology.
Results:
Analyses showed that certain MRI characteristics were frequently associated with Gleason 3 plus 3 pathology. Several patients who proceeded to surgery had inconspicuous, equivocal or low probability MRI lesions such as PI-RADS 2 or 3 that lacked features typically associated with clinically significant cancer. Recognising these signatures may refine patient selection and support broader use of active surveillance.
Conclusion:
Assessment of MRI appearance, lesion conspicuity and PI-RADS category may improve selection for surgery and reduce overtreatment. MRI-based identification of low-grade prostate cancer could help avoid unnecessary prostatectomies and the associated morbidity while preserving oncologic safety.