Radiomics Based on MRI as a Biomarker to Guide Therapy by Predicting Upgrading of Prostate Cancer From Biopsy to...

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Gleason score (GS) from prostate cancer (PCa) biopsy is critical for treatment decisions.
• GS upgrading from biopsy to radical prostatectomy (RP) can lead to patient undertreatment.

Purpose of the Study:

• To develop and validate a radiomics model using multiparametric MRI (mp-MRI) to predict PCa upgrading.
• Enhance the accuracy of treatment decisions by identifying potential GS discrepancies.

Main Methods:

• Retrospective analysis of 166 patients with RP-confirmed PCa.
• Radiomic features extracted from T2-weighted, ADC, and DCE MRI sequences.
• Multivariate logistic regression used to build predictive models and a radiomics nomogram.

Main Results:

• The radiomics model integrating all three MRI sequences achieved an AUC of 0.868.
• A combined model including radiomics, clinical stage, and time from biopsy to RP showed superior performance (AUC 0.910).
• The nomogram demonstrated good predictive performance and calibration.

Conclusions:

• Radiomics analysis of mp-MRI shows significant potential for predicting PCa upgrading.
• This approach can aid in more precise treatment planning for prostate cancer patients.