Clinical Scorecard: Combining Radiomic Features from Biparametric MRI with Clinical Data Enhances Prediction of Prostate Cancer Recurrence Before Treatment
At a Glance
Category
Detail
Condition
Prostate Cancer
Key Mechanisms
Integration of radiomic features from biparametric MRI with clinical variables to enhance prediction of biochemical recurrence.
Target Population
Men diagnosed with prostate cancer undergoing radical prostatectomy.
Care Setting
Oncology and Radiology
Key Highlights
Combined model achieved AUC of 0.85, outperforming clinical-only and radiomics-only models.
Calibration of the model was strong with a slope of 1.01.
High-risk patients had significantly shorter recurrence-free survival.
Most influential predictors included Gleason Grade Group and PSA levels.
The combined model showed higher clinical net benefit than the D’Amico classification.
Guideline-Based Recommendations
Diagnosis
Utilize biparametric MRI for enhanced imaging in prostate cancer diagnosis.
Management
Consider integrating radiomic features with clinical variables for recurrence risk assessment.
Monitoring & Follow-up
Monitor PSA levels post-radical prostatectomy to assess for biochemical recurrence.
Risks
Up to 30% of patients may experience biochemical recurrence after radical prostatectomy.
Patient & Prescribing Data
395 male patients with prostate cancer undergoing radical prostatectomy.
Integration of imaging and clinical data may refine treatment planning and surveillance strategies.
Clinical Best Practices
Incorporate radiomic analysis in pre-treatment assessments.
Utilize a stacked ensemble model for improved prediction accuracy.
Regularly evaluate model performance using calibration and decision-curve analysis.
by Selma Bozorgpana, Indri Desiati, Mohammed R. S. Sunoqrot, Petter Davik, Guro F. Giskeødegård, Gabriel Addio Nketiah, Mattijs Elschot, May-Britt Tessem, Tone F. Bathen
