Related Experiment Video
Updated: Jun 9, 2026
Use of MRI-ultrasound Fusion to Achieve Targeted Prostate Biopsy
Published on: April 9, 2019
Development and validation of an MRI radiomics-based model for predicting progression risk in prostate cancer after
Ke Ding1, Qiong Chen2, Lifeng Huang1
1Department of Radiology, Third Affiliated Hospital of Guangxi Medical University, Nanning, China.
Background:
Prostate cancer often progresses to castration-resistant disease despite initial response to endocrine therapy, necessitating better predictive tools like magnetic resonance imaging (MRI) radiomics. This study aimed to develop a predictive model using MRI radiomics and clinicopathological factors to assess tumor progression risk after endocrine therapy in prostate cancer patients, and to create a nomogram for evaluating progression-free survival (PFS).
Methods:
A total of 136 prostate cancer patients receiving endocrine therapy were retrospectively analyzed and randomly split into training (n=95) and internal validation (n=41) sets (7:3). A radiomics-clinical nomogram was developed and validated internally and externally (n=52). Performance was assessed for discrimination, calibration, and clinical utility.
Results:
Independent predictors for tumor progression included time to prostate-specific antigen (PSA) nadir, Gleason score, tumor T stage, and bone metastasis. The combined prediction model achieved C-index values of 0.884, 0.839, and 0.795 in training, internal validation, and external validation sets, respectively. Calibration curves indicated accuracy; decision curve analysis confirmed clinical utility. Kaplan-Meier analysis showed that using a nomogram score of 79.44 as the cutoff effectively stratified prostate cancer patients into high-risk (>79.44) and low-risk (≤79.44) groups, with significantly shorter PFS in the high-risk group (log-rank test, P<0.001).
Conclusions:
The model incorporating MRI radiomics features with clinicopathological factors effectively predicts progression risk post-endocrine therapy in prostate cancer patients, aiding personalized clinical decisions to improve prognosis.
