Presented by: Linda Huynh` MSc
University of Nebraska Medical Center

Introduction

Multi-parametric magnetic resonance imaging (mpMRI)-derived radiomics have been shown to capture sub-visual patterns for quantitative characterization of prostate cancer (PC) phenotypes. The present study seeks to develop, test, and compare the performance of an MRI-derived radiomic model for the prediction of PC recurrence following definitive treatment with radical prostatectomy (RP).

Materials

mpMRI was obtained from 76 patients who had a minimum of 2 years follow-up following RP. The prostate was manually delineated as the region of interest and 924 radiomic features were extracted. All features were tested for stability via intraclass correlation coefficient (ICC) and image normalization via histogram matching.

Results

,

Six important and non-redundant features were found to be predictors of PC recurrence at a median of 4.1 (2.2) years and were aggregated into a radiomic model. Five-fold, ten-run cross-validation yielded a receiver-operator characteristic area under the curve (ROC-AUC) of 0.95±0.06 in the training set (n=56). In comparison, the UCSF Cancer of the Prostate Risk Assessment score and MSKCC Pre-Radical Prostatectomy nomograms yielded AUC of 0.72±0.07 and 0.82±0.07, respectively. Finally, when the radiomic model was applied to the test set (n=20), ROC-AUC was 0.67 and sensitivity, specificity, positive predictive value, and negative predictive value were 33%, 100%, 40% and 100%, respectively.

Conclusion

The present study is a proof of concept for the use of an mpMRI-derived radiomic model in predicting PC recurrence in 76 prostate cancer patients, yielding six radiomic features significantly associated with recurrence following RP. When these features were aggregated into a radiomic signature, this signature predicted recurrence well in cross-validation and predicted patients at low-risk for recurrence with 100% specificity. Furthermore, when predictive capability was compared with the UCSF-CAPRA score, the radiomic model illustrated significantly higher ROC-AUC.

Funding

N/A

Co-Authors

Jacob Marasco, MS
Creighton University

Thomas Ahlering, MD, FACS
University of California, Irvine

Shuo Wang, PhD
University of Nebraska Medical Center

Michael Baine, MD, PhD
University of Nebraska Medical Center