Noninvasive Prediction of High‐Grade Prostate Cancer via Biparametric MRI Radiomics

Abstract

Background Gleason score (GS) is a histologic prognostic factor and the basis of treatment decision‐making for prostate cancer (PCa). Treatment regimens between lower‐grade (GS ≤7) and high‐grade (GS >7) PCa differ largely and have great effects on cancer progression. Purpose To investigate the use of different sequences in biparametric MRI (bpMRI) of the prostate gland for noninvasively distinguishing high‐grade PCa. Study Type Retrospective. Population In all, 489 patients (training cohort: N = 326; test cohort: N = 163) with PCa between June 2008 and January 2018. Field Strength/Sequence 3.0T, pelvic phased‐array coils, bpMRI including T2‐weighted imaging (T2WI) and diffusion‐weighted imaging (DWI); apparent diffusion coefficient map extracted from DWI. Assessment The whole prostate gland was delineated. Radiomic features were extracted and selected using the Kruskal–Wallis test, the minimum redundancy‐maximum relevance, and the sequential backward elimination algorithm. Two single‐sequence radiomic (T2WI, DWI) and two combined (T2WI‐DWI, T2WI‐DWI‐Clinic) models were respectively constructed and validated via logistic regression. Statistical Tests The Kruskal–Wallis test and chi‐squared test were utilized to evaluate the differences among variable groups. P < 0.05 determined statistical significance. The area under the receiver operating characteristic curve (AUC), specificity, sensitivity, and accuracy were used to evaluate model performance. The Delong test was conducted to compare the differences between the AUCs of all models. Result All radiomic models showed significant (P < 0.001) predictive performances. Between the single‐sequence radiomic models, the DWI model achieved the most encouraging results, with AUCs of 0.801 and 0.787 in the training and test cohorts, respectively. For the combined models, the T2WI‐DWI models acquired an AUC of 0.788, which was almost the same with DWI in the test cohort, and no significant difference was found between them (training cohort: P = 0.199; test cohort: P = 0.924). Data Conclusion Radiomics based on bpMRI can noninvasively identify high‐grade PCa before the operation, which is helpful for individualized diagnosis of PCa. Level of Evidence 4 Technical Efficacy Stage 2 J. Magn. Reson. Imaging 2020;52:1102–1109.

Radiomics Models Based on Apparent Diffusion Coefficient Maps for the Prediction of High‐Grade Prostate Cancer at Radical Prostatectomy: Comparison With Preoperative Biopsy

“…Second, some factors or receptors related to PCa progression, such as CXC chemokine receptor 4 (CXCR4), are overexpressed in areas composed of approximately 40–60% of tumor glands rather than at the tumor center 39 . Although many studies have used MR radiomics analyses for the assessment of GS grading, only Gong et al 40 used manual segmentation of the whole prostate gland to construct a biparametric (T2WI + DWI) MR radiomics model for the prediction of high‐grade PCa. Their model achieved an AUC of 0.788 in the test cohort, but the performance did not improve after the incorporation of clinical information.…”

Detection of Prostate Cancer Using Biparametric Prostate MRI, Radiomics, and Kallikreins: A Retrospective Multicenter Study of Men With a Clinical Suspicion of Prostate Cancer

“…We hypothesize that the choice of large set of features may help in distinguishing disease pathophysiology better and hence likely improve prediction than a small set of features, which may miss certain tissue characteristics. [34][35][36][37][38] Furthermore, we employed opensource software and provide access to datasets as well as details on our feature pruning/selection process to ensure comparability and transparency. However, direct comparison of our results with prior studies is limited since none of prior studies applying radiomics and machine learning methods for prostate MRI provide access to datasets and postprocessing code.…”

Diffusion-Weighted Imaging in Oncology: An Update

“…Previous studies reported similar ( p = 0.83) diagnostic performance of biparametric MRI without DCE (pooled sensitivity and specificity of 0.74 and 0.90, respectively) and mpMRI (pooled sensitivity and specificity of 0.76 and 0.89, respectively) for the diagnosis of PCa, suggesting the former as a valuable first-line imaging test due to its robust sensitivity [ 84 ]. Notably, Gong et al have recently reported the high diagnostic performance of radiomic models based on biparametric MRI to non-invasively identify high-grade PCa, with accuracy ranging from 0.787 and 0.801 [ 85 ]. However, despite the limited role of DCE in determining the overall PI-RADS category, in some instances DCE may assist in the detection of PCa and the PI-RADS Committee suggests to reserve biparametric MRI only for selected clinical scenarios.…”