Radiomic Machine Learning for Characterization of Prostate Lesions with MRI: Comparison to ADC Values.

Purpose To compare biparametric contrast-free radiomic machine learning (RML), mean apparent diffusion coefficient (ADC), and radiologist assessment for characterization of prostate lesions detected during prospective MRI interpretation. Materials and Methods This single-institution study included 316 men (mean age ± standard deviation, 64.0 years ± 7.8) with an indication for MRI-transrectal US fusion biopsy between May 2015 and September 2016 (training cohort, 183 patients; test cohort, 133 patients). Lesions identified by prospective clinical readings were manually segmented for mean ADC and radiomics analysis. Global and zone-specific random forest RML and mean ADC models for classification of clinically significant prostate cancer (Gleason grade group ≥ 2) were developed on the training set and the fixed models tested on an independent test set. Clinical readings, mean ADC, and radiomics were compared by using the McNemar test and receiver operating characteristic (ROC) analysis. Results In the test set, radiologist interpretation had a per-lesion sensitivity of 88% (53 of 60) and specificity of 50% (79 of 159). Quantitative measurement of the mean ADC (cut-off 732 mm2/sec) significantly reduced false-positive (FP) lesions from 80 to 60 (specificity 62% [99 of 159]) and false-negative (FN) lesions from seven to six (sensitivity 90% [54 of 60]) (P = .048). Radiologist interpretation had a per-patient sensitivity of 89% (40 of 45) and specificity of 43% (38 of 88). Quantitative measurement of the mean ADC reduced the number of patients with FP lesions from 50 to 43 (specificity 51% [45 of 88]) and the number of patients with FN lesions from five to three (sensitivity 93% [42 of 45]) (P = .496). Comparison of the area under the ROC curve (AUC) for the mean ADC (AUCglobal = 0.84; AUCzone-specific ≤ 0.87) vs the RML (AUCglobal = 0.88, P = .176; AUCzone-specific ≤ 0.89, P ≥ .493) showed no significantly different performance. Conclusion Quantitative measurement of the mean apparent diffusion coefficient (ADC) improved differentiation of benign versus malignant prostate lesions, compared with clinical assessment. Radiomic machine learning had comparable but not better performance than mean ADC assessment. © RSNA, 2018 Online supplemental material is available for this article.

Combined Clinical Parameters and Multiparametric Magnetic Resonance Imaging for the Prediction of Extraprostatic Disease-A Risk Model for Patient-tailored Risk Stratification When Planning Radical Prostatectomy.

BACKGROUND: Multiparametric magnetic resonance imaging (mpMRI) facilitates the detection of significant prostate cancer. Therefore, addition of mpMRI to clinical parameters might improve the prediction of extraprostatic extension (EPE) in radical prostatectomy (RP) specimens. OBJECTIVE: To investigate the accuracy of a novel risk model (RM) combining clinical and mpMRI parameters to predict EPE in RP specimens. DESIGN, SETTING, AND PARTICIPANTS: We added prebiopsy mpMRI to clinical parameters and developed an RM to predict individual side-specific EPE (EPE-RM). Clinical parameters of 264 consecutive men with mpMRI prior to MRI/transrectal ultrasound fusion biopsy and subsequent RP between 2012 and 2015 were retrospectively analysed. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Multivariate regression analyses were used to determine significant EPE predictors for RM development. The prediction performance of the novel EPE-RM was compared with clinical T stage (cT), MR-European Society of Urogenital Radiology (ESUR) classification for EPE, two established nomograms (by Steuber et al and Ohori et al) and a clinical nomogram based on the coefficients of the established nomograms, and was constructed based on the data of the present cohort, using receiver operating characteristics (ROCs). For comparison, models' likelihood ratio (LR) tests and Vuong tests were used. Discrimination and calibration of the EPE-RM were validated based on resampling methods using bootstrapping. RESULTS AND LIMITATIONS: International society of Urogenital Pathology grade on biopsy, ESUR criteria, prostate-specific antigen, cT, prostate volume, and capsule contact length were included in the EPE-RM. Calibration of the EPE-RM was good (error 0.018). The ROC area under the curve for the EPE-RM was larger (0.87) compared with cT (0.66), Memorial Sloan Kettering Cancer Center nomogram (0.73), Steuber nomogram (0.70), novel clinical nomogram (0.79), and ESUR classification (0.81). Based on LR and Vuong tests, the EPE-RM's model fit was significantly better than that of cT, all clinical models, and ESUR classification alone (p<0.001). Limitations include monocentric design and expert reading of MRI. CONCLUSIONS: This novel EPE-RM, incorporating clinical and MRI parameters, performed better than contemporary clinical RMs and MRI predictors, therefore providing an accurate patient-tailored preoperative risk stratification of side-specific EPE. PATIENT SUMMARY: Extraprostatic extension of prostate cancer can be predicted accurately using a combination of magnetic resonance imaging and clinical parameters. This novel risk model outperforms magnetic resonance imaging and clinical predictors alone and can be useful when planning nerve-sparing radical prostatectomy.