PURPOSE: This study aims to combine multiparametric magnetic resonance imaging (MRI) and digitized pathology with machine learning to generate predictive maps of histologic features for prostate cancer localization. METHODS AND MATERIALS: Thirty-nine patients underwent MRI prior to prostatectomy. After surgery, tissue was sliced according to MRI orientation using patient-specific 3-dimensionally printed slicing jigs. Whole-mount sections were annotated by our pathologist and digitally contoured to differentiate the lumen and epithelium. Slides were co-registered to the T2-weighted MRI scan. A learning curve was generated to determine the number of patients required for a stable machine-learning model. Patients were randomly stratified into 2 training sets and 1 test set. Two partial least-squares regression models were trained, each capable of predicting lumen and epithelium density. Predicted density values were calculated for each patient in the test dataset, mapped into the MRI space, and compared between regions confirmed as high-grade prostate cancer. RESULTS: The learning-curve analysis showed that a stable fit was achieved with data from 10 patients. Maps indicated that regions of increased epithelium and decreased lumen density, generated from each independent model, corresponded with pathologist-annotated regions of high-grade cancer. CONCLUSIONS: We present a radio-pathomic approach to mapping prostate cancer. We find that the maps are useful for highlighting high-grade tumors. This technique may be relevant for dose-painting strategies in prostate radiation therapy.

Prostate cancer is the most common noncutaneous cancer in men in the United States. The current paradigm for screening and diagnosis is imperfect, with relatively low specificity, high cost, and high morbidity. This study aims to generate new image contrasts by learning a distribution of unique image signatures associated with prostate cancer. In total, 48 patients were prospectively recruited for this institutional review board-approved study. Patients underwent multiparametric magnetic resonance imaging 2 weeks before surgery. Postsurgical tissues were annotated by a pathologist and aligned to the in vivo imaging. Radiomic profiles were generated by linearly combining 4 image contrasts (T2, apparent diffusion coefficient [ADC] 0-1000, ADC 50-2000, and dynamic contrast-enhanced) segmented using global thresholds. The distribution of radiomic profiles in high-grade cancer, low-grade cancer, and normal tissues was recorded, and the generated probability values were applied to a naive test set. The resulting Gleason probability maps were stable regardless of training cohort, functioned independent of prostate zone, and outperformed conventional clinical imaging (area under the curve [AUC] = 0.79). Extensive overlap was seen in the most common image signatures associated with high- and low-grade cancer, indicating that low- and high-grade tumors present similarly on conventional imaging.

Purpose: Our study predictively maps epithelium density in magnetic resonance imaging (MRI) space while varying the ground truth labels provided by five pathologists to quantify the downstream effects of interobserver variability. Approach: Clinical imaging and postsurgical tissue from 48 recruited prospective patients were used in our study. Tissue was sliced to match the MRI orientation and whole-mount slides were stained and digitized. Data from 28 patients ( n = 33 slides) were sent to five pathologists to be annotated. Slides from the remaining 20 patients ( n = 123 slides) were annotated by one of the five pathologists. Interpathologist variability was measured using Krippendorffs alpha. Pathologist-specific radiopathomic mapping models were trained using a partial least-squares regression using MRI values to predict epithelium density, a known marker for disease severity. An analysis of variance characterized intermodel means difference in epithelium density. A consensus model was created and evaluated using a receiver operator characteristic classifying high grade versus low grade and benign, and was statistically compared to apparent diffusion coefficient (ADC). Results: Interobserver variability ranged from low to acceptable agreement (0.31 to 0.69). There was a statistically significant difference in mean predicted epithelium density values ( p < 0.001 ) between the five models. The consensus model outperformed ADC (areas under the curve = 0.80 and 0.71, respectively, p < 0.05 ). Conclusion: We demonstrate that radiopathomic maps of epithelium density are sensitive to the pathologist annotating the dataset; however, it is unclear if these differences are clinically significant. The consensus model produced the best maps, matched the performance of the best individual model, and outperformed ADC.