Recent conflicting reports have found both brain tumor hypercellularity and necrosis in regions of restricted diffusion on MRI-derived apparent diffusion coefficient (ADC) images. This study precisely compares ADC and cell density voxel by voxel using postmortem human whole brain samples. Patients with meningioma were evaluated to determine a normative ADC distribution within benign fluid attenuated inversion recovery (FLAIR) T2/hyperintensity surrounding tumor. This distribution was used to calculate a minimum ADC threshold to define regions of ADC-FLAIR mismatch (AFMM), where restricted diffusion presented in conjunction with T2/FLAIR hyperintensity. Contrast-enhancing voxels were excluded from this analysis. AFMM maps were generated using imaging acquired prior to death in 7 patients with high-grade glioma who eventually donated their brains upon death. Histological samples were taken from numerous regions of abnormal FLAIR and AFMM. Each sample was computationally processed to determine cell density. Custom software was then used to downsample coregistered microscopic histology to the more coarse MRI resolution. A voxel-by-voxel evaluation comparing ADC and cellularity was then performed. An ADC threshold of 0.929 × 10(-3) mm(2)/s was calculated from meningioma-induced edema and was used to define AFMM. Regions of AFMM showed significantly greater cell density in 6 of 7 high-grade glioma cases compared with regions of hyperintense FLAIR alone (P < .0001). Two patients had small regions of diffusion-restricted necrosis that had significantly lower ADC than nearby hypercellularity. Regions of AFMM contain hypercellularity except for regions with extremely restricted diffusion, where necrosis is present.

Magnetic resonance imaging (MRI) is used to diagnose and monitor brain tumors. Extracting additional information from medical imaging and relating it to a clinical variable of interest is broadly defined as radiomics. Here, multiparametric MRI radiomic profiles (RPs) of de novo glioblastoma (GBM) brain tumors is related with patient prognosis. Clinical imaging from 81 patients with GBM before surgery was analyzed. Four MRI contrasts were aligned, masked by margins defined by gadolinium contrast enhancement and T2/fluid attenuated inversion recovery hyperintensity, and contoured based on image intensity. These segmentations were combined for visualization and quantification by assigning a 4-digit numerical code to each voxel to indicate the segmented RP. Each RP volume was then compared with overall survival. A combined classifier was then generated on the basis of significant RPs and optimized volume thresholds. Five RPs were predictive of overall survival before therapy. Combining the RP classifiers into a single prognostic score predicted patient survival better than each alone (P < .005). Voxels coded with 1 RP associated with poor prognosis were pathologically confirmed to contain hypercellular tumor. This study applies radiomic profiling to de novo patients with GBM to determine imaging signatures associated with poor prognosis at tumor diagnosis. This tool may be useful for planning surgical resection or radiation treatment margins.

BACKGROUND AND PURPOSE: Patients with recurrent glioblastoma often exhibit regions of diffusion restriction following the initiation of bevacizumab therapy. Studies suggest that these regions represent either diffusion-restricted necrosis or hypercellular tumor. This study explored postmortem brain specimens and a population analysis of overall survival to determine the identity and implications of such lesions. MATERIALS AND METHODS: Postmortem examinations were performed on 6 patients with recurrent glioblastoma on bevacizumab with progressively growing regions of diffusion restriction. ADC values were extracted from regions of both hypercellular tumor and necrosis. A receiver operating characteristic analysis was performed to define optimal ADC thresholds for differentiating tissue types. A retrospective population study was also performed comparing the overall survival of 64 patients with recurrent glioblastoma treated with bevacizumab. Patients were separated into 3 groups: no diffusion restriction, diffusion restriction that appeared and progressed within 5 months of bevacizumab initiation, and delayed or stable diffusion restriction. An additional analysis was performed assessing tumor O⁶-methylguanine-DNA-methyltransferase methylation. RESULTS: The optimal ADC threshold for differentiation of hypercellularity and necrosis was 0.736 × 10⁻³mm²/s. Progressively expanding diffusion restriction was pathologically confirmed to be coagulative necrosis surrounded by viable tumor. Progressive lesions were associated with the worst overall survival, while stable lesions showed the greatest overall survival (P < .05). Of the 40% of patients with O⁶-methylguanine-DNA-methyltransferase methylated tumors, none developed diffusion-restricted lesions. CONCLUSIONS: Progressive diffusion-restricted lesions were pathologically confirmed to be coagulative necrosis surrounded by viable tumor and associated with decreased overall survival. Stable lesions were, however, associated with increased overall survival. All lesions were associated with O⁶-methylguanine-DNA-methyltransferase unmethylated tumors.

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. 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. 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. 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.