BackgroundGenetic profiling represents the future of neuro-oncology but suffers from inadequate biopsies in heterogeneous tumors like Glioblastoma (GBM). Contrast-enhanced MRI (CE-MRI) targets enhancing core (ENH) but yields adequate tumor in only ~60% of cases. Further, CE-MRI poorly localizes infiltrative tumor within surrounding non-enhancing parenchyma, or brain-around-tumor (BAT), despite the importance of characterizing this tumor segment, which universally recurs. In this study, we use multiple texture analysis and machine learning (ML) algorithms to analyze multi-parametric MRI, and produce new images indicating tumor-rich targets in GBM.MethodsWe recruited primary GBM patients undergoing image-guided biopsies and acquired pre-operative MRI: CE-MRI, Dynamic-Susceptibility-weighted-Contrast-enhanced-MRI, and Diffusion Tensor Imaging. Following image coregistration and region of interest placement at biopsy locations, we compared MRI metrics and regional texture with histologic diagnoses of high- vs low-tumor content (≥80% vs <80% tumor nuclei) for corresponding samples. In a training set, we used three texture analysis algorithms and three ML methods to identify MRI-texture features that optimized model accuracy to distinguish tumor content. We confirmed model accuracy in a separate validation set.ResultsWe collected 82 biopsies from 18 GBMs throughout ENH and BAT. The MRI-based model achieved 85% cross-validated accuracy to diagnose high- vs low-tumor in the training set (60 biopsies, 11 patients). The model achieved 81.8% accuracy in the validation set (22 biopsies, 7 patients).ConclusionMulti-parametric MRI and texture analysis can help characterize and visualize GBM’s spatial histologic heterogeneity to identify regional tumor-rich biopsy targets.

BACKGROUND AND PURPOSE:Because sinonasal inverted papilloma can harbor squamous cell carcinoma, differentiating these tumors is relevant. The objectives of this study were to determine whether MR imaging-based texture analysis can accurately classify cases of noncoexistent squamous cell carcinoma and inverted papilloma and to compare this classification performance with neuroradiologists' review.

Neoadjuvant endocrine therapy (NET) is increasingly used for the treatment of estrogen receptor positive, HER2 negative breast cancer. We evaluated whether MRI phenotype and background parenchymal enhancement (BPE) can predict response to NET. Patients with localized breast cancer treated with NET and had a pre-treatment breast MRI were identified. Baseline MRI phenotype and BPE was interpreted by a single radiologist blinded to the results of systemic therapy. Response was defined as stable disease or reduction in tumor size on clinical and/or ultrasound examination. Of the 21 patients identified, 17 were responders; all patients with minimal/mild BPE had a response compared to 5/9 (56%) patients with moderate/marked BPE (P = 0.02). All four nonresponders had moderate/marked BPE as compared to 5/17 (29%) responders (P = 0.02). This pilot study suggests that minimal/mild BPE may be predictive of a positive response to NET. A higher degree of background enhancement was significantly predictive of negative response to NET.

Useful diagnostic information regarding HPV infection can be extracted from the CT appearance of OPSCC beyond what is apparent to the trained human eye.