Pre-operative MRI radiomics model non-invasively predicts key genomic markers and survival in glioblastoma patients.

PURPOSE: Although glioblastoma (GBM) is the most common primary brain malignancy, few tools exist to pre-operatively risk-stratify patients by overall survival (OS) or common genetic alterations. We developed an MRI-based radiomics model to identify patients with EGFR amplification, MGMT methylation, GBM subtype, and OS greater than 12 months. METHODS: We retrospectively identified 235 patients with pathologically confirmed GBMs from the Cancer Genome Atlas (88; TCGA) and MD Anderson Cancer Center (147; MDACC). After two neuroradiologists segmented MRI tumor volumes, we extracted first-order and second-order radiomic features (gray-level co-occurrence matrices). We used the Maximum Relevance Minimum Redundancy technique to identify the 100 most relevant features and validated models using leave-one-out-cross-validation and validation on external datasets (i.e., TCGA). Our results were reported as the area under the curve (AUC). RESULTS: The MDACC patient cohort had significantly higher OS (22 months) than the TCGA dataset (14 months). On both LOOCV and external validation, our radiomics models were able to identify EGFR amplification (all AUCs > 0.83), MGMT methylation (all AUCs > 0.85), GBM subtype (all AUCs > 0.92), and OS (AUC > 0.91 on LOOCV and 0.71 for TCGA validation). CONCLUSIONS: Our robust radiomics pipeline has the potential to pre-operatively discriminate common genetic alterations and identify patients with favorable survival.

Brain extraction on MRI scans in presence of diffuse glioma: Multi-institutional performance evaluation of deep learning methods and robust modality-agnostic training.

Brain extraction, or skull-stripping, is an essential pre-processing step in neuro-imaging that has a direct impact on the quality of all subsequent processing and analyses steps. It is also a key requirement in multi-institutional collaborations to comply with privacy-preserving regulations. Existing automated methods, including Deep Learning (DL) based methods that have obtained state-of-the-art results in recent years, have primarily targeted brain extraction without considering pathologically-affected brains. Accordingly, they perform sub-optimally when applied on magnetic resonance imaging (MRI) brain scans with apparent pathologies such as brain tumors. Furthermore, existing methods focus on using only T1-weighted MRI scans, even though multi-parametric MRI (mpMRI) scans are routinely acquired for patients with suspected brain tumors. In this study, we present a comprehensive performance evaluation of recent deep learning architectures for brain extraction, training models on mpMRI scans of pathologically-affected brains, with a particular focus on seeking a practically-applicable, low computational footprint approach, generalizable across multiple institutions, further facilitating collaborations. We identified a large retrospective multi-institutional dataset of n=3340 mpMRI brain tumor scans, with manually-inspected and approved gold-standard segmentations, acquired during standard clinical practice under varying acquisition protocols, both from private institutional data and public (TCIA) collections. To facilitate optimal utilization of rich mpMRI data, we further introduce and evaluate a novel ''modality-agnostic training'' technique that can be applied using any available modality, without need for model retraining. Our results indicate that the modality-agnostic approach1 obtains accurate results, providing a generic and practical tool for brain extraction on scans with brain tumors.

Radiomics to predict immunotherapy-induced pneumonitis: proof of concept.

We present the first reported work that explores the potential of radiomics to predict patients who are at risk for developing immunotherapy-induced pneumonitis. Despite promising results with immunotherapies, immune-related adverse events (irAEs) are challenging. Although less common, pneumonitis is a potentially fatal irAE. Thus, early detection is critical for improving treatment outcomes; an urgent need to identify biomarkers that predict patients at risk for pneumonitis exists. Radiomics, an emerging field, is the automated extraction of high fidelity, high-dimensional imaging features from standard medical images and allows for comprehensive visualization and characterization of the tissue of interest and corresponding microenvironment. In this pilot study, we sought to determine whether radiomics has the potential to predict development of pneumonitis. We performed radiomic analyses using baseline chest computed tomography images of patients who did (N = 2) and did not (N = 30) develop immunotherapy-induced pneumonitis. We extracted 1860 radiomic features in each patient. Maximum relevance and minimum redundancy feature selection method, anomaly detection algorithm, and leave-one-out cross-validation identified radiomic features that were significantly different and predicted subsequent immunotherapy-induced pneumonitis (accuracy, 100% [p = 0.0033]). This study suggests that radiomic features can classify and predict those patients at baseline who will subsequently develop immunotherapy-induced pneumonitis, further enabling risk-stratification that will ultimately lead to better treatment outcomes.

Multi-center study finds postoperative residual non-enhancing component of glioblastoma as a new determinant of patient outcome.

INTRODUCTION: The aim of the present study is to assess whether postoperative residual non-enhancing volume (PRNV) is correlated and predictive of overall survival (OS) in glioblastoma (GBM) patients. METHODS: We retrospectively analyzed a total 134 GBM patients obtained from The University of Texas MD Anderson Cancer Center (training cohort, n = 97) and The Cancer Genome Atlas (validation cohort, n = 37). All patients had undergone postoperative magnetic resonance imaging immediately after surgery. We evaluated the survival outcomes with regard to PRNV. The role of possible prognostic factors that may affect survival after resection, including age, sex, preoperative Karnofsky performance status, postoperative nodular enhancement, surgically induced enhancement, and postoperative necrosis, was investigated using univariate and multivariate Cox proportional hazards regression analyses. Additionally, a recursive partitioning analysis (RPA) was used to identify prognostic groups. RESULTS: Our analyses revealed that a high PRNV (HR 1.051; p-corrected = 0.046) and old age (HR 1.031; p-corrected = 0.006) were independent predictors of overall survival. This trend was also observed in the validation cohort (higher PRNV: HR 1.127, p-corrected = 0.002; older age: HR 1.034, p-corrected = 0.022). RPA analysis identified two prognostic risk groups: low-risk group (PRNV < 70.2 cm3; n = 55) and high-risk group (PRNV ≥ 70.2 cm3; n = 42). GBM patients with low PRNV had a significant survival benefit (5.6 months; p = 0.0037). CONCLUSION: Our results demonstrate that high PRNV is associated with poor OS. Such results could be of great importance in a clinical setting, particularly in the postoperative management and monitoring of therapy.

Ex vivo MR volumetry of human brain hemispheres.

PURPOSE: The aims of this work were to (a) develop an approach for ex vivo MR volumetry of human brain hemispheres that does not contaminate the results of histopathological examination, (b) longitudinally assess regional brain volumes postmortem, and (c) investigate the relationship between MR volumetric measurements performed in vivo and ex vivo. METHODS: An approach for ex vivo MR volumetry of human brain hemispheres was developed. Five hemispheres from elderly subjects were imaged ex vivo longitudinally. All datasets were segmented. The longitudinal behavior of volumes measured ex vivo was assessed. The relationship between in vivo and ex vivo volumetric measurements was investigated in seven elderly subjects imaged both antemortem and postmortem. RESULTS: This approach for ex vivo MR volumetry did not contaminate the results of histopathological examination. For a period of 6 months postmortem, within-subject volume variation across time points was substantially smaller than intersubject volume variation. A close linear correspondence was detected between in vivo and ex vivo volumetric measurements. CONCLUSION: Regional brain volumes measured with this approach for ex vivo MR volumetry remain relatively unchanged for a period of 6 months postmortem. Furthermore, the linear relationship between in vivo and ex vivo MR volumetric measurements suggests that this approach captures information linked to antemortem macrostructural brain characteristics.

A Radiomics Model Based on Synthetic MRI Acquisition for Predicting Neoadjuvant Systemic Treatment Response in Triple-Negative Breast Cancer.

Purpose To determine if a radiomics model based on quantitative maps acquired with synthetic MRI (SyMRI) is useful for predicting neoadjuvant systemic therapy (NAST) response in triple-negative breast cancer (TNBC). Materials and Methods In this prospective study, 181 women diagnosed with stage I-III TNBC were scanned with a SyMRI sequence at baseline and at midtreatment (after four cycles of NAST), producing T1, T2, and proton density (PD) maps. Histopathologic analysis at surgery was used to determine pathologic complete response (pCR) or non-pCR status. From three-dimensional tumor contours drawn on the three maps, 310 histogram and textural features were extracted, resulting in 930 features per scan. Radiomic features were compared between pCR and non-pCR groups by using Wilcoxon rank sum test. To build a multivariable predictive model, logistic regression with elastic net regularization and cross-validation was performed for texture feature selection using 119 participants (median age, 52 years [range, 26-77 years]). An independent testing cohort of 62 participants (median age, 48 years [range, 23-74 years]) was used to evaluate and compare the models by area under the receiver operating characteristic curve (AUC). Results Univariable analysis identified 15 T1, 10 T2, and 12 PD radiomic features at midtreatment that predicted pCR with an AUC greater than 0.70 in both the training and testing cohorts. Multivariable radiomics models of maps acquired at midtreatment demonstrated superior performance over those acquired at baseline, achieving AUCs as high as 0.78 and 0.72 in the training and testing cohorts, respectively. Conclusion SyMRI-based radiomic features acquired at midtreatment are potentially useful for identifying early NAST responders in TNBC. Keywords: MR Imaging, Breast, Outcomes Analysis ClinicalTrials.gov registration no. NCT02276443 Supplemental material is available for this article. © RSNA, 2023 See also the commentary by Houser and Rapelyea in this issue.

The Brain Tumor Segmentation (BraTS) Challenge 2023: Brain MR Image Synthesis for Tumor Segmentation (BraSyn).

Automated brain tumor segmentation methods have become well-established and reached performance levels offering clear clinical utility. These methods typically rely on four input magnetic resonance imaging (MRI) modalities: T1-weighted images with and without contrast enhancement, T2-weighted images, and FLAIR images. However, some sequences are often missing in clinical practice due to time constraints or image artifacts, such as patient motion. Consequently, the ability to substitute missing modalities and gain segmentation performance is highly desirable and necessary for the broader adoption of these algorithms in the clinical routine. In this work, we present the establishment of the Brain MR Image Synthesis Benchmark (BraSyn) in conjunction with the Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2023. The primary objective of this challenge is to evaluate image synthesis methods that can realistically generate missing MRI modalities when multiple available images are provided. The ultimate aim is to facilitate automated brain tumor segmentation pipelines. The image dataset used in the benchmark is diverse and multi-modal, created through collaboration with various hospitals and research institutions.

Radiometal-Based PET/MRI Contrast Agents for Sensing Tumor Extracellular pH.

Acidosis is a useful biomarker for tumor diagnoses and for evaluating early response to anti-cancer treatments. Despite these useful applications, there are few methods for non-invasively measuring tumor extracellular pH, and none are routinely used in clinics. Responsive MRI contrast agents have been developed, and they undergo a change in MRI signal with pH. However, these signal changes are concentration-dependent, and it is difficult to accurately measure the concentration of an MRI contrast agent in vivo. PET/MRI provides a unique opportunity to overcome this concentration dependence issue by using the PET component to report on the concentration of the pH-responsive MRI agent. Herein, we synthesized PET/MRI co-agents based on the design of a pH-dependent MRI agent, and we have correlated pH with the r1 relaxivity of the MRI co-agent. We have also developed a procedure that uses PET radioactivity measurements and MRI R1 relaxation rate measurements to determine the r1 relaxivity of the MRI co-agent, which can then be used to estimate pH. This simultaneous PET/MRI procedure accurately measured pH in solution, with a precision that depended on the concentration of the MRI co-agent. We used our procedure to measure extracellular pH in a subcutaneous flank model of MIA PaCa-2 pancreatic cancer. Although the PET co-agents were stable in serum, post-imaging studies showed evidence that the PET co-agents were degraded in vivo. These results showed that tumor acidosis can be evaluated with simultaneous PET/MRI, although improvements are needed to more precisely measure MRI R1 relaxation rates, and ensure the in vivo stability of the agents.

Development of a Nanoscale Chemical Exchange Saturation Transfer Magnetic Resonance Imaging Contrast Agent That Measures pH.

AcidoCEST MRI can measure the extracellular pH (pHe) of the tumor microenvironment in mouse models of human cancers and in patients who have cancer. However, chemical exchange saturation transfer (CEST) is an insensitive magnetic resonance imaging (MRI) contrast mechanism, requiring a high concentration of small-molecule agent to be delivered to the tumor. Herein, we developed a nanoscale CEST agent that can measure pH using acidoCEST MRI, which may decrease the requirement for high delivery concentrations of agent. We also developed a monomer agent for comparison to the polymer. After optimizing CEST experimental conditions, we determined that the polymer agent could be used during acidoCEST MRI studies at 125-fold and 488-fold lower concentration than the monomer agent and iopamidol, respectively. We also determined that both agents can measure pH with negligible dependence on temperature. However, pH measurements with both agents were dependent on concentration, which may be due to concentration-dependent changes in hydrogen bonding and/or steric hindrance. We performed in vivo acidoCEST MRI studies using the three agents to study a xenograft MDA-MB-231 model of mammary carcinoma. The tumor pHe measurements were 6.33 ± 0.12, 6.70 ± 0.15, and 6.85 ± 0.15 units with iopamidol, the monomer agent, and polymer agent, respectively. The higher pHe measurements with the monomer and polymer agents were attributed to the concentration dependence of these agents. This study demonstrated that nanoscale agents have merit for CEST MRI studies, but consideration should be given to the dependence of CEST contrast on the concentration of these agents.

Simultaneous Evaluations of pH and Enzyme Activity with a CEST MRI Contrast Agent.

The extracellular tumor microenvironment of many solid tumors has high acidosis and high protease activity. Simultaneously assessing both characteristics may improve diagnostic evaluations of aggressive tumors and the effects of anticancer treatments. Noninvasive imaging methods have previously been developed that measure extracellular pH or can detect enzyme activity using chemical exchange saturation transfer (CEST) magnetic resonance imaging (MRI). Herein, we developed a single-hybrid CEST agent that can simultaneously measure pH and evaluate protease activity using a combination of dual-power acidoCEST MRI and catalyCEST MRI. Our agent showed CEST signals at 9.2 ppm from a salicylic acid moiety and at 5.0 ppm from an aryl amide. The CEST signal at 9.2 ppm could be measured after selective saturation was applied at 1 and 4 µT, and these measurements could be used with a ratiometric analysis to determine pH. The CEST signal at 5.0 ppm from the aryl amide disappeared after the agent was treated with cathepsin B, while the CEST signal at 9.2 ppm remained, indicating that the agent could detect protease activity through the amide bond cleavage. Michaelis-Menten kinetics studies with catalyCEST MRI demonstrated that the binding affinity (as shown with the Michaelis constant KM), the catalytic turnover rate (kcat), and catalytic efficiency (kcat/KM) were each higher for cathepsin B at lower pH. The kcat rates measured with catalyCEST MRI were lower than the comparable rates measured with liquid chromatography-mass spectrometry (LC-MS), which reflected a limitation of inherently noisy and relatively insensitive CEST MRI analyses. Although this level of precision limited catalyCEST MRI to semiquantitative evaluations, these semiquantitative assessments of high and low protease activity still had value by demonstrating that high acidosis and high protease activity can be used as synergistic, multiparametric biomarkers.

Assessment of Early Response to Neoadjuvant Systemic Therapy in Triple-Negative Breast Cancer Using Amide Proton Transfer-weighted Chemical Exchange Saturation Transfer MRI: A Pilot Study.

Purpose To determine if amide proton transfer-weighted chemical exchange saturation transfer (APTW CEST) MRI is useful in the early assessment of treatment response in persons with triple-negative breast cancer (TNBC). Materials and Methods In this prospective study, a total of 51 participants (mean age, 51 years [range, 26-79 years]) with TNBC were included who underwent APTW CEST MRI with 0.9- and 2.0-µT saturation power performed at baseline, after two cycles (C2), and after four cycles (C4) of neoadjuvant systemic therapy (NAST). Imaging was performed between January 31, 2019, and November 11, 2019, and was a part of a clinical trial (registry number NCT02744053). CEST MR images were analyzed using two methods-magnetic transfer ratio asymmetry (MTRasym) and Lorentzian line shape fitting. The APTW CEST signals at baseline, C2, and C4 were compared for 51 participants to evaluate the saturation power levels and analysis methods. The APTW CEST signals and their changes during NAST were then compared for the 26 participants with pathology reports for treatment response assessment. Results A significant APTW CEST signal decrease was observed during NAST when acquisition at 0.9-µT saturation power was paired with Lorentzian line shape fitting analysis and when the acquisition at 2.0 µT was paired with MTRasym analysis. Using 0.9-µT saturation power and Lorentzian line shape fitting, the APTW CEST signal at C2 was significantly different from baseline in participants with pathologic complete response (pCR) (3.19% vs 2.43%; P = .03) but not with non-pCR (2.76% vs 2.50%; P > .05). The APTW CEST signal change was not significant between pCR and non-pCR at all time points. Conclusion Quantitative APTW CEST MRI depended on optimizing acquisition saturation powers and analysis methods. APTW CEST MRI monitored treatment effects but did not differentiate participants with TNBC who had pCR from those with non-pCR. © RSNA, 2021 Clinical trial registration no. NCT02744053 Supplemental material is available for this article.Keywords Molecular Imaging-Cancer, Molecular Imaging-Clinical Translation, MR-Imaging, Breast, Technical Aspects, Tumor Response, Technology Assessment.

Diffusion-weighted MR imaging histogram analysis in HIV positive and negative patients with primary central nervous system lymphoma as a predictor of outcome and tumor proliferation.

INTRODUCTION: Ki-67 expression, a marker of tumor proliferation, is considered a prognostic factor in primary CNS lymphoma (PCNSL). Apparent diffusion coefficient (ADC) parameters have also been proposed as imaging biomarkers for tumor progression and proliferative activity in various malignancies. The aim of this study is to investigate the correlation between ADC parameters, Ki-67 expression, overall survival (OS) and progression free survival (PFS) in PCNSL. MATERIALS AND METHODS: Patients diagnosed with PCNSL at MD Anderson Cancer Center between Mar 2000 and Jul 2016 and at Ben Taub Hospital between Jan 2012 and Dec 2016 were retrospectively studied. Co-registered ADC maps and post-contrast images underwent whole tumor segmentation. Normalized ADC parameters (nADC) were calculated as the ratio to normal white matter. Percentiles of nADC were calculated and were correlated with Ki-67 using Pearson's correlation coefficient and clinical outcomes (OS and PFS) using Cox proportional hazards models. RESULTS: Selection criteria yielded 90 patients, 23 patients living with HIV (PLWH) and 67 immunocompetent patients. Above median values for nADCmean, nADC15, nADC75 and nADC95 were associated with improved OS in all patients (p < 0.05). Above median values for nADCmin, nADCmean, nADC1, nADC5 and kurtosis were associated with improved PFS in all patients (p < 0.05). In patients with available Ki-67 expression data (n = 22), nADCmean, nADC15 and nADC75 inversely correlated with Ki-67 expression (p < 0.05). For PLWH, there was no correlation between ADC parameters and Ki-67 expression or clinical outcomes. CONCLUSIONS: ADC histogram analysis can predict tumor proliferation and survival in immunocompetent patients with PCNSL, but with limited utility in PLWH.

Magnetic Resonance-Based Radiomic Analysis of Radiofrequency Lesion Predicts Outcomes After Percutaneous Cordotomy: A Feasibility Study.

BACKGROUND: To date, there is limited data on evaluation of the cordotomy lesion and predicting clinical outcome. OBJECTIVE: To evaluate the utility of magnetic resonance (MR)-based radiomic analysis to quantify microstructural changes created by the cordotomy lesion and predict outcome in patients undergoing percutaneous cordotomy for medically refractory cancer pain. METHODS: This is a retrospective interpretation of prospectively acquired data in 10 patients (5 males, age range 43-76 yr) who underwent percutaneous computed tomography-guided high cervical cordotomy for medically refractory cancer pain between 2015 and 2016. All patients underwent magnetic resonance imaging (MRI) of the cordotomy lesion on postoperative day 1. After segmentation of T2-weighted images, 310 radiomic features were extracted. Pain outcomes were recorded on postoperative day 1 and day 7 using the visual analog scale. R software was used to build statistical models based on MRI radiomic features for prediction of pain outcomes. RESULTS: A total of 20 relevant radiomic features were identified using the maximum relevance minimum redundanc method. Radiomics predicted postoperative day 1 pain scores with an accuracy of 90% (P = .046), 100% sensitivity, 75% specificity, 85.7% positive predictive value, and 100% negative predictive value. The radiomics model also predicted if the postoperative day 1 pain score was sustained on postoperative day 7 with an accuracy of 100% (P = .028), 100% sensitivity, 100% specificity, and 100% positive and negative predictive value. CONCLUSION: MR-based radiomic analysis of the cordotomy lesion was predictive of pain outcomes at 1 wk after percutaneous cordotomy for intractable cancer pain.

Federated learning in medicine: facilitating multi-institutional collaborations without sharing patient data.

Several studies underscore the potential of deep learning in identifying complex patterns, leading to diagnostic and prognostic biomarkers. Identifying sufficiently large and diverse datasets, required for training, is a significant challenge in medicine and can rarely be found in individual institutions. Multi-institutional collaborations based on centrally-shared patient data face privacy and ownership challenges. Federated learning is a novel paradigm for data-private multi-institutional collaborations, where model-learning leverages all available data without sharing data between institutions, by distributing the model-training to the data-owners and aggregating their results. We show that federated learning among 10 institutions results in models reaching 99% of the model quality achieved with centralized data, and evaluate generalizability on data from institutions outside the federation. We further investigate the effects of data distribution across collaborating institutions on model quality and learning patterns, indicating that increased access to data through data private multi-institutional collaborations can benefit model quality more than the errors introduced by the collaborative method. Finally, we compare with other collaborative-learning approaches demonstrating the superiority of federated learning, and discuss practical implementation considerations. Clinical adoption of federated learning is expected to lead to models trained on datasets of unprecedented size, hence have a catalytic impact towards precision/personalized medicine.

Measuring Efficiency of Semi-automated Brain Tumor Segmentation by Simulating User Interaction.

Traditionally, radiologists have crudely quantified tumor extent by measuring the longest and shortest dimension by dragging a cursor between opposite boundary points across a single image rather than full segmentation of the volumetric extent. For algorithmic-based volumetric segmentation, the degree of radiologist experiential involvement varies from confirming a fully automated segmentation, to making a single drag on an image to initiate semi-automated segmentation, to making multiple drags and clicks on multiple images during interactive segmentation. An experiment was designed to test an algorithm that allows various levels of interaction. Given the ground-truth of the BraTS training data, which delimits the brain tumors of 285 patients on multi-spectral MR, a computer simulation mimicked the process that a radiologist would follow to perform segmentation with real-time interaction. Clicks and drags were placed only where needed in response to the deviation between real-time segmentation results and assumed radiologist's goal, as provided by the ground-truth. Results of accuracy for various levels of interaction are presented along with estimated elapsed time, in order to measure efficiency. Average total elapsed time, including loading the study through confirming 3D contours, was 46 s.

Phase I study of intraventricular infusions of autologous ex vivo expanded NK cells in children with recurrent medulloblastoma and ependymoma.

BACKGROUND: Recurrent pediatric medulloblastoma and ependymoma have a grim prognosis. We report a first-in-human, phase I study of intraventricular infusions of ex vivo expanded autologous natural killer (NK) cells in these tumors, with correlative studies. METHODS: Twelve patients were enrolled, 9 received protocol therapy up to 3 infusions weekly, in escalating doses from 3 × 106 to 3 × 108 NK cells/m2/infusion, for up to 3 cycles. Cerebrospinal fluid (CSF) was obtained for cellular profile, persistence, and phenotypic analysis of NK cells. Radiomic characterization on pretreatment MRI scans was performed in 7 patients, to develop a non-invasive imaging-based signature. RESULTS: Primary objectives of NK cell harvest, expansion, release, and safety of 112 intraventricular infusions of NK cells were achieved in all 9 patients. There were no dose-limiting toxicities. All patients showed progressive disease (PD), except 1 patient showed stable disease for one month at end of study follow-up. Another patient had transient radiographic response of the intraventricular tumor after 5 infusions of NK cell before progressing to PD. At higher dose levels, NK cells increased in the CSF during treatment with repetitive infusions (mean 11.6-fold). Frequent infusions of NK cells resulted in CSF pleocytosis. Radiomic signatures were profiled in 7 patients, evaluating ability to predict upfront radiographic changes, although they did not attain statistical significance. CONCLUSIONS: This study demonstrated feasibility of production and safety of intraventricular infusions of autologous NK cells. These findings support further investigation of locoregional NK cell infusions in children with brain malignancies.

Whole Tumor Histogram Analysis Using DW MRI in Primary Central Nervous System Lymphoma Correlates with Tumor Biomarkers and Outcome.

The ability to non-invasively predict outcomes and monitor treatment response in primary central nervous system lymphoma (PCNSL) is important as treatment regimens are constantly being trialed. The aim of this study was to assess the validity of using apparent diffusion coefficient (ADC) histogram values to predict Ki-67 expression, a tumor proliferation marker, and patient outcomes in PCNSL in both immunocompetent patients and patients living with HIV (PLWH). Qualitative PCNSL magnetic resonance imaging (MRI) characteristics from 93 patients (23 PLWH and 70 immunocompetent) were analyzed, and whole tumor segmentation was performed on the ADC maps. Quantitative histogram analyses of the segmentations were calculated. These measures were compared to PCNSL Ki-67 expression. Progression-free survival (PFS) and overall survival (OS) were analyzed via comparison to the International Primary Central Nervous System Lymphoma Collaboration Group Response Criteria. Associations between ADC measures and clinical outcomes were assessed using univariate and multivariate Cox proportional hazards models. Normalized ADC (nADC)Min, nADCMean, nADC1, nADC5, and nADC15 values were significantly associated with a poorer OS. nADCMax, nADCMean, nADC5, nADC15, nADC75, nADC95, nADC99 inversely correlated with Ki-67 expression. OS was also significantly associated with lesion hemorrhage. PFS was not significantly associated with ADC values but with lesion hemorrhage. ADC histogram values and related parameters can predict the degree of tumor proliferation and patient outcomes for primary central nervous system lymphoma patients and in both immunocompetent patients and patients living with HIV.

Associations of amygdala volume and shape with transactive response DNA-binding protein 43 (TDP-43) pathology in a community cohort of older adults.

Transactive response DNA-binding protein 43 (TDP-43) pathology is common in old age and is strongly associated with cognitive decline and dementia above and beyond contributions from other neuropathologies. TDP-43 pathology in aging typically originates in the amygdala, a brain region also affected by other age-related neuropathologies such as Alzheimer's pathology. The purpose of this study was two-fold: to determine the independent effects of TDP-43 pathology on the volume, as well as shape, of the amygdala in a community cohort of older adults, and to determine the contribution of amygdala volume to the variance of the rate of cognitive decline after accounting for the contributions of neuropathologies and demographics. Cerebral hemispheres from 198 participants of the Rush Memory and Aging Project and the Religious Orders Study were imaged with MRI ex vivo and underwent neuropathologic examination. Measures of amygdala volume and shape were extracted for all participants. Regression models controlling for neuropathologies and demographics showed an independent negative association of TDP-43 with the volume of the amygdala. Shape analysis revealed a unique pattern of amygdala deformation associated with TDP-43 pathology. Finally, mixed-effects models showed that amygdala volume explained an additional portion of the variance of the rate of decline in global cognition, episodic memory, semantic memory, and perceptual speed, above and beyond what was explained by demographics and neuropathologies.

Multicenter study demonstrates radiomic features derived from magnetic resonance perfusion images identify pseudoprogression in glioblastoma.

Pseudoprogression (PsP) is a diagnostic clinical dilemma in cancer. In this study, we retrospectively analyse glioblastoma patients, and using their dynamic susceptibility contrast and dynamic contrast-enhanced perfusion MRI images we build a classifier using radiomic features obtained from both Ktrans and rCBV maps coupled with support vector machines. We achieve an accuracy of 90.82% (area under the curve (AUC) = 89.10%, sensitivity = 91.36%, 67 specificity = 88.24%, p = 0.017) in differentiating between pseudoprogression (PsP) and progressive disease (PD). The diagnostic performances of the models built using radiomic features from Ktrans and rCBV separately were equally high (Ktrans: AUC = 94%, 69 p = 0.012; rCBV: AUC = 89.8%, p = 0.004). Thus, this MR perfusion-based radiomic model demonstrates high accuracy, sensitivity and specificity in discriminating PsP from PD, thus provides a reliable alternative for noninvasive identification of PsP versus PD at the time of clinical/radiologic question. This study also illustrates the successful application of radiomic analysis as an advanced processing step on different MR perfusion maps.

A Coclinical Radiogenomic Validation Study: Conserved Magnetic Resonance Radiomic Appearance of Periostin-Expressing Glioblastoma in Patients and Xenograft Models.

PURPOSE: Radiomics is the extraction of multidimensional imaging features, which when correlated with genomics, is termed radiogenomics. However, radiogenomic biological validation is not sufficiently described in the literature. We seek to establish causality between differential gene expression status and MRI-extracted radiomic-features in glioblastoma. EXPERIMENTAL DESIGN: Radiogenomic predictions and validation were done using the Cancer Genome Atlas and Repository of Molecular Brain Neoplasia Data glioblastoma patients (n = 93) and orthotopic xenografts (OX; n = 40). Tumor phenotypes were segmented, and radiomic-features extracted using the developed radiome-sequencing pipeline. Patients and animals were dichotomized on the basis of Periostin (POSTN) expression levels. RNA and protein levels confirmed RNAi-mediated POSTN knockdown in OX. Total RNA of tumor cells isolated from mouse brains (knockdown and control) was used for microarray-based expression profiling. Radiomic-features were utilized to predict POSTN expression status in patient, mouse, and interspecies. RESULTS: Our robust pipeline consists of segmentation, radiomic-feature extraction, feature normalization/selection, and predictive modeling. The combination of skull stripping, brain-tissue focused normalization, and patient-specific normalization are unique to this study, providing comparable cross-platform, cross-institution radiomic features. POSTN expression status was not associated with qualitative or volumetric MRI parameters. Radiomic features significantly predicted POSTN expression status in patients (AUC: 76.56%; sensitivity/specificity: 73.91/78.26%) and OX (AUC: 92.26%; sensitivity/specificity: 92.86%/91.67%). Furthermore, radiomic features in OX were significantly associated with patients with similar POSTN expression levels (AUC: 93.36%; sensitivity/specificity: 82.61%/95.74%; P = 02.021E-15). CONCLUSIONS: We determined causality between radiomic texture features and POSTN expression levels in a preclinical model with clinical validation. Our biologically validated radiomic pipeline also showed the potential application for human-mouse matched coclinical trials.