Search

Brain cancers pose a novel set of difficulties due to the limited accessibility of human brain tumor tissue. For this reason, clinical decision-making relies heavily on MR imaging interpretation, yet the mapping between MRI features and underlying biology remains ambiguous. Standard tissue sampling fails to capture the full heterogeneity of the disease. Biopsies are required to obtain a pathological diagnosis and are predominantly taken from the tumor core, which often has different traits to the surrounding invasive tumor that typically leads to recurrent disease. One approach to solving this issue is to characterize the spatial heterogeneity of molecular, genetic, and cellular features of glioma through the intraoperative collection of multiple image-localized biopsy samples paired with multi-parametric MRIs. We have adopted this approach and are currently actively enrolling patients for our ‘Image-Based Mapping of Brain Tumors’ study. Patients are eligible for this research study (IRB #16-002424) if they are 18 years or older and undergoing surgical intervention for a brain lesion. Once identified, candidate patients receive dynamic susceptibility contrast (DSC) perfusion MRI and diffusion tensor imaging (DTI), in addition to standard sequences (T1, T1Gd, T2, T2-FLAIR) at their presurgical scan. During surgery, sample locations are tracked using neuronavigation and genetic aberrations are later quantified through whole-exome and RNA sequencing. The collected specimens from this NCI-funded research study will be primarily used to generate regional maps of the spatial distribution of tumor cell density and/or treatment-related key genetic marker status across tumors, within clinically feasible time frames, to identify biopsy and/or treatment targets based on insight from the entire tumor makeup regional histologic and genetic makeup. This type of methodology, when delivered within clinically feasible time frames, has the potential to further inform medical decision-making by improving surgical intervention, radiation, and targeted drug therapy for patients with glioma. From October 1, 2017 to October 31, 2022, this study has enrolled 186 patients with 197 surgeries, of which 163 resulted in the successful collection of image-guided biopsy samples. A total of 995 biopsies have been collected of which 962 are image localized, with a mean of 5.90 image-localized samples per surgery.

Automatic brain tumor segmentation is particularly challenging on magnetic resonance imaging (MRI) with marked pathologies, such as brain tumors, which usually cause large displacement, abnormal appearance, and deformation of brain tissue. Despite an abundance of previous literature on learning-based methodologies for MRI segmentation, few works have focused on tackling MRI skull stripping of brain tumor patient data. This gap in literature can be associated with the lack of publicly available data (due to concerns about patient identification) and the labor-intensive nature of generating ground truth labels for model training. In this retrospective study, we assessed the performance of Dense-Vnet in skull stripping brain tumor patient MRI trained on our large multi-institutional brain tumor patient dataset. Our data included pretreatment MRI of 668 patients from our in-house institutional review board–approved multi-institutional brain tumor repository. Because of the absence of ground truth, we used imperfect automatically generated training labels using SPM12 software. We trained the network using common MRI sequences in oncology: T1-weighted with gadolinium contrast, T2-weighted fluid-attenuated inversion recovery, or both. We measured model performance against 30 independent brain tumor test cases with available manual brain masks. All images were harmonized for voxel spacing and volumetric dimensions before model training. Model training was performed using the modularly structured deep learning platform NiftyNet that is tailored toward simplifying medical image analysis. Our proposed approach showed the success of a weakly supervised deep learning approach in MRI brain extraction even in the presence of pathology. Our best model achieved an average Dice score, sensitivity, and specificity of, respectively, 94.5, 96.4, and 98.5% on the multi-institutional independent brain tumor test set. To further contextualize our results within existing literature on healthy brain segmentation, we tested the model against healthy subjects from the benchmark LBPA40 dataset. For this dataset, the model achieved an average Dice score, sensitivity, and specificity of 96.2, 96.6, and 99.2%, which are, although comparable to other publications, slightly lower than the performance of models trained on healthy patients. We associate this drop in performance with the use of brain tumor data for model training and its influence on brain appearance.

Glioblastomas (GBMs) are biologically heterogeneous within and between patients. Many previous attempts to characterize this heterogeneity have classified tumors according to their omics similarities. These discrete classifications have predominantly focused on characterizing malignant cells, neglecting the immune and other cell populations that are known to be present. We leverage a manifold learning algorithm to define a low-dimensional transcriptional continuum along which heterogeneous GBM samples organize. This reveals three polarized states: invasive, immune/inflammatory, and proliferative. The location of each sample along this continuum correlates with the abundance of eighteen malignant, immune, and other cell populations. We connect these cell abundances with magnetic resonance imaging and find that the relationship between contrast enhancement and tumor composition varies with patient sex and treatment status. These findings suggest that GBM transcriptional biology is a predictably constrained continuum that contains a limited spectrum of viable cell cohabitation ecologies. Since the relationships between this ecological continuum and imaging vary with patient sex and tumor treatment status, studies that integrate imaging features with tumor biology should incorporate these variables in their design.

BACKGROUND AND PURPOSE: Perfusion MR imaging measures of relative CBV can distinguish recurrent tumor from posttreatment radiation effects in high-grade gliomas. Currently, relative CBV measurement requires normalization based on user-defined reference tissues. A recently proposed method of relative CBV standardization eliminates the need for user input. This study compares the predictive performance of relative CBV standardization against relative CBV normalization for quantifying recurrent tumor burden in high-grade gliomas relative to posttreatment radiation effects. MATERIALS AND METHODS: We recruited 38 previously treated patients with high-grade gliomas (World Health Organization grades III or IV) undergoing surgical re-resection for new contrast-enhancing lesions concerning for recurrent tumor versus posttreatment radiation effects. We recovered 112 image-localized biopsies and quantified the percentage of histologic tumor content versus posttreatment radiation effects for each sample. We measured spatially matched normalized and standardized relative CBV metrics (mean, median) and fractional tumor burden for each biopsy. We compared relative CBV performance to predict tumor content, including the Pearson correlation (r), against histologic tumor content (0%–100%) and the receiver operating characteristic area under the curve for predicting high-versus-low tumor content using binary histologic cutoffs (≥50%; ≥80% tumor). RESULTS: Across relative CBV metrics, fractional tumor burden showed the highest correlations with tumor content (0%–100%) for normalized (r = 0.63, P

INTRODUCTION Dendritic cells (DCs) are potent antigen presenting cells that can be exploited to initiate an adaptive anti-tumoral immune response. DC vaccine clinical trials for primary glioblastoma (GBM) have reported prolonged progression-free survival without any impact on overall survival (OS). We report a radiomics approach that identifies a subpopulation of patients with prolonged OS in clinical trial MC1272. METHODS Twenty adults with primary GBM undergoing standard-of-care therapy were enrolled in MC1272. Autologous DCs were pulsed with allogenic GBM cell lines to generate vaccines that were administered for up to twelve cycles. Standard brain imaging was obtained at the initiation of treatment and two months afterwards. An independent cohort of image-localized biopsies underwent RNA sequencing followed by cellular deconvolution to estimate T cell abundance. A machine learning model was trained to predict intratumoral T cell abundance from imaging features, and the model was applied to MC1272 patient imaging. RESULTS Voxelwise predictions of T cell abundance were generated for each patient’s pre- and post-treatment images. The difference in total intratumoral T cell abundance between imaging timepoints classified patients into increasing or decreasing T cell groups. Patients whose T cells increased had longer OS (median, 21 months) than those whose T cells decreased (median, 10 months; p=0.0035). The significance remained in a Cox proportional hazards analysis that adjusted for patient age and sex (p=0.011). CONCLUSIONS A spatially-resolved radiomics model detected that an intratumoral T cell influx after DC vaccine therapy was associated with prolonged OS. The “post-treatment” imaging in this study was obtained a mere two months after DC vaccine initiation, suggesting that our radiomics model can provide an early indication of treatment responsiveness and prognosis in these patients.

Intra-tumor genetic heterogeneity is an important cause of treatment failure of GBM. Using MRI and image-localized biopsies, it is possible to train machine learning (ML) models to predict regional genetic status. However, biopsy samples are limited, making it difficult to train a robust ML model. We proposed a data-inclusive model called Weakly Supervised Ordinal Support Vector Machines (WSO-SVM) which leverages the vast amount of MRI data outside the sparsely sampled biopsy regions to augment the biopsy samples to improve ML accuracy. Our study included a unique dataset of 104 image-localized biopsies with spatially matched multiparametric MRI from 30 untreated Glioblastoma (GBM) patients. Each biopsy sample went through genetic sequencing analysis and our study focused on two GBM hallmark genes, EGFR and PDGFRA. For each gene, a biopsy sample was labeled as “altered” if the copy number of this gene was amplified or a mutation was found, and “non-altered” otherwise. From the localized region of six MRI contrasts from T1gd, T2w, diffusion and perfusion imaging, over 300 texture features were extracted. To account for biopsy sample location uncertainty, six neighboring regions of the biopsy sample including four neighbors two pixels away from the biopsy location and two neighbors on adjacent slices were also included in model training. WSO-SVM achieved 0.83 accuracy, 0.77 sensitivity, and 0.86 specificity for classifying EGFR; 0.77 accuracy, 0.74 sensitivity, and 0.79 specificity for classifying PDGFRA, based on 10-fold cross validation. Furthermore, using the trained models, we generated regional EGFR and PDGFRA alteration maps for each patient within the enhancing and non-enhancing tumoral areas. On average we found a greater proportion of enhancing tumor with co-alteration than non-enhancing tumor, while this trend was reversed when considering only one altered gene. The ratio of EGFR vs PDGFRA alterations was higher in non-enhancing tumor than enhancing tumor.

Glioblastoma (GBM) is a devastating primary brain tumor known for its heterogeneity with a median survival of 15 months. Clinical imaging remains the primary modality to assess brain tumor response, but it is nearly impossible to distinguish between tumor growth and treatment response. Ki67 is a marker of active cell proliferation that shows inter- and intra-patient heterogeneity and should change under many therapies. In this work, we assessed the utility of a semi-supervised deep learning approach for regionally predicting high-vs-low Ki67 in GBM patients based on MRI. We used both labeled and unlabeled datasets to train the model. Labeled data included 114 MRI-localized biopsies from 43 unique GBM patients with available immunohistochemistry Ki67 labels. Unlabeled data included nine repeat routine pretreatment paired scans of newly-diagnosed GBM patients acquired within three days. Data augmentation techniques were utilized to enhance the size of our data and increase generalizability. Data was split between training, validation, and testing sets using 65-15-20 percent ratios. Model inputs were 16x16x3 patches around biopsies on T1Gd and T2 MRIs for labeled data, and around randomly selected patches inside the T2 abnormal region for unlabeled data. The network was a 4-conv layered VGG-inspired architecture. Training objective was accurate prediction of Ki67 in labeled patches and consistency in predictions across repeat unlabeled patches. We measured final model accuracy on held-out test samples. Our promising preliminary results suggest potential for deep learning in deconvolving the spatial heterogeneity of proliferative GBM subpopulations. If successful, this model can provide a non-invasive readout of cell proliferation and reveal the effectiveness of a given cytotoxic therapy dynamically during the patient's routine follow up. Further, the spatial resolution of our approach provides insights into the intra-tumoral heterogeneity of response which can be related to heterogeneity in localization of therapies (e.g. radiation therapy, drug dose heterogeneity).

Choosing effective chemotherapies for intravenous delivery to brain tumors is challenging, especially given the protective nature of the blood brain barrier (BBB). Connecting drug distribution to non-invasive, pre-surgical magnetic resonance imaging (MRI) could allow for predictive insight into drug distribution. In a previous study, we found that T2Gd images were predictive of a low BBB penetrant drug (Cefazolin), and FLAIR images were predictive of a high BBB penetrant drug (Levetiracetam). While these results are promising, we further seek to explore how advanced MRI sequences might inform image-based models of drug distribution. Prior to surgery, we acquired advanced dynamic contrast enhanced (DCE) and diffusion weighted imaging (DWI) MRI sequences for eight brain tumor patients (7 gliomas and 1 metastatic adenocarcinoma) in addition to the anatomic MRIs. All resulting quantitative maps and acquired images were co-registered. Prior to incision, patients received injections of cefazolin and levetiracetam. Next, multiple blood samples and biopsies were collected during surgery. Biopsies and plasma samples were analyzed for drug concentration using liquid chromatography mass spectrometry (LCMS), and biopsy drug levels were reported as Brain-Plasma Ratio (BPR). Mean image intensity was extracted from a 15x15 voxel window surrounding the biopsy location. We performed linear regression analyses to determine which combination of images were predictive of BPR. We found that considering quantitative imaging improved our initial ability to predict BPR for both drugs. For cefazolin, the third diffusion tensor eigenvalue (L3) map was significantly correlated with BPR (p< 0.001, R2= 0.36). For levetiracetam, the best model consisted of a combination of images and maps with the L3 map and the isotropic diffusion map (P) being the most influential (p= 0.001, R2= 0.63). Advanced MRI-based modeling is a promising tool for forecasting drug distribution in brain tumors and could be of great importance for understanding efficacy and selecting therapeutic strategies.

OBJECTIVE Glioblastoma (GBM) is the most common malignant primary brain tumor in adults, with males more commonly affected than females(1.6:1). Despite advancements in treatments, prognosis is dismal with a median overall survival of 15 months. Our aim was to investigate sex as a variable in GBM patient survival after receiving incremental levels of standard-of-care treatment regimens – different extents of surgical resection and different numbers of cycles of adjuvant temozolomide chemotherapy. METHODS Drawing from our extensive multi-institutional brain tumor repository, we investigated GBM subjects with overall survival (OS), extent of resection (EOR), number of temozolomide (TMZ) cycles, and sex data (n=620, males: n=387, females: n=233). Cox proportional hazard ratios were computed to investigate the multivariable predictive value of the patient variables with OS. Patients were then divided into groups based on their sex, EOR (either biopsy, subtotal resection (STR) or gross total resection (GTR)), and TMZ cycles (I: < 6 cycles, II: 7-11 cycles and III: >12 cycles). RESULTS We observed that STR was beneficial for females (HR=0.52; CI=0.33-0.83; p-value=0.013), while for males the benefit was not detected (HR=0.73; CI=0.46-1.15; p-value=0.173) for STR but was detectable for GTR (HR=0.58, CI=0.37-0.90; p-value=0.014). Females receiving 7-11 cycles of TMZ showed a survival benefit (HR=0.52; CI=0.12-0.53; p-value=0.048) while males in the same group did not (HR=0.74; CI=0.46-1.19; p-value=0.21), in comparison to those in group I of TMZ cycles. No sex differences were identified in patients receiving < =6 cycles or >=12 cycles. CONCLUSION Together, our results contribute to the growing literature that sex differences exist in GBM patients, even in response to standard-of-care therapies. This should be accounted for when designing clinical trials for GBM so that we may advance our pursuit to deliver personalized medicine.

Purpose: Deep learning (DL) algorithms have shown promising results for brain tumor segmentation in MRI. However, validation is required prior to routine clinical use. We report the first randomized and blinded comparison of DL and trained technician segmentations. Approach: We compiled a multi-institutional database of 741 pretreatment MRI exams. Each contained a postcontrast T1-weighted exam, a T2-weighted fluid-attenuated inversion recovery exam, and at least one technician-derived tumor segmentation. The database included 729 unique patients (470 males and 259 females). Of these exams, 641 were used for training the DL system, and 100 were reserved for testing. We developed a platform to enable qualitative, blinded, controlled assessment of lesion segmentations made by technicians and the DL method. On this platform, 20 neuroradiologists performed 400 side-by-side comparisons of segmentations on 100 test cases. They scored each segmentation between 0 (poor) and 10 (perfect). Agreement between segmentations from technicians and the DL method was also evaluated quantitatively using the Dice coefficient, which produces values between 0 (no overlap) and 1 (perfect overlap). Results: The neuroradiologists gave technician and DL segmentations mean scores of 6.97 and 7.31, respectively (p

Background Accurate assessments of patient response to therapy are a critical component of personalized medicine. In glioblastoma (GBM), the most aggressive form of brain cancer, tumor growth dynamics are heterogenous across patients, complicating assessment of treatment response. This study aimed to analyze days gained (DG), a burgeoning model-based dynamic metric, for response assessment in patients with recurrent GBM who received bevacizumab-based therapies. Methods DG response scores were calculated using volumetric tumor segmentations for patients receiving bevacizumab with and without concurrent cytotoxic therapy (N = 62). Kaplan–Meier and Cox proportional hazards analyses were implemented to examine DG prognostic relationship to overall (OS) and progression-free survival (PFS) from the onset of treatment for recurrent GBM. Results In patients receiving concurrent bevacizumab and cytotoxic therapy, Kaplan–Meier analysis showed significant differences in OS and PFS at DG cutoffs consistent with previously identified values from newly diagnosed GBM using T1-weighted gadolinium-enhanced magnetic resonance imaging (T1Gd). DG scores for bevacizumab monotherapy patients only approached significance for PFS. Cox regression showed that increases of 25 DG on T1Gd imaging were significantly associated with a 12.5% reduction in OS hazard for concurrent therapy patients and a 4.4% reduction in PFS hazard for bevacizumab monotherapy patients. Conclusion DG has significant meaning in recurrent therapy as a metric of treatment response, even in the context of anti-angiogenic therapies. This provides further evidence supporting the use of DG as an adjunct response metric that quantitatively connects treatment response and clinical outcomes.

BACKGROUNDRadiogenomics uses machine-learning (ML) to directly connect the morphologic and physiological appearance of tumors on clinical imaging with underlying genomic features. Despite extensive growth in the area of radiogenomics across many cancers, and its potential role in advancing clinical decision making, no published studies have directly addressed uncertainty in these model predictions.METHODSWe developed a radiogenomics ML model to quantify uncertainty using transductive Gaussian Processes (GP) and a unique dataset of 95 image-localized biopsies with spatially matched MRI from 25 untreated Glioblastoma (GBM) patients. The model generated predictions for regional EGFR amplification status (a common and important target in GBM) to resolve the intratumoral genetic heterogeneity across each individual tumor - a key factor for future personalized therapeutic paradigms. The model used probability distributions for each sample prediction to quantify uncertainty, and used transductive learning to reduce the overall uncertainty. We compared predictive accuracy and uncertainty of the transductive learning GP model against a standard GP model using leave-one-patient-out cross validation.RESULTSPredictive uncertainty informed the likelihood of achieving an accurate sample prediction. When stratifying predictions based on uncertainty, we observed substantially higher performance in the group cohort (75% accuracy, n=95) and amongst sample predictions with the lowest uncertainty (83% accuracy, n=72) compared to predictions with higher uncertainty (48% accuracy, n=23), due largely to data interpolation (rather than extrapolation).CONCLUSIONWe present a novel approach to quantify radiogenomics uncertainty to enhance model performance and clinical interpretability. This should help integrate more reliable radiogenomics models for improved medical decision-making.

BackgroundTemozolomide (TMZ) has been the standard-of-care chemotherapy for glioblastoma (GBM) patients for more than a decade. Despite this long time in use, significant questions remain regarding how best to optimize TMZ therapy for individual patients. Understanding the relationship between TMZ response and factors such as number of adjuvant TMZ cycles, patient age, patient sex, and image-based tumor features, might help predict which GBM patients would benefit most from TMZ, particularly for those whose tumors lack O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation.Methods and findingsUsing a cohort of 90 newly-diagnosed GBM patients treated according to the standard of care, we examined the relationships between several patient and tumor characteristics and volumetric and survival outcomes during adjuvant chemotherapy. Volumetric changes in MR imaging abnormalities during adjuvant therapy were used to assess TMZ response. T1Gd volumetric response is associated with younger patient age, increased number of TMZ cycles, longer time to nadir volume, and decreased tumor invasiveness. Moreover, increased adjuvant TMZ cycles corresponded with improved volumetric response only among more nodular tumors, and this volumetric response was associated with improved survival outcomes. Finally, in a subcohort of patients with known MGMT methylation status, methylated tumors were more diffusely invasive than unmethylated tumors, suggesting the improved response in nodular tumors is not driven by a preponderance of MGMT methylated tumors.ConclusionsOur finding that less diffusely invasive tumors are associated with greater volumetric response to TMZ suggests patients with these tumors may benefit from additional adjuvant TMZ cycles, even for those without MGMT methylation.

INTRODUCTION Glioblastoma (GBM) is the most common malignant primary brain tumor in adults with a median overall survival (OS) of 15months. Despite advancements in treatments, prognosis is dismal and the prognostic significance of tumor location is not entirely understood. METHODOLOGY: In our study, we investigated sex-specific volumetric, tumor growth kinetics, and outcome differences among GBMs in various brain locations. Primary GBM patients with pretreatment magnetic resonance imaging (MRI) data (N=289, 173 males, 116 females) were selected from our brain tumor repository. Tumor abnormality was segmented on T1-weighted post-gadolinium contrast agent (T1Gd) MRIs. We utilized the Harvard-Oxford brain atlases to determine the location of GBMs. RESULTS Overall, our study found smaller tumors in the left hemisphere. This may be expected as left-hemispheric GBM symptoms could present earlier, leading to earlier diagnosis and treatment. However, when the cohort was split by sex, we found this observation significant for females only in the parietal lobe (p < 0.0001). Further, female GBMs demonstrated smaller necrotic volume in the left hemisphere (p = 0.030). Sex-specific differences in incidence were noted in the temporal and occipital lobes (2M:1F). Comparing tumor growth kinetics in different brain locations and hemispheres, females had significantly lower tumor proliferation rates in the left hemisphere (p = 0.009) and lower tumor proliferation rates in the left frontal lobe (p = 0.031). Controlling for treatment, patients with frontal lobe tumors had significantly longer OS compared to those with GBMs in the temporal lobe (p = 0.046, 312 days). Differences in growth velocities were noted between frontal and parietal lobe with frontal GBMs having lower velocities in comparison to parietal lobe GBMs. CONCLUSION Together, our results demonstrate that tumor growth and proliferation rates may vary based on location and sex. Additional research is needed to further explore the clinical significance of tumor location.

OBJECTIVES Dogma suggests that for brain tumors, regions of enhancement on T1-weighted gadolinium contrast enhanced (T1Gd) magnetic resonance imaging (MRI) correlate with intravenously delivered drug distribution as enhancement indicates a compromised blood-brain barrier (BBB). However, poor response to intravenous therapies highlights the importance of the diffuse disease beyond enhancing regions. This study investigated whether imaging features can provide an accurate prediction of drug distribution. METHODS Eight brain tumor patients (7 gliomas and 1 metastatic adenocarcinoma) were included in this Phase 0 trial. Presurgery T1-weighted, T1Gd, T2-weighted gadolinium contrast enhanced (T2Gd), and T2-weighted Fluid Attenuated Inversion Recovery (T2FLAIR) MRIs were acquired. All images underwent bias correction using the N4 algorithm, standardization of intensities, and registration. Prior to incision, patients received both an antibiotic cefazolin (6% BBB penetrance) and levetiracetam (80% BBB penetrance), an anti-seizure drug. Subsequently, multiple blood samples and image-guided biopsies were taken and analyzed for drug concentration using liquid chromatography mass spectrometry. Biopsy drug levels are reported as Brain-Plasma Ratio (BPR), the ratio of biopsy concentration relative to plasma concentration. Mean image intensity was extracted from an 8x8 mm window surrounding each biopsy location. Regression analysis was performed to determine which combination of image types were linearly predictive of BPR for both drugs. Correlations were also analyzed according to the biopsy location radiographic appearance. RESULTS Regression analysis revealed that T2Gd intensity was linearly predictive of cefazolin BPR and FLAIR intensity was linearly predictive of levetiracetam BPR (p=0.009 and 0.041, respectively). Grouping samples according the the radiographic appearance revealed that levetiracetam BPR had a similar pattern of values to that of FLAIR intensities and cefazolin BPR had a similar pattern to T2, further supporting the regression analysis results. CONCLUSIONS Local concentrations of drug may be related to T2-weighted signals (T2Gd and T2FLAIR) rather than the gadolinium distribution on T1Gd images.

The explosion of medical imaging data along with the advent of big data analytics has launched an exciting era for clinical research. One factor affecting the ability to aggregate large medical image collections for research is the lack of infrastructure for automated data annotation. Among all imaging modalities, annotation of magnetic resonance (MR) images is particularly challenging due to the non-standard labeling of MR image types. In this work, we aimed to train a deep neural network to annotate MR image sequence type for scans of brain tumor patients. We focused on the four most common MR sequence types within neuroimaging: T1-weighted (T1W), T1-weighted post-gadolinium contrast (T1Gd), T2-weighted (T2W), and T2-weighted fluid-attenuated inversion recovery (FLAIR). Our repository contains images acquired using a variety of pulse sequences, sequence parameters, field strengths, and scanner manufacturers. Image selection was agnostic to patient demographics, diagnosis, and the presence of tumor in the imaging field of view. We used a total of 14,400 two-dimensional images, each visualizing a different part of the brain. Data was split into train, validation, and test sets (9600, 2400, and 2400 images, respectively) and sets consisted of equal-sized groups of image types. Overall, the model reached an accuracy of 99% on the test set. Our results showed excellent performance of deep learning techniques in predicting sequence types for brain tumor MR images. We conclude deep learning models can serve as tools to support clinical research and facilitate efficient database management.

Seizures are common presenting symptoms of primary brain tumors. Mechanisms of epileptogenesis are still unknown and are believed to be multifactorial. Previous studies have indicated correlation of seizure with tumor location. Recent investigations of our group have shown image-based parameters have sex-specific implications for patient outcome. In this retrospective study, we examined the association of tumor location with the probability and risk of seizure in male and female glioma patients.

Glioblastoma (GBM) is a heterogeneous and lethal brain cancer. These tumors are followed using magnetic resonance imaging (MRI), which is unable to precisely identify tumor cell invasion, impairing effective surgery and radiation planning. We present a novel hybrid model, based on multiparametric intensities, which combines machine learning (ML) with a mechanistic model of tumor growth to provide spatially resolved tumor cell density predictions. The ML component is an imaging data-driven graph-based semi-supervised learning model and we use the Proliferation-Invasion (PI) mechanistic tumor growth model. We thus refer to the hybrid model as the ML-PI model. The hybrid model was trained using 82 image-localized biopsies from 18 primary GBM patients with pre-operative MRI using a leave-one-patient-out cross validation framework. A Relief algorithm was developed to quantify relative contributions from the data sources. The ML-PI model statistically significantly outperformed (p

BACKGROUND: MRI-based modeling of tumor cell density (TCD) can significantly improve targeted treatment of Glioblastoma (GBM). Unfortunately, interpatient variability limits the predictive ability of many modeling approaches. We present a Transfer Learning (TL) method that generates individualized patient models, grounded in the wealth of population data, while also detecting and adjusting for interpatient variabilities based on each patient’s own histologic data. METHODS: We recruited primary GBM patients undergoing image-guided biopsies and preoperative imaging including contrast-enhanced MRI (CE-MRI), Dynamic-Susceptibility-Contrast (DSC)-MRI, and Diffusion Tensor Imaging (DTI). We calculated relative cerebral blood volume (rCBV) from DSC-MRI and mean diffusivity (MD) and fractional anisotropy (FA) from DTI. Following image coregistration, we assessed TCD for each biopsy and identified corresponding localized MRI measurements. We then explored a range of univariate and multivariate predictive models of TCD based on MRI measurements in a generalized one-model-fits-all (OMFA) approach. We then implemented both univariate and multivariate individualized TL predictive models, which harness the available population level data but allow for individual variability in their predictions. Finally, we compared Pearson correlation coefficients and mean absolute error between the individualized TL versus generalized OMFA models. RESULTS: TCD significantly correlated with rCBV (r=0.33,p

Temozolomide (TMZ) has been the standard-of-care chemotherapy for glioblastoma (GBM) patients for more than a decade. Despite this long time in use, significant questions remain regarding how best to optimize TMZ therapy for individual patients. Understanding the relationship between TMZ response and factors such as number of adjuvant TMZ cycles, patient age, patient sex, and image-based tumor features, might help predict which GBM patients would benefit most from TMZ, particularly for those whose tumors are not MGMT methylated. Using a cohort of 90 newly-diagnosed GBM patients treated according to the Stupp protocol, we examined the relationships between several patient and tumor characteristics and volumetric and survival outcomes during adjuvant chemotherapy. Volumetric changes in MR imaging abnormalities during adjuvant therapy were used to assess TMZ response. T1Gd volumetric response is associated with younger patient age, increased number of TMZ cycles, longer time to nadir volume, and decreased tumor invasiveness. Moreover, increased adjuvant TMZ cycles corresponded with improved volumetric response only among more nodular tumors, and this volumetric response was associated with improved survival outcomes. Finally, in a subcohort of patients with known MGMT methylation status, MGMT methylated tumors were more diffusely invasive than unmethylated tumors, suggesting that the improved response in nodular tumors is not driven by a preponderance of MGMT methylated tumors. Our finding that less diffusely invasive tumors are associated with greater volumetric response to TMZ suggests that patients with these tumors may benefit from additional cycles of adjuvant TMZ, even for those without MGMT methylation.

Sex differences in the incidence and outcome of human disease are broadly recognized but, in most cases, not sufficiently understood to enable sex-specific approaches to treatment. Glioblastoma (GBM), the most common malignant brain tumor, provides a case in point. Despite well-established differences in incidence and emerging indications of differences in outcome, there are few insights that distinguish male and female GBM at the molecular level or allow specific targeting of these biological differences. Here, using a quantitative imaging-based measure of response, we found that standard therapy is more effective in female compared with male patients with GBM. We then applied a computational algorithm to linked GBM transcriptome and outcome data and identified sex-specific molecular subtypes of GBM in which cell cycle and integrin signaling are the critical determinants of survival for male and female patients, respectively. The clinical relevance of cell cycle and integrin signaling pathway signatures was further established through correlations between gene expression and in vitro chemotherapy sensitivity in a panel of male and female patient-derived GBM cell lines. Together, these results suggest that greater precision in GBM molecular subtyping can be achieved through sex-specific analyses and that improved outcomes for all patients might be accomplished by tailoring treatment to sex differences in molecular mechanisms.

INTRODUCTION: Brain tumors are difficult to segment in MRI scans. Consequently, we have developed a new system for completely automatic brain tumor segmentation by combining a state-of-the-art 3D deep convolutional neural network (CNN) with a large collection of curated segmentations of brain tumors. METHODS: Our brain tumor database holds 74,722 MRI series from 2,742 unique patients. Over the last 15 years our image analysis team has segmented brain tumors in 35,710 of these series. This preliminary experiment identified 741 pre-treatment studies that included a T1GD and FLAIR scan, and at least one adjudicated brain tumor segmentation. These studies were randomly assigned into 600 training, 41 validation, and 100 test cases. CNN training was performed in two stages: 1) 50 epochs on minimally modified MRI volumes; and, 2) 24 epochs to tune the CNN on skull-stripped volumes. Whole-tumor Dice coefficients (1=perfect overlap, 0=no overlap) were calculated by comparing CNN segmentations against adjudicated segmentations from trained measurers. Training was performed in-the-cloud using an Amazon Machine Instance equipped with an NVidia Tesla V100 GPU, 8 Intel Xeon processors, and 64 GB of RAM. RESULTS: Training required 74 hours. Afterwards, our network required 800 seconds to segment 100 studies in the test set (8 seconds/study). The mean whole-tumor Dice coefficient on the test studies was 0.885. DISCUSSION: The best result on the highly cited 2017 BraTS brain tumor segmentation challenge was a whole-tumor Dice of 0.886, achieved by an ensemble of 7 CNNs. BraTS included 274 studies, each with T1GD, FLAIR, T1 and T2 contrasts. The performance of our single CNN may be due to our comparatively large training set. Our goal is to train our CNN on all series in our database. This may provide useful tools to monitor each patients journey, from diagnosis through treatment.

INTRODUCTION: While a growing body of research on the intrinsic biological differences between male and female tumors and their environments continues to evolve, few deep learning techniques have been trained with a sex-specific focus. To investigate the influence of sex differences on automated tumor segmentation, a set of Male and Female 3D deep convolutional neural networks were trained and evaluated (MaleDNN and FemaleDNN). METHODS: A balanced data set was obtained from our brain tumor database of 518 cases with known sex, pretreatment T1GD and FLAIR MRI, and at least one brain tumor segmentation. Cases were split by sex to create training cohorts of 200 male and female cases to train the MaleDNN and FemaleDNN. A set of 59 unseen test cases for each sex were reserved for evaluation. Both networks were used to segment tumor volumes from male and female test cases. Whole-tumor Dice coefficients (1=perfect overlap, 0=no overlap) were calculated by comparing network segmentations against segmentations from trained measurers. RESULTS: The MaleDNN had higher overall performance on all cases (Dice male=0.8416; female=0.7800) compared to the FemaleDNN (Dice male=0.8269, female=0.7639). The difference in performance between networks was significant for male tumors (p=0.0466), but not female tumors (p=0.1872). Both networks performed better on male tumors. Average dice scores were significantly lower for the MaleDNN (Dice decrease=0.0616, p=0.0273) and FemaleDNN (Dice decrease=0.629, p=0.0422) when evaluating female tumors. DISCUSSION: It was anticipated that the MaleDNN and FemaleDNN would have highest performance on test cases from the same sex. However, the FemaleDNN performed better on males than females and the MaleDNN had comparable performance for female tumors. The significant differences between these sex-specific deep neural networks indicate that male and female gliomas differ on imaging and that female tumors are more difficult to segment at initial presentation.

We show the application of a minimally based, patient-specific mathematical model in the evaluation of glioblastoma response to therapy. Days Gained uses computational models of glioblastoma growth dynamics derived from clinically acquired magnetic resonance imaging (MRI) to compare the post-treatment tumor lesion to the expected untreated tumor lesion at the same time point. It accounts for the inter-patient variability in growth dynamics and response to therapy. This allows for the accurate assessment of therapeutic response and provides insight into overall survival as it relates to treatment response.

BACKGROUND Conventional magnetic resonance imaging (MR) guides patient care in GBM. However, there is mounting awareness that MR enhancement is non-specific reflecting either tumor progression or non-tumoral inflammatory changes. Histological evaluation of GBM is held as the gold standard for disease assessment. However, the invasiveness of this methodology and the sample sparsity limit its usefulness. Methods to infer histological underpinnings of MRI are needed to improve clinical care. METHODS A transfer learning mixed effects model based on T1Gd and FLAIR MR voxel based image features trained and cross-validated to predict FPKM values of CD68, CASP3, CD8A and IL13RA2. Training data included RNAseq from 38 image-localized biopsies from 15 newly-diagnosed GBM patients. These models were then applied to two independent patients, chosen based on therapy and the quality of image registration, at three different time points, just prior to receiving IL13RA2 targeted CAR-T therapy for rGBM, after 2 cycles and after 4 cycles, resulting in a voxel-based prediction of the relative expression levels of these genes. RESULTS Cross-validation of the proposed machine learning models demonstrated a Pearson Correlation coefficient between predicted and observed FPKM values of 0.89 (CD68), 0.92 (CASP3), 0.93 (CD8A), and 0.88 (IL13RA2). When the models were applied to the two CAR-T patients, CD68-modeled expression levels were seen to increase throughout therapy for one patient, while remaining stable for the other. Survival from first CAR-T infusion was, respectively, 412 and 83 days suggesting that the increased CD68 was indicative of therapeutic effectivity. Further, the spatial activity predictions were consistent with expected therapeutic action as the correlation coefficient between CASP3 and the product of IL13R2 and CD8A expression was 0.81, suggesting the T-cells targeting IL13RA2 were inducing cell death. CONCLUSIONS Preliminary results with our model highlights the potential of spatially resolved radiomic maps to provide insight into regional therapeutic effectivity.

INTRODUCTION The quantification of intratumoral heterogeneity – through radiomics-based approaches - can help resolve the regionally distinct genetic drug targets that may co-exist within a single Glioblastoma (GBM) tumor. While this offers potential diagnostic value under the paradigm of individualized oncology, clinical decision-making must also consider the degree of uncertainty associated with each model. In this study, we evaluate the performance of a novel machine-learning (ML) algorithm, called Gaussian Process (GP) modeling, that can quantify the impact of multiple sources of uncertainty in ML model development and prediction accuracy, including variabilities in the copy number measurement, radiomics features, training sample characteristics, and training sample size. METHOD We collected 95 image-localized biopsies from 25 primary GBM patients. We coregistered stereotactic locations with preoperative multi-parametric MRI features (conventional MRI, DSC perfusion, Diffusion Tensor Imaging) to generate spatially matched pairs of MRI and copy number variants (CNV) for for each biopsy. We developed a Gaussian Process (GP) model to predict CNV for Epidermal Growth Factor Receptor (EGFR) based on MRI radiomic features in each patient. We used leave-one-patient-out cross validation to quantify prediction accuracy and model uncertainty. Spatial prediction and uncertainty (p-value) maps were overlaid to help visualize regional genetic variation of EGFR and uncertainty of the radiomic predictions. RESULT: The initial GP radiomics model for EGFR amplification (CNV > 3.5) produced a sensitivity of 0.8 and specificity of 0.8. Samples/regions associated with high uncertainty (p-value >0.05) correlated with either 1) extrapolation of radiomic features from the training set-defined feature space or 2) insufficient training samples in the feature space. CONCLUSION We present a ML-based model that quantifies spatial genetic heterogeneity in GBM, while also estimating model uncertainties that result from multi-source data variabilities. This approach lays the groundwork for prospective clinical integration of modeling-based diagnostic approaches in the paradigm of individualized medicine.

Sex differences in the incidence and outcome of human disease are broadly recognized but in most cases not adequately understood to enable sex-specific approaches to treatment. Glioblastoma (GBM), the most common malignant brain tumor, provides a case in point. Despite well-established differences in incidence, and emerging indications of differences in outcome, there are few insights that distinguish male and female GBM at the molecular level, or allow specific targeting of these biological differences. Here, using a quantitative imaging-based measure of response, we found that temozolomide chemotherapy is more effective in female compared to male GBM patients. We then applied a novel computational algorithm to linked GBM transcriptome and outcome data, and identified novel sex-specific molecular subtypes of GBM in which cell cycle and integrin signaling were identified as the critical determinants of survival for male and female patients, respectively. The clinical utility of cell cycle and integrin signaling pathway signatures was further established through correlations between gene expression and in vitro chemotherapy sensitivity in a panel of male and female patient-derived GBM cell lines. Together these results suggest that greater precision in GBM molecular subtyping can be achieved through sex-specific analyses, and that improved outcome for all patients might be accomplished via tailoring treatment to sex differences in molecular mechanisms.One Sentence SummaryMale and female glioblastoma are biologically distinct and maximal chances for cure may require sex-specific approaches to treatment.

In glioblastoma (GBM), contrast enhanced (CE)-MRI delineates bulk tumor with contrast-enhancement but poorly characterizes invasive tumor in the nonenhancing T2W abnormality. There is extensive literature in both machine-learning (ML) and mechanistic mathematical oncology seeking to accurately predict diffuse tumor invasion from multi-parametric MRI. ML offers strengths of a data-driven iterative approach, while mechanistic (proliferation-invasion, PI) modeling incorporates spatial relationships with expected drop-offs of tumor cell density from central regions of MRI enhancement. In this study, we build and cross-validate a first-of-its-kind hybrid (ML-PI) model. We collected 82 image-guided biopsies from 18 primary GBM patients throughout CE T1W and nonenhancing T2W regions. For each biopsy, we obtained neuropathologist estimates of tumor cell density and spatially matched MRI (CE T1W, T2W) from which we extracted texture features. PI maps of tumor invasion were generated from MRI-based patient-specific estimates of the net rates of invasion and proliferation. Then, the PI maps were incorporated with a ML model that uses texture features to predict cell density to minimize the prediction error on biopsy samples (ML strength) while making sure tumor-wide prediction (biopsied and unbiopsied regions) conformed with glioblastoma biology (PI strength). We optimized this hybrid ML-PI model using leave-one-out-cross-validation, and compared its performance with PI and ML alone. We used Pearson correlation (r) between cross-validated predicted tumor cell density and true tumor cell density. We focused on prediction within the nonenhancing zone (n=32) because accurate estimation of invasive tumor cell density in this zone has important clinical value for radiation and surgical planning. PI-ML showed significantly stronger correlation (r=0.76) compared to PI (r=0.44) and ML (r=0.04) models alone (p value

The current guideline to assess radiological changes in response to immunotherapy, iRANO, faces critical challenges in the determination of progressive disease during the first 6 months from the start of therapy. In patients with recurrent glioblastoma (GBM), immunotherapy induces an inflammatory response that mimics progressive disease, which can lead to premature discontinuation of potentially effective therapy. We investigated the dynamics of tumor growth within the first month of starting immunotherapy with the goal of identifying early indicators that could suggest clinical response. We performed a retrospective review of 8 patients from a Phase 1 Trial of CAR T-Cell therapy that had one MRI scan prior to treatment, during treatment and following treatment. Spherically equivalent radii were extracted from tumor volumes segmented on T1Gd and FLAIR MRIs to quantify the dynamics of tumor growth. Kaplan Meier and Cox Proportional Hazard analysis were performed to the role of tumor dynamics in predicting patient survival. Initial tumor growth velocity during CAR T-cell therapy of >50 mm/year (P=0.0067) was prognostic for overall survival from the time of treatment. An increase in radial T1Gd tumor size of >25% (P=0.046) after receiving two CAR T-Cell cycles also discriminated patient survival from the start of treatment. In addition to the above noted optimal cutoffs, initial T1Gd tumor growth velocity was significant as a continuous variable in cox proportional hazard (HR=0.976, P=0.0404). This significance was retained when controlling for age and sex (HR=0.961, P=0.0429). In this case, more aggressive imageable growth was beneficial to the patient. This increased tumor growth rate may be suggestive of a robust immunological response early in therapy that may be important to anticipate in the clinical management of GBM patients. Increased initial tumor growth velocity during treatments shows promise in identifying patients experiencing clinically relevant response that is predictive of patient survival.

BackgroundQUIT is the only primary care-based brief intervention that has previously shown efficacy for reducing risky drug use in the United States (Gelberg et al., 2015). This pilot study replicated the QUIT protocol in one of the five original QUIT clinics primarily serving Latinos.DesignSingle-blind, two-arm, randomized controlled trial of patients enrolled from March-October 2013 with 3-month follow-up.SettingPrimary care waiting room of a federally qualified health center (FQHC) in East Los Angeles.ParticipantsAdult patients with risky drug use (4-26 on the computerized WHO ASSIST): 65 patients (32 intervention, 33 control); 51 (78%) completed follow-up; mean age 30.8 years; 59% male; 94% Latino.Interventions and measuresIntervention patients received: 1) brief (typically 3-4 minutes) clinician advice to quit/reduce their risky drug use, 2) video doctor message reinforcing the clinician's advice, 3) health education booklet, and 4) up to two 20-30 minute follow-up telephone drug use reduction coaching sessions. Control patients received usual care and cancer screening information. Primary outcome was reduction in number of days of drug use in past 30days of the highest scoring drug (HSD) on the baseline ASSIST, from baseline to 3-month follow-up.ResultsControls reported unchanged HSD use between baseline and 3-month follow-up whereas Intervention patients reported reducing their use by 40% (p

PURPOSE: Despite a growing body of research on the etiology of seizures in the brain tumor population, it is not yet clear why approximately one third of this population presents with seizure. In this retrospective study we seek to assess the impact of tumor location on the risk of seizure among male and female glioma patients, respectively. METHODS: From our multi-institutional database, we selected adult patients with contrast-enhancing gliomas (any grade) and known seizure-presentation status at initial diagnosis. Tumors were segmented on pretreatment T2-FLAIR, T2, and post-gadolinium T1 (T1Gd) MR sequences. We warped cortical and subcortical probabilistic atlases to patient images and estimated tumor burden (%) on structures within each lobe and the deep brain. We calculated risk of seizure for various levels of tumor burden and used the result to create sex-specific risk-maps RESULTS: Our cohort included 128 patients (47 females, 81 males) among whom tumors presented with seizure in 34% of females (n=18) and 58% of males (n=47). Patients with tumors in deep brain structures were identified as at risk for seizure in both males and females. In female patients T2 hyperintensity on 44% of right frontal lobe or 28% of left parietal lobe seem to have an estimated 50%+ risk of seizure. Data indicate that in male patients with left-sided tumors, T2 hyperintensity on 30% or more of frontal, temporal, or parietal lobes results in 50%+ risk of seizure. Male patients with right side tumors, showed a similar level of risk for T2 presence in lateral ventricles and parietal lobe. CONCLUSION: This study reveals that the risk of seizure is specific to the location of glioma, the percentage of tumor burden, and patient sex. These differences could be considered in patient management decisions to more selectively prescribe anticonvulsants and optimize patient care.

INTRODUCTION: Evaluating treatment response in glioblastoma is a difficult task for clinicians, particularly in recurrent disease. Current response metrics focus on changes in imageable tumor burden which can be ambiguously affected by anti-angiogenic therapy. In previous work, we reported the ability of the Days Gained (DG) response metric to discriminate patients receiving bevacizumab into long and short-term survival groups. In this work, we investigate how the DG metric performs for patients who received bevacizumab with a concurrent cytotoxic agent (e.g. CCNU) compared to patients who received bevacizumab alone. METHODS: We identified a set of 38 patients with recurrent glioblastoma who received bevacizumab therapy (21 patients) or bevacizumab with concurrent CCNU (17 patients). Each case had tumor volumes segmented on two dates prior and one day post therapy on T1GD and FLAIR MR imaging. DG scores were calculated using tumor growth characteristics and were used to evaluate discrimination of patient survival and time to progression (TTP) using Kaplan-Meier curves and logistic regression. RESULTS: An optimal threshold of 162 DG (p= 0.0393, log rank) on T1GD imaging was discriminative for TTP in bevacizumab cases. Combination therapy with CCNU showed a similar trend for the correlation between DG and TTP outcomes as for bevacizumab alone. Patients receiving bevacizumab alone had significant DG thresholds for survival following treatment (T1GD: 162 DG, p=0.01857; FLAIR: 216 DG, p=0.0387). Although predictive of TTP, combination therapy was not significantly associated with increased overall survival from time of treatment. DISCUSSION: Days gained was able to discriminate survival and TTP for patients receiving bevacizumab therapy alone. Discriminative power for combination therapy with CCNU demonstrated the same trend for TTP outcomes. CONCLUSION: Growing evidence supports DG as a clinically meaningful metric of treatment response even in the context of anti-angiogenic therapies that are known to ambiguously modulate imaging features on MRI.

Determining effective therapies for patients with recurrent glioblastoma remains difficult especially in small early phase clinical trials. Currently utilized estimators of treatment response rely on two-dimensional measurements of tumor progression (e.g. RANO, RECIST, MacDonald) that have often fallen short in reliably connecting patients who respond with positive survival outcomes. This has prompted consideration of advanced imaging and volumetric measurements by the RANO working group and others. Our proposed approach, Days Gained (DG), uses volumetric measurements to calculate a personalized response metric based on change in tumor growth characteristics pre/post-treatment. DG has shown promise for connecting imageable treatment response with clinically meaningful (statistically significant) survival benefit when applied in upfront and recurrent therapy settings, even in the context of coincident bevacizumab treatment. In this study, we seek to consider a new cohort of patients receiving bevacizumab in combination with other novel therapies to distinguish any subpopulation of patients with meaningful response. In this work, we identified a subset of 16 patients from bevacizumab clinical trials with data available for DG analysis requiring serial imaging (T1Gd and FLAIR) prior to and after the first cycle of therapy. Each trial evaluated bevacizumab alone versus bevacizumab plus either dasatinib or TRC105 in recurrent disease. Discrimination of time to progression (TTP) was evaluated at DG thresholds identified from our prior analysis of patients treated with bevacizumab. Using previously identified DG thresholds for clinically significant treatment benefit, in our cohort, DG measured by serial FLAIR imaging was significant in predicting longer TTP (120 DG, p=0.014), but T1Gd DG scores did not attain significance (p=0.630). Given the limited insights in connecting response metrics to accurately predicted outcomes, these data further support the use of DG as a burgeoning marker of treatment response (or lack of response) that may be useful for more routine integration into clinical trials.

BACKGROUND An important challenge in radiomics research is reproducibility. Images are collected on different image scanners and protocols, which introduces significant variability even for the same type of image across institutions. In the present proof-of-concept study, we address the reproducibility issue by using domain adaptation – an algorithm that transforms the radiomic features of each new patient to align with the distribution of features formed by the patient samples in a training set. METHOD Our dataset included 18 patients in training with a total of 82 biopsy sample. The pathological tumor cell density was available for each sample. Radiomic (statistical + texture) features were extracted from the region of six image contrasts locally matched with each biopsy sample. A Gaussian Process (GP) classifier was built to predict tumor cell density using radiomic features. Another 6 patients were used to test the training model. These patients had a total of 31 biopsy samples. The images of each test patient were purposely normalized using a different approach, i.e., using the CSF instead of the whole brain as the reference. This was to mimic the practical scenario of image source discrepancy between different patients. Domain adaptation was applied to each test patient. RESULTS Among the 18 training patients, the leave-one-patient-out cross validation accuracy is 0.81 AUC, 0.78 sensitivity, and 0.83 specificity. When the trained model was applied to the 6 test patients (purposely normalized using a different approach than that of the training data), the accuracy dramatically reduced to 0.39 AUC, 0.08 sensitivity, and 0.61 specificity. After using domain adaption, the accuracy improved to 0.68 AUC, 0.62 sensitivity, and 0.72 specificity. CONCLUSION We provide candidate enabling tools to address reproducibility in radiomics models by using domain adaption algorithms to account for discrepancy of the images between different patients.

High-grade gliomas (HGGs) nearly always recur after standard initial treatment, and the resulting mixture of recurrent tumor and treatment-induced reactive changes presents major diagnostic challenges. Anatomical imaging, such as MRI, cannot adequately distinguish progressive disease from treatment effect (pseudo-progression). Furthermore, there is marked intra-tumoral heterogeneity, such that some areas of a tumor may demonstrate necrotic treatment effect and others frank recurrence. Due to this difficulty reliably differentiating between these two clinical findings, analytic methods using multiple modalities are necessary to further our understanding of this disease process. To this end, we sought to correlate radiographic, histopathologic and molecular features of surgically sampled post-treatment suspected recurrence to identify markers distinguishing tumor growth from treatment effect. We performed Stimulated Raman Histology (SRH) imaging and highly multiplexed RNA-sequencing (PLATE-seq) on 84 MRI-localized biopsies from 39 patients with clinically suspected recurrent HGG. The SRH images were classified as recurrent tumor or gliotic/reactive tissue using a convolutional neural network trained on an independent cohort including a large set of recurrent HGG, and an automated cell-counting algorithm was used to quantify cellularity from the SRH image of each sample. Differential gene expression analysis of the PLATE-seq data was used to identify gene sets that distinguish recurrent tumor from treatment effect, and single sample gene set variation analysis (GSVA) was used to further assess the molecular and cellular composition of each MRI-localized sample. The histopathologic and molecular features of each sample were also correlated with the MRI features of the corresponding biopsy sites, and this data is currently being used to train machine learning models that predict the distribution of recurrent tumor and treatment-induced reactive changes within a patient’s radiographic lesion. These predictive radiomic models will help to guide neurosurgical sampling, and improve our ability to monitor glioma progression and response to therapy.

Recurrent glioblastoma is challenging to distinguish from so called “treatment effect” on routine clinical imaging. Further, within tumor heterogeneity reveals that some regions can be histologically dominated by tumor progression whilst others can be dominated by secondary effects of treatment response. Apparent tumor progression on MRI can be very difficult to manage clinically as it is unclear the degree to which the imaging changes are actually tumor progression vs response to treatment (including inflammatory response and necrosis). In this analysis, we study a unique cohort of patients for whom image localized-biopsies reveal heterogeneity in response vs progression. Our dataset included 70 biopsy samples from 32 patients with GBM each histolopatholgically characterized for tumor abundance vs immune infiltrate. Six multiparametric MRI contrasts were available, including T1, T1gd, T2, FLAIR, SWI, and ADC. Images were co-registered. Radiomic (statistical + texture) features were extracted from the region of six image contrasts locally matched with each biopsy sample. Machine learning models were built to predict each biomarker using radiomic features. Leave-one-out cross validation was used to evaluate the prediction accuracy. Radiomic features were found to be informative to the prediction of biomarkers. ANOVA tests show significant improvement of using radiomic features compared with the null model. The prediction accuracy was higher when considering the biomarkers on a binary scale using the median as the cutoff than on a numerical scale. Spatially-informed radiomics models for tumor progression vs treatment effect are possible and can play an instrumental role in navigating confounding imaging changed common during treatment progression.

BACKGROUND: MRI-based modeling can help characterize the intratumoral genetic heterogeneity of Glioblastoma (GBM). Yet, published models to date have neglected the potential impact of sex-differences on the accuracy of MRI-genetic correlations. Specifically, there is growing awareness that female GBM patients can display different genetic/molecular aberrations and phenotypic expression compared to male counterparts. In this exploratory study, we compare MRI signal and key GBM driver alterations across a cohort of male and female GBM patients, using image-guided biopsies and spatially matched multi-parametric MRI. METHODS: We collected 61 image-guided biopsies from 18 primary GBM patients (9/9 male/female). For each biopsy, we analyzed DNA copy number variants (CNV) for 6 core GBM driver genes reported by TCGA: amplifications (++) for EGFR and PDGFRA and deletions (--) for PTEN, CDKN2A, RB1, TP53. We compared regional CNV status with spatially matched MRI texture measurements from co-registered biopsy locations. Advanced MRI features included relative cerebral blood volume (rCBV) on DSC-perfusion, mean diffusivity (MD) and fractional anisotropy (FA) on diffusion tensor imaging. We identified univariate correlations for combined and sex-specific (male, female) subgroups. We also built multivariate predictive decision-tree models for each GBM driver gene and used leave-one-out-cross-validation (LOOCV) to determine area-under-curve (AUC) on ROC analysis to compare accuracies across combined and sex-specific models. RESULTS: We identified multiple univariate correlations between regional CNV status and spatially matched MRI texture features that were specific to either male or female GBM tumors. For instance, EGFR++ specifically correlated with T2W image textures in male biopsies but rCBV textures in female biopsies. In general, sex-specific analyses on decision-tree modeling improved predictive accuracies (AUC) compared to combined (male+female) modeling, particularly for EGFR++ (p

INTRODUCTION: Glioblastoma (GBM) is the most common malignant primary brain tumor in adults with a median survival of 14–16 months. Patient sex plays an important role in GBM as there is a difference in incidence rates and outcome between males and females, which may be attributable to differences in genetic makeup and physiology. OBJECTIVE: Investigate the impact of tumor location and sex differences on survival outcomes based on tumor location, laterality, age, handedness, and extent of resection. METHODS: Patients (129 males and 87 females) who received standard-of-care were included. Analyses were performed using Cox proportional hazard modeling and Kaplan-Meier analysis (log-rank test) to determine which variables impacted patient survival. RESULTS: Overall survival was significantly longer in females in comparison to males (197 days, p = 0.0391). Investigating specific tumor locations, females with a tumor in the left frontal lobe (n = 12) showed a survival advantage compared to females with a right frontal (n = 15) GBM (2853 days, p = 0.0160). Significant differences in median OS were also associated with age. Female patients below the age of fifty showed significantly longer survival (2602 days, n = 84, p

Despite the growing ubiquity of data in the medical domain, it remains difficult to apply results from experimental and observational studies to additional populations suffering from the same disease. Many methods are employed for testing internal validity; yet limited effort is made in testing generalizability, or external validity. The development of disease models often suffers from this lack of validity testing and trained models frequently have worse performance on different populations, rendering them ineffective. In this work, we discuss the use of transportability theory, a causal graphical model examination, as a mechanism for determining what elements of a data resource can be shared or moved between a source and target population. A simplified Bayesian model of glioblastoma multiforme serves as the example for discussion and preliminary analysis. Examination over data collection hospitals from the TCGA dataset demonstrated improvement of prediction in a transported model over a baseline model.

To assess the effect of a multi-component primary care delivered brief intervention for reducing risky psychoactive drug use (RDU) among patients identified by screening.Multicenter single-blind two-arm randomized controlled trial of patients enrolled from February 2011 to November 2012 with 3-month follow-up. Randomization and allocation to trial group were computer-generated.Primary care waiting rooms of five federally qualified health centers in Los Angeles County (LAC), USA.A total of 334 adult primary care patients (171 intervention; 163 control) with RDU scores (4-26) on the World Health Organization (WHO) Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) self-administered on tablet computers. 261 (78%) completed follow-up. Mean age was 41.7 years; 62.9% were male; 37.7% were Caucasian.Intervention patients received brief (typically 3-4 minutes) clinician advice to quit/reduce their drug use reinforced by a video doctor message, health education booklet and up to two 20-30-minute follow-up telephone drug use coaching sessions. Controls received usual care and cancer screening information. Primary outcome was patient self-reported use of highest scoring drug (HSD) at follow-up.Intervention and control patients reported equivalent baseline HSD use at 3-month follow-up. After adjustment for covariates, in the complete sample linear regression model, intervention patients used their HSD on 3.5 fewer days in the previous month relative to controls (P0.001), and in the completed sample model, intervention patients used their HSD 2.2 fewer days than controls (P 0.005). No compensatory increases in use of other measured substances were found.A primary-care based, clinician-delivered brief intervention with follow-up coaching calls may decrease risky psychoactive drug use.

The growing amount of electronic data collected from patient care and clinical trials is motivating the creation of national repositories where multiple institutions share data about their patient cohorts. Such efforts aim to provide sufficient sample sizes for data mining and predictive modeling, ultimately improving treatment recommendations and patient outcome prediction. While these repositories offer the potential to improve our understanding of a disease, potential issues need to be addressed to ensure that multi-site data and resultant predictive models are useful to non-contributing institutions. In this paper we examine the challenges of utilizing National Cancer Institute datasets for modeling glioblastoma multiforme. We created several types of prognostic models and compared their results against models generated using data solely from our institution. While overall model performance between the data sources was similar, different variables were selected during model generation, suggesting that mapping data resources between models is not a straightforward issue.

The accurate and expeditious collection of survey data by coordinators in the field is critical in the support of research studies. Early methods that used paper documentation have slowly evolved into electronic capture systems. Indeed, tools such as REDCap and others illustrate this transition. However, many current systems are tailored web-browsers running on desktop/laptop computers, requiring keyboard and mouse input. We present a system that utilizes a touch screen interface running on a tablet PC with consideration for portability, limited screen space, wireless connectivity, and potentially inexperienced and low literacy users. The system was developed using C#, ASP.net, and SQL Server by multiple programmers over the course of a year. The system was developed in coordination with UCLA Family Medicine and is currently deployed for the collection of data in a group of Los Angeles area clinics of community health centers for a study on drug addiction and intervention.

Predicting patient outcome is an important task in medical decision making, as clinician expectations of outcome drive testing and treatment decisions. Accurate models can assist clinicians by capitalizing on information from a broad spectrum of features to predict outcome. In an article in this journal, ‘A study in transfer learning: leveraging data from multiple hospitals to enhance hospital-specific predictions,’ Wiens, Guttag, and Horvitz1 explore the use of transfer learning for improving a predictive model of Clostridium difficile infection (CDI). Their discussion focuses on the need to aggregate data for studying rare diseases but notes the failure of global models to predict accurately for specific institutions. Transfer learning attempts to rectify the generalizability problem by applying evidence from multiple sources on a related target task. Their work demonstrates how transfer learning can be utilized to create a ‘source+target’ model matching or outperforming models trained with source or target data alone. A number of important considerations when pooling data are raised by the authors. Here, we note the need for further discussion by revisiting two points raised in their paper affecting transfer: (1) feature similarity and (2) feature selection. We briefly discuss limitations of transfer learning tied to a lack of causal knowledge and posit that causal information can complement transfer learning to improve model generalization. Weins et al combined datasets by comparing the overlap of features in data collected across hospitals. However, it is important to note that overlapping features do not guarantee feature similarity. This issue was explored in the section, ‘Not all transfer is created equal’; hospital …