Standardized evaluation of the extent of resection in glioblastoma with automated early post-operative segmentation.

Standard treatment of patients with glioblastoma includes surgical resection of the tumor. The extent of resection (EOR) achieved during surgery significantly impacts prognosis and is used to stratify patients in clinical trials. In this study, we developed a U-Net-based deep-learning model to segment contrast-enhancing tumor on post-operative MRI exams taken within 72 h of resection surgery and used these segmentations to classify the EOR as either maximal or submaximal. The model was trained on 122 multiparametric MRI scans from our institution and achieved a mean Dice score of 0.52 ± 0.03 on an external dataset (n = 248), a performance -on par with the interrater agreement between expert annotators as reported in literature. We obtained an EOR classification precision/recall of 0.72/0.78 on the internal test dataset (n = 462) and 0.90/0.87 on the external dataset. Furthermore, Kaplan-Meier curves were used to compare the overall survival between patients with maximal and submaximal resection in the internal test dataset, as determined by either clinicians or the model. There was no significant difference between the survival predictions using the model's and clinical EOR classification. We find that the proposed segmentation model is capable of reliably classifying the EOR of glioblastoma tumors on early post-operative MRI scans. Moreover, we show that stratification of patients based on the model's predictions offers at least the same prognostic value as when done by clinicians.

No association between metformin initiation and incident dementia in older adults newly diagnosed with diabetes.

BACKGROUND: Metformin has been suggested to reduce dementia risk; however, most epidemiologic studies have been limited by immortal time bias or confounding due to disease severity. OBJECTIVES: To investigate the association of metformin initiation with incident dementia using strategies that mitigate these important sources of bias. METHODS: Residents of Ontario, Canada ≥66 years newly diagnosed with diabetes from January 1, 2008 to December 31, 2017 entered this retrospective population-based cohort. To consider the indication for metformin monotherapy initiation, people with hemoglobin A1c of 6.5%-8.0% and estimated glomerular filtration rate ≥45 mL/min/1.73 m2 were selected. Using the landmark method to address immortal time bias, exposure was grouped into "metformin monotherapy initiation within 180 days after new diabetes diagnosis" or "no glucose-lowering medications within 180 days." To address disease latency, 1-year lag time was applied to the end of the 180-day landmark period. Incident dementia was defined using a validated algorithm for Alzheimer's disease and related dementias. Adjusted hazard ratios (aHR) and confidence intervals (CIs) were estimated from propensity-score weighted Cox proportional hazard models. RESULTS: Over mean follow-up of 6.77 years from cohort entry, metformin initiation within 180 days after new diabetes diagnosis (N = 12,331; 978 events; 65,762 person-years) showed no association with dementia risk (aHR [95% CI] = 1.05 [0.96-1.15]), compared to delayed or no glucose-lowering medication initiation (N = 22,369; 1768 events; 117,415 person-years). CONCLUSION: Early metformin initiation was not associated with incident dementia in older adults newly diagnosed with diabetes. The utility of metformin to prevent dementia was not supported.

Vascular Endothelial Growth Factor Polymorphism rs699947 Is Associated with Neurostructural Phenotypes in Youth with Bipolar Disorder.

Background: Vascular endothelial growth factor (VEGF) may be relevant to bipolar disorder (BD) and brain structure. We evaluated VEGF rs699947 single-nucleotide polymorphism in relation to structural neuroimaging phenotypes in youth BD. Methods: We collected 3 T anatomical magnetic resonance images from 154 youth (79 BD and 75 healthy control [HC]) genotyped for VEGF rs699947. The participants were age (BD = 17.28 ± 1.40 and HC = 17.01 ± 1.83, t = -1.02, p = 0.31) and sex (BD = 63.3% females and HC = 52.0% females, χ2 = 2.01, p = 0.16) matched. Cortical thickness, surface area (SA), and volume were examined by region-of-interest (ROI) and vertex-wise analyses using general linear models (GLMs). ROI investigations selected for the prefrontal cortex (PFC), amygdala, and hippocampus. Vertex-wise analyses controlled for age, sex, and intracranial volume. Results: ROI results found lower PFC SA (p = 0.003, ηp2 = 0.06) and volume (p = 0.04, ηp2 = 0.03) in BD and a main effect of rs699947 on hippocampal volume (p = 0.03, ηp2 = 0.05). The latter two findings did not survive multiple comparisons. Vertex-wise analyses found rs699947 main effects on left postcentral gyrus volume (p < 0.001), right rostral anterior cingulate SA (p = 0.004), and right superior temporal gyrus thickness (p = 0.004). There were significant diagnosis-by-genotype interactions in the left superior temporal, left caudal middle frontal, left superior frontal, right fusiform, and right lingual gyri, and the left insular cortex. Posthoc analyses revealed the AA allele was associated with larger brain structures among HC, but smaller brain structures in BD for most clusters. Conclusions: Overall, we found preliminary evidence of divergent associations between BD and HC youth in terms of neurostructural correlates of VEGF rs699947 encompassing highly relevant frontotemporal regions.

Automated generation of cerebral blood flow and arterial transit time maps from multiple delay arterial spin-labeled MRI.

PURPOSE: Develop and evaluate a deep learning approach to estimate cerebral blood flow (CBF) and arterial transit time (ATT) from multiple post-labeling delay (PLD) ASL MRI. METHODS: ASL MRI were acquired with 6 PLDs on a 1.5T or 3T GE system in adults with and without cognitive impairment (N = 99). Voxel-level CBF and ATT maps were quantified by training models with distinct convolutional neural network architectures: (1) convolutional neural network (CNN) and (2) U-Net. Models were trained and compared via 5-fold cross validation. Performance was evaluated using mean absolute error (MAE). Model outputs were trained on and compared against a reference ASL model fitting after data cleaning. Minimally processed ASL data served as another benchmark. Model output uncertainty was estimated using Monte Carlo dropout. The better-performing neural network was subsequently re-trained on inputs with missing PLDs to investigate generalizability to different PLD schedules. RESULTS: Relative to the CNN, the U-Net yielded lower MAE on training data. On test data, the U-Net MAE was 8.4 ± 1.4 mL/100 g/min for CBF and 0.22 ± 0.09 s for ATT. A significant association was observed between MAE and Monte Carlo dropout-based uncertainty estimates. Neural network performance remained stable despite systematically reducing the number of input images (i.e., up to 3 missing PLD images). Mean processing time was 10.77 s for the U-Net neural network compared to 10 min 41 s for the reference pipeline. CONCLUSION: It is feasible to generate CBF and ATT maps from 1.5T and 3T multi-PLD ASL MRI with a fast deep learning image-generation approach.

Prevalence of white matter hyperintensities is not elevated in a large sample of adolescents and young adults with bipolar disorder

Objective: The increased prevalence rate of white matter hyperintensities is one of the most consistently reported brain abnormalities in adults with bipolar disorder. However, findings in children and adolescents with bipolar disorder are less consistent. Prior studies have been constrained by small sample sizes and/or poor age- and sex-matching of healthy controls. We examined this topic in the largest sample of adolescents with bipolar disorder to date. Methods: T2-weighted 3-Tesla magnetic resonance imaging data were acquired for 83 adolescents with bipolar disorder diagnosed via the Kiddie Schedule for Affective Disorders and the Schizophrenia, Present and Lifetime version semi-structured interview and 64 age- and sex-matched healthy controls. All acquired scans were examined by neuroradiologists and the presence or absence of white matter hyperintensities was determined for each participant. Results: The prevalence of white matter hyperintensities did not differ between adolescents with bipolar disorder (13.3%) and controls (21.9%; χ2 = 1.90; p = 0.168). Conclusion: In contrast to the study hypothesis, the prevalence of white matter hyperintensities was not higher in adolescents with bipolar disorder than controls. The large sample size and good matching for age and sex bolster the reliability of this negative finding. Future studies are warranted to evaluate the prevalence, incidence, and predictors of white matter hyperintensities in early-onset bipolar disorder prospectively.

Assessment of cognitive and neural recovery in survivors of pediatric brain tumors in a pilot clinical trial using metformin.

We asked whether pharmacological stimulation of endogenous neural precursor cells (NPCs) may promote cognitive recovery and brain repair, focusing on the drug metformin, in parallel rodent and human studies of radiation injury. In the rodent cranial radiation model, we found that metformin enhanced the recovery of NPCs in the dentate gyrus, with sex-dependent effects on neurogenesis and cognition. A pilot double-blind, placebo-controlled crossover trial was conducted (ClinicalTrials.gov, NCT02040376) in survivors of pediatric brain tumors who had been treated with cranial radiation. Safety, feasibility, cognitive tests and MRI measures of white matter and the hippocampus were evaluated as endpoints. Twenty-four participants consented and were randomly assigned to complete 12-week cycles of metformin (A) and placebo (B) in either an AB or BA sequence with a 10-week washout period at crossover. Blood draws were conducted to monitor safety. Feasibility was assessed as recruitment rate, medication adherence and procedural adherence. Linear mixed modeling was used to examine cognitive and MRI outcomes as a function of cycle, sequence and treatment. We found no clinically relevant safety concerns and no serious adverse events associated with metformin. Sequence effects were observed for all cognitive outcomes in our linear mixed models. For the subset of participants with complete data in cycle 1, metformin was associated with better performance than placebo on tests of declarative and working memory. We present evidence that a clinical trial examining the effects of metformin on cognition and brain structure is feasible in long-term survivors of pediatric brain tumors and that metformin is safe to use and tolerable in this population. This pilot trial was not intended to test the efficacy of metformin for cognitive recovery and brain growth, but the preliminary results are encouraging and warrant further investigation in a large multicenter phase 3 trial.

Amyloid-beta burden predicts prospective decline in body mass index in clinically normal adults.

In the present study, we tested the hypothesis that higher amyloid-beta (Aß) burden at baseline is associated with greater longitudinal decline in body mass index (BMI) in clinically normal adults. Participants from the Harvard Aging Brain Study (n = 312) and the Alzheimer's Disease Neuroimaging Initiative (n = 336) underwent Aß positron emission tomography at baseline. BMI was assessed longitudinally over a median of >4 years. Linear mixed models showed that higher baseline Aß burden was significantly associated with greater decline in BMI in both the Harvard Aging Brain Study (t = -1.93; p = 0.05) and Alzheimer's Disease Neuroimaging Initiative cohorts (t = -2.54; p = 0.01), after adjusting for covariates, including cognitive performance and depressive symptoms. In addition, the association of Aß burden with longitudinal decline in BMI persisted in both cohorts after excluding participants with diabetes/endocrine disturbances and participants classified as underweight or obese (BMI <18.5 or >30). These findings suggest that decline in BMI in clinically normal adults may be an early manifestation related to cerebral amyloidosis that precedes objective cognitive impairment. Therefore, unintentional BMI decline in otherwise healthy individuals might alert clinicians to increased risk of Alzheimer's disease.

Effects of cannabis on cognition in patients with MS: a psychometric and MRI study.

OBJECTIVE: To determine functional and structural neuroimaging correlates of cognitive dysfunction associated with cannabis use in multiple sclerosis (MS). METHODS: In a cross-sectional study, 20 subjects with MS who smoked cannabis and 19 noncannabis users with MS, matched on demographic and neurologic variables, underwent fMRI while completing a test of working memory, the N-Back. Resting-state fMRI and structural MRI data (lesion and normal-appearing brain tissue volumes, diffusion tensor imaging metrics) were also collected. Neuropsychological data pertaining to verbal (Selective Reminding Test Revised) and visual (10/36 Spatial Recall Test) memory, information processing speed (Paced Auditory Serial Addition Test [2- and 3-second versions] and Symbol Digit Modalities Test), and attention (Word List Generation) were obtained. RESULTS: The cannabis group performed more poorly on the more demanding of the Paced Auditory Serial Addition Test tasks (i.e., 2-second version) (p < 0.02) and the 10/36 Spatial Recall Test (p < 0.03). Cannabis users had more diffuse cerebral activation across all N-Back trials and made more errors on the 2-Back task (p < 0.006), during which they displayed increased activation relative to nonusers in parietal (p < 0.007) and anterior cingulate (p < 0.001) regions implicated in working memory. No group differences in resting-state networks or structural MRI variables were found. CONCLUSIONS: Patients with MS who smoke cannabis are more cognitively impaired than nonusers. Cannabis further compromises cerebral compensatory mechanisms, already faulty in MS. These imaging data boost the construct validity of the neuropsychological findings and act as a cautionary note to cannabis users and prescribers.

Evaluating quantitative approaches to dynamic susceptibility contrast MRI among carotid endarterectomy patients.

PURPOSE: To evaluate two dynamic susceptibility contrast (DSC) quantification methods in symptomatic carotid artery disease patients undergoing carotid endarterectomy (CEA) surgery by comparing methods directly and assessing the reliability of each method in the hemisphere contralateral to surgery. MATERIALS AND METHODS: Absolute cerebral blood flow (CBF) and volume (CBV) was calculated in putamen and sensorimotor gray matter of 17 patients using two methods: 1) The Bookend method that scales relative DSC images to CBV values calculated from the ratio of pre- and postcontrast T1-weighted images, and 2) the Tail-scaling method that uses the ratio of area under the tails of the venous and arterial concentration time-courses to scale the DSC images. RESULTS: There was a positive correlation between the methods with significant correlation post-CEA (P < 0.035). Intersession correlation was greater when using the Tail-scaling method contralateral to surgery (P < 0.004). CONCLUSION: We have demonstrated correlation between methods that is significant after surgery and have found that the Tail-scaling method produces better test-retest reliability than our implementation of the Bookend method. Results from this study suggest that DSC has the potential to measure hemodynamic changes after endarterectomy and future work is required to establish clinical value.