Citations

Background and Purpose— We evaluated deep learning algorithms’ segmentation of acute ischemic lesions on heterogeneous multi-center clinical diffusion-weighted magnetic resonance imaging (MRI) data sets and explored the potential role of this tool for phenotyping acute ischemic stroke. Methods— Ischemic stroke data sets from the MRI-GENIE (MRI-Genetics Interface Exploration) repository consisting of 12 international genetic research centers were retrospectively analyzed using an automated deep learning segmentation algorithm consisting of an ensemble of 3-dimensional convolutional neural networks. Three ensembles were trained using data from the following: (1) 267 patients from an independent single-center cohort, (2) 267 patients from MRI-GENIE, and (3) mixture of (1) and (2). The algorithms’ performances were compared against manual outlines from a separate 383 patient subset from MRI-GENIE. Univariable and multivariable logistic regression with respect to demographics, stroke subtypes, and vascular risk factors were performed to identify phenotypes associated with large acute diffusion-weighted MRI volumes and greater stroke severity in 2770 MRI-GENIE patients. Stroke topography was investigated. Results— The ensemble consisting of a mixture of MRI-GENIE and single-center convolutional neural networks performed best. Subset analysis comparing automated and manual lesion volumes in 383 patients found excellent correlation (ρ=0.92; P <0.0001). Median (interquartile range) diffusion-weighted MRI lesion volumes from 2770 patients were 3.7 cm 3 (0.9–16.6 cm 3 ). Patients with small artery occlusion stroke subtype had smaller lesion volumes ( P <0.0001) and different topography compared with other stroke subtypes. Conclusions— Automated accurate clinical diffusion-weighted MRI lesion segmentation using deep learning algorithms trained with multi-center and diverse data is feasible. Both lesion volume and topography can provide insight into stroke subtypes with sufficient sample size from big heterogeneous multi-center clinical imaging phenotype data sets.

AbstractObjectiveOnset of multiple sclerosis (MS) occurs in childhood for approximately 5% of cases (pediatricMS, or ped‐MS). Epigenetic influences are strongly implicated inMSpathogenesis in adults, including the contribution from microRNAs (miRNAs), small noncodingRNAs that affect gene expression by binding target genemRNAs. Few studies have specifically examined miRNAs in ped‐MS, but individuals developingMSat an early age may carry a relatively high burden of genetic risk factors, and miRNAdysregulation may therefore play a larger role in the development of ped‐MSthan in adult‐onsetMS. This study aimed to look for evidence of miRNAinvolvement in ped‐MSpathogenesis.MethodsGWASresults from 486 ped‐MScases and 1362 controls from the U.S. PediatricMSNetwork and Kaiser Permanente Northern California membership were investigated for miRNA‐specific signals. First, enrichment of miRNA‐target gene network signals was evaluated usingMIGWASsoftware. Second,SNPs in miRNAgenes and in target gene binding sites (miR−SNPs) were tested for association with ped‐MS, and pathway analysis was performed on associated target genes.ResultsMIGWASanalysis showed that miRNA‐target gene signals were enriched inGWAS(P = 0.038) and identified 39 candidate biomarker miRNA‐target gene pairs, including immune and neuronal signaling genes. The miR‐SNPanalysis implicated dysregulation of miRNAbinding to target genes in five pathways, mainly involved in immune signaling.InterpretationEvidence fromGWASsuggests that miRNAs play a role in ped‐MSpathogenesis by affecting immune signaling and other pathways. Candidate biomarker miRNA‐target gene pairs should be further studied for diagnostic, prognostic, and/or therapeutic utility.

Objective: The aim of this study was to determine whether a vitamin D genetic risk score (vitDGRS) is associated with 25-hydroxyvitamin D (25(OH)D) level and multiple sclerosis (MS) relapses in children. Methods: DNA samples were typed for single nucleotide polymorphisms (SNPs) from four genes previously identified to be associated with 25(OH)D levels. SNPs with strong associations with 25(OH)D after multiple comparison correction were used to create a genetic risk score (vitDGRS). Cox regression models tested associations of vitDGRS with relapse hazard. Results: Two independent SNPs within or near GC and NADSYN1/DHCR7 genes were strongly associated with 25(OH)D levels in the discovery cohort ( n = 182) after Bonferroni correction. The vitDGRS of these SNPs explained 4.5% of the variance of 25(OH)D level after adjustment for genetic ancestry. Having the highest versus lowest vitDGRS was associated with 11 ng/mL lower 25(OH)D level (95% confidence interval (CI) = −17.5, −4.5, p = 0.001) in the discovery cohort. Adjusting for ancestry, sex, disease-modifying therapy (DMT), and HLA-DRB1*15 carrier status, the highest versus lowest vitDGRS was associated with 2.6-fold (95% CI = 1.37, 5.03, p = 0.004) and 2.0-fold (95% CI = 0.75, 5.20, p = 0.16) higher relapse hazard in the discovery and replication cohorts, respectively. Conclusion: The vitDGRS identifies children at greater risk of relapse. These findings support a causal role for vitamin D in MS course.

Background Steps towards the development of diagnostic criteria are needed for children with the radiologically isolated syndrome to identify children at risk of clinical demyelination. Objectives To evaluate the 2005 and 2016 MAGNIMS magnetic resonance imaging criteria for dissemination in space for multiple sclerosis, both alone and with oligoclonal bands in cerebrospinal fluid added, as predictors of a first clinical event consistent with central nervous system demyelination in children with radiologically isolated syndrome. Methods We analysed an international historical cohort of 61 children with radiologically isolated syndrome (≤18 years), defined using the 2010 magnetic resonance imaging dissemination in space criteria (Ped-RIS) who were followed longitudinally (mean 4.2 ± 4.7 years). All index scans also met the 2017 magnetic resonance imaging dissemination in space criteria. Results Diagnostic indices (95% confidence intervals) for the 2005 dissemination in space criteria, with and without oligoclonal bands, were: sensitivity 66.7% (38.4–88.2%) versus 72.7% (49.8–89.3%); specificity 83.3% (58.6–96.4%) versus 53.9% (37.2–69.9%). For the 2016 MAGNIMS dissemination in space criteria diagnostic indices were: sensitivity 76.5% (50.1–93.2%) versus 100% (84.6–100%); specificity 72.7% (49.8–89.3%) versus 25.6% (13.0–42.1%). Conclusions Oligoclonal bands increased the specificity of magnetic resonance imaging criteria in children with Ped-RIS. Clinicians should consider testing cerebrospinal fluid to improve diagnostic certainty. There is rationale to include cerebrospinal fluid analysis for biomarkers including oligoclonal bands in planned prospective studies to develop optimal diagnostic criteria for radiologically isolated syndrome in children.

Introduction: Risk of lobar and non-lobar intracerebral hemorrhage (ICH) varies among blacks, whites and Hispanics. We sought to determine whether these differences could be due to variability in the effects of Apolipoprotein E (APOE) epsilon (ε) alleles, the most potent genetic risk factor for ICH. Methods: Primary ICH cases and controls were collected from US and European sites contributing to the International Stroke Genetic Consortium (ISGC). We meta-analyzed the effects of APOE allele status on ICH risk applying a two-stage clustering approach based on race/ethnicity and the contributing study. Models were adjusted for age, sex, history of hypertension, hypercholesterolemia, warfarin, statin and antiplatelet use, smoking and alcohol use. A propensity score analysis was used to model the influence of APOE against the burden of hypertension across races/ethnicities. Results: 13,124 subjects (54.5% male, median age 66 years) were included. In whites, APOE ε2 (odds ratio (OR)=1.85, 95% confidence interval (CI)=1.27-2.69, p<0.001) and APOE ε4 (OR=1.94, 95% CI=1.58-2.38, p<0.001) were independently associated with lobar ICH risk, however within self-identified Hispanics and blacks, no associations were found (Figure). After propensity score-matching for hypertension burden, APOE ε4 was associated with lobar ICH risk among whites (OR=1.12, 95% CI=1.08-1.17) and Hispanics (OR=1.07, 95% CI=1.01-1.15, p=0.01), but not blacks (OR=1.02 95% CI=0.98-1.07, p=0.251). APOE ε2 and ε4 did not show an effect on non-lobar ICH risk in any race/ethnicity. Conclusion: APOE ε4 and ε2 alleles affect lobar ICH risk variably by race and ethnicity. Associations are confirmed in whites but can be shown in Hispanics only when the excess burden of hypertension is propensity score-matched. Further studies are needed to explore interactions between APOE alleles and environmental exposures that vary by race and ethnicity in representative populations at risk for ICH.

627 Background: The SLI of the BEACON CRC phase 3 study assessed the safety and efficacy of the combination of the BRAF inhibitor encorafenib (ENCO) + MEK inhibitor binimetinib (BINI) + anti-EGFR antibody cetuximab (CETUX) in pts with BRAFV600E-mutant mCRC after 1 or 2 prior regimens. CEA and CA19-9 are tumor markers that are widely used to monitor the effectiveness of systemic therapies for mCRC; here we report on the association of CEA and CA19-9 changes while on treatment with clinical outcomes in pts from the SLI. Methods: Pts in the SLI received the triplet of ENCO 300 mg QD + BINI 45 mg BID + CETUX 400 mg/m2 (initial dose) then 250 mg/m2 QW in 28-day cycles. Pts were evaluated for safety, radiographic response, and change in tumor markers CEA and CA19-9. Results: 30 pts were treated. The triplet was generally well tolerated, and adverse events were consistent with known BRAF, MEK, and EGFR inhibitor toxicities. The rate of severe skin toxicities (grade 3/4) was lower than generally observed for CETUX in mCRC. Of the 29 pts with a BRAFV600E mutation, the median time on study treatment was 5.6 mo (range, 1.0–9.3 mo), and 22 (76%) remained on study treatment at the time of data cutoff. The confirmed overall response rate (ORR) was 41%, with 1 complete and 11 partial responses. In addition, 9 pts had prolonged stable disease (SD) up to 9.3 mo.CEA and CA19-9 were analyzed in 28 pts. CEA and CA19-9 decreased in 96% and 82% of these pts, respectively. Among 15 pts with treatment duration ≥5.6 mo, median/mean % decreases were 97%/79% for CEA and 92%/82% for CA19-9 in confirmed responders (n=6). Respective decreases in pts with prolonged SD (n=9) were similar: 84%/68% for CEA and 89%/68% for CA19-9. Updated safety, efficacy, and tumor marker results will be provided. Conclusions: ENCO + BINI + CETUX is generally well tolerated and has encouraging clinical activity in BRAFV 600E mCRC, with a confirmed ORR of 41%. In pts with study treatment duration ≥5.6 mo, the tumor markers CEA and CA19-9 decreased markedly and to the same degree in responders vs pts with prolonged SD, providing additional evidence of the meaningful clinical activity of this regimen. Clinical trial information: NCT02928224.

Background and Purpose— We aimed to assess if renal function can aid in risk stratification for ischemic stroke or transient ischemic attack (TIA) recurrence and death in patients with embolic stroke of undetermined source (ESUS). Methods— We pooled 12 ESUS datasets from Europe and America. Renal function was evaluated using the estimated glomerular filtration rate (eGFR) and analyzed in continuous, binary, and categorical way. Cox-regression analyses assessed if renal function was independently associated with the risk for ischemic stroke/TIA recurrence and death. The Kaplan-Meier product limit method estimated the cumulative probability of ischemic stroke/TIA recurrence and death. Results— In 1530 patients with ESUS followed for 3260 patient-years, there were 237 recurrences (15.9%) and 201 deaths (13.4%), corresponding to 7.3 ischemic stroke/TIA recurrences and 5.6 deaths per 100 patient-years, respectively. Renal function was not associated with the risk for ischemic stroke/TIA recurrence when forced into the final multivariate model, regardless if it was analyzed as continuous (hazard ratio, 1.00; 95% CI, 0.99–1.00 for every 1 mL/min), binary (hazard ratio, 1.27; 95% CI, 0.87–1.73) or categorical covariate (likelihood-ratio test 2.59, P =0.63 for stroke recurrence). The probability of ischemic stroke/TIA recurrence across stages of renal function was 11.9% for eGFR ≥90, 16.6% for eGFR 60–89, 21.7% for eGFR 45–59, 19.2% for eGFR 30–44, and 24.9% for eGFR <30 (likelihood-ratio test 2.59, P =0.63). The results were similar for the outcome of death. Conclusions— The present study is the largest pooled individual patient-level ESUS dataset, and does not provide evidence that renal function can be used to stratify the risk of ischemic stroke/TIA recurrence or death in patients with ESUS.

Children with pediatric-onset multiple sclerosis and pediatric controls were enrolled across 16 pediatric multiple sclerosis centers in the United States and completed questionnaires that addressed time of first unaided walking and acquisition of 2-word phrases. A total of 467 (308 female) cases and 428 (209 female) controls were enrolled. Pediatric multiple sclerosis (n = 467) were not delayed in walking or using 2-word phrases compared to healthy controls (n = 428) (2.2% vs 5.7%, respectively). Children with disease onset before age 11 versus onset at 11 years or after were more likely to need an individualized education plan ( P = .002), reading assistance ( P = .0003), and math assistance ( P = .001). Children with multiple sclerosis onset prior to age 18 are not delayed in meeting the 2 major early developmental milestones but do have a significantly increased use of special services or learning assistance at school. Further research will need to address whether other measures of development (eg, rate of language acquisition or fine motor skills) differ between pediatric multiple sclerosis and controls.

Abstract Pyruvate kinase deficiency (PKD) is the most common enzyme defect of glycolysis and an important cause of hereditary, nonspherocytic hemolytic anemia. The disease has a worldwide geographical distribution but there are no verified data regarding its frequency. Difficulties in the diagnostic workflow and interpretation of PK enzyme assay likely play a role. By the creation of a global PKD International Working Group in 2016, involving 24 experts from 20 Centers of Expertise we studied the current gaps in the diagnosis of PKD in order to establish diagnostic guidelines. By means of a detailed survey and subsequent discussions, multiple aspects of the diagnosis of PKD were evaluated and discussed by members of Expert Centers from Europe, USA, and Asia directly involved in diagnosis. Broad consensus was reached among the Centers on many clinical and technical aspects of the diagnosis of PKD. The results of this study are here presented as recommendations for the diagnosis of PKD and used to prepare a diagnostic algorithm. This information might be helpful for other Centers to deliver timely and appropriate diagnosis and to increase awareness in PKD.