Search

Background and Objectives: The SLC35A2 gene, located at chromosome Xp11.23, encodes for a uridine diphosphate-galactose transporter. We describe clinical, genetic, neuroimaging, EEG, and histopathologic findings and assess possible predictors of postoperative seizure and cognitive outcome in 47 patients with refractory epilepsy and brain somatic SLC35A2 gene variants., Methods: This is a retrospective multicenter study where we performed a descriptive analysis and classical hypothesis testing. We included the variables of interest significantly associated with the outcomes in the generalized linear models., Results: Two main phenotypes were associated with brain somatic SLC35A2 variants: (1) early epileptic encephalopathy (EE, 39 patients) with epileptic spasms as the predominant seizure type and moderate to severe intellectual disability and (2) drug-resistant focal epilepsy (DR-FE, 8 patients) associated with normal/borderline cognitive function and specific neuropsychological deficits. Brain MRI was abnormal in all patients with EE and in 50% of those with DR-FE. Histopathology review identified mild malformation of cortical development with oligodendroglial hyperplasia in epilepsy in 44/47 patients and was inconclusive in 3. The 47 patients harbored 42 distinct mosaic SLC35A2 variants, including 14 (33.3%) missense, 13 (30.9%) frameshift, 10 (23.8%) nonsense, 4 (9.5%) in-frame deletions/duplications, and 1 (2.4%) splicing variant. Variant allele frequencies (VAFs) ranged from 1.4% to 52.6% (mean VAF: 17.3 ± 13.5). At last follow-up (35.5 ± 21.5 months), 30 patients (63.8%) were in Engel Class I, of which 26 (55.3%) were in Class IA. Cognitive performances remained unchanged in most patients after surgery. Regression analyses showed that the probability of achieving both Engel Class IA and Class I outcomes, adjusted by age at seizure onset, was lower when the duration of epilepsy increased and higher when postoperative EEG was normal or improved. Lower brain VAF was associated with improved postoperative cognitive outcome in the analysis of associations, but this finding was not confirmed in regression analyses., Discussion: Brain somatic SLC35A2 gene variants are associated with 2 main clinical phenotypes, EE and DR-FE, and a histopathologic diagnosis of MOGHE. Additional studies will be needed to delineate any possible correlation between specific genetic variants, mutational load in the epileptogenic tissue, and surgical outcomes., (© 2022 American Academy of Neurology.)

Background and Objective: To test the hypothesis that a multicenter-validated computer deep learning algorithm detects MRI-negative focal cortical dysplasia (FCD)., Methods: We used clinically acquired 3-dimensional (3D) T1-weighted and 3D fluid-attenuated inversion recovery MRI of 148 patients (median age 23 years [range 2-55 years]; 47% female) with histologically verified FCD at 9 centers to train a deep convolutional neural network (CNN) classifier. Images were initially deemed MRI-negative in 51% of patients, in whom intracranial EEG determined the focus. For risk stratification, the CNN incorporated bayesian uncertainty estimation as a measure of confidence. To evaluate performance, detection maps were compared to expert FCD manual labels. Sensitivity was tested in an independent cohort of 23 cases with FCD (13 ± 10 years). Applying the algorithm to 42 healthy controls and 89 controls with temporal lobe epilepsy disease tested specificity., Results: Overall sensitivity was 93% (137 of 148 FCD detected) using a leave-one-site-out cross-validation, with an average of 6 false positives per patient. Sensitivity in MRI-negative FCD was 85%. In 73% of patients, the FCD was among the clusters with the highest confidence; in half, it ranked the highest. Sensitivity in the independent cohort was 83% (19 of 23; average of 5 false positives per patient). Specificity was 89% in healthy and disease controls., Discussion: This first multicenter-validated deep learning detection algorithm yields the highest sensitivity to date in MRI-negative FCD. By pairing predictions with risk stratification, this classifier may assist clinicians in adjusting hypotheses relative to other tests, increasing diagnostic confidence. Moreover, generalizability across age and MRI hardware makes this approach ideal for presurgical evaluation of MRI-negative epilepsy., Classification of Evidence: This study provides Class III evidence that deep learning on multimodal MRI accurately identifies FCD in patients with epilepsy initially diagnosed as MRI negative., (© 2021 American Academy of Neurology.)