Lack of response to quinidine in KCNT1-related neonatal epilepsy

Objective To evaluate the clinical efficacy and safety of quinidine in patients with KCNT1-related epilepsy of infancy with migrating focal seizures (EIMFS) in the infantile period and to compare with the effect of quinidine on mutant channels in vitro. Methods Results We identified 4 patients with EIMFS with onset in the neonatal period, pathogenic variants in the KCNT1 gene, and lack of response to AEDs. Patients were prospectively enrolled, treated with quinidine, and monitored according to a predefined protocol. Electroclinical, neuroimaging, and genetic data were reviewed. Two patients had novel variants in the KCNT1 gene that were modeled in Xenopus oocytes with channel properties characterized using electrophysiology recordings. Three of four patients were treated with quinidine early in their disease course, prior to 6 months of age. No significant side effects were noted with quinidine therapy. In addition, there were no significant changes in electroencephalography (EEG)-confirmed seizure burden during therapy, and patients had near hourly seizures before, during, and after treatment. Two patients had previously reported gain-of-function mutations, which demonstrated sensitivity to high levels of quinidine in the oocyte assay. Two patients with novel variants, showed characteristic gain-of-function and were thus predicted to be pathogenic. Of interest, these variants were essentially insensitive to high levels of quinidine. Significance Patients had no reported benefit to quinidine therapy despite age at treatment initiation. Pharmacogenetic results in oocytes were consistent with clinical treatment failure in 2 patients, suggesting that single-dose pharmacologic assessment may be helpful in predicting which patients are exceedingly unlikely to achieve benefit with quinidine. In the 2 patients who had a lack of therapeutic benefit despite sensitivity to high concentrations of quinidine with in vitro oocyte assay, it is likely that the achievable exposure levels in the brain were too low to cause significant in vivo channel blockade.