DNA methylation of antiseizure medication metabolism genes and pharmacokinetic changes in adult epilepsy patients treated with the modified Atkins diet

The modified Atkins diet is increasingly used as an adjunctive therapy for drug-resistant epilepsy. Recent evidence suggests that such dietary interventions may affect the pharmacokinetics of antiseizure medications and induce DNA methylation changes. However, it remains unclear whether epigenetic regulation of drug-metabolizing enzymes contributes to variability in medication serum levels. We analyzed 58 adults with drug-resistant epilepsy treated with a modified Atkins diet for 12 weeks. DNA methylation was profiled in peripheral blood before diet initiation (baseline) and after 4 weeks and 12 weeks using a genome-wide array. A total of 131 cytosine-phosphate-guanine (CpG) sites mapped to 13 candidate genes involved in antiseizure medication metabolism were assessed. One CpG in the CES1 gene showed significant methylation changes from baseline to 4 weeks, 12 weeks, and across all timepoints (FDR = 0.0018), indicating consistent longitudinal effects. Repeated measures correlation analyses were used to evaluate within-individual associations between DNA methylation and antiseizure medication serum concentrations. Two CpG–drug associations showed strong positive intra-individual correlations: CYP3A5 methylation was significantly associated with carbamazepine serum levels (repeated measures correlation coefficient = 0.81, FDR = 0.023), whereas CYP3A4 methylation showed a similarly strong but borderline association with topiramate (correlation coefficient = 0.76, FDR = 0.051). This study finds that a modified Atkins diet may be associated with DNA methylation variation at genes involved in drug metabolism. Furthermore, DNA methylation levels were strongly associated with individual serum concentrations of certain antiseizure medications. These findings are consistent with the concept of a diet–epigenome–drug metabolism axis and warrant further investigation in larger studies to assess clinical implications for personalized dietary and pharmacological management in epilepsy.