Influence of the glutamate-glutamine cycle on valproic acid-associated hepatotoxicity in pediatric patients with epilepsy.

Introduction: Although valproic acid is generally well tolerated, hepatotoxicity is a common side effect in patients receiving long-term treatment. However, the mechanisms underlying valproic acid-associated hepatotoxicity remain elusive.

Methods: To investigate the mechanisms and explore the potential risk factors for valproic acid-associated hepatotoxicity, 165 age-matched pediatric patients were recruited for laboratory tests and glutamate-glutamine cycle analysis.

Results: The concentration of glutamate in patients with hepatotoxicity was significantly greater than that in control patients, while the concentration of glutamine in patients with hepatotoxicity was significantly lower than that in control patients (P <0.05). In addition, the frequencies of the heterozygous with one mutant allele and homozygous with two mutant alleles genotypes in glutamate-ammonia ligase rs10911021 were significantly higher in the hepatotoxicity group than those in the control group (47.1 percent versus 32.5 percent, P = 0.010; 17.6 percent versus 5.2 percent, P = 0.001, respectively). Moreover, heterozygous carriers with one mutant allele and homozygous carriers with two mutant alleles genotypes of glutamate-ammonia ligase rs10911021 exhibited significant differences in the concentrations of glutamine and glutamate concentrations (P ˂ 0.001 and P = 0.001, respectively) and liver function indicators (activities of aspartate aminotransferase, alanine aminotransferase, and gamma-glutamyl transferase, P <0.001, respectively). Furthermore, logistic regression analysis indicated that glutamate-ammonia ligase rs10911021 (P = 0.002, odds ratio: 3.027, 95 percent confidence interval, 1.521 - 6.023) and glutamate (P = 0.001, odds ratio: 2.235, 95 percent confidence interval, 1.369 - 3.146) were associated with a greater risk for hepatotoxicity, while glutamine concentrations were negatively associated with hepatotoxicity (P = 0.001, odds ratio: 0.711, 95 percent confidence interval, 0.629 - 0.804).

Discussion: Understanding pharmacogenomic risks for valproic acid induced hepatotoxicity might help direct patient specific care. Limitations of our study include the exclusive use of children from one location and concomitant medication use in many patients.

Conclusion: Perturbation of the glutamate-glutamine cycle is associated with valproic acid-associated hepatotoxicity. Moreover, glutamate-ammonia ligase rs10911021, glutamate and glutamine concentrations are potential risk factors for valproic acid-associated hepatotoxicity.