Prevalence of UGT1A6 polymorphisms in children with epilepsy on valproate monotherapy
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/0028-3886.152631
Source of Support: None, Conflict of Interest: None
Background: Valproate is a commonly used anticonvulsant drug. Uridine 5΄-diphospho (UDP)-glucuronosyltransferase (UGT) contributes to around 50% of valproate metabolism and its polymorphisms may be important for explaining the considerable variation in valproate levels in patients with epilepsy.
Keywords: Epilepsy; pharmacogenetics; UGT1A6 polymorphisms; valproate
Pharmacogenetics refers to the science about how genetic variations affect the drug metabolism, drug targets or disease pathways leading to varying response to the drug with regard to its efficacy or adverse effect profile. , The concept of "individualized medicine" is evolving and there has been a paradigm shift from the concept of "one drug fits all" to "right drug for the right patient at the right dose and time."
Valproate is one of the most commonly used anti-epileptic drugs in childhood epilepsy. It is extensively metabolized in liver through microsomal glucuronide conjugation [mediated by uridine 5'-diphospho (UDP) glucuronosyltransferase (UGT)]; mitochondrial β-oxidation; and cytochrome P450 (CYP)-dependent ω-, (ω-1)-, and (ω-2)-oxidation.  Less than 5% is excreted unchanged in urine. The CYP isoforms involved in the metabolism of valproate include CYP2C9, CYP2A6, CYP2B6, and possibly CYP2C19.  UGT1A6, UGT1A9, and UGT2B7 are the major UGT isoforms capable of glucuronidating valproic acid and this pathway contributes to around 50% of valproate metabolism. 
Previous research has focused on CYP polymorphisms. ,, However, UGT polymorphisms may be more important for explaining the considerable variation observed in valproate levels in patients with epilepsy. The three most common non-synonymous UGT1A6 polymorphisms include S7A, T181A, and R184S. ,, The literature supports either an increased or a similar activity between wild-type and non-wild-type isozymes. However, few studies report conflicting results and the majority of these studies are in Chinese population. ,,,, Thus, the results cannot be generalized. This formed the rationale of the current study. The purpose of the study was to determine the prevalence of UGT1A6 polymorphisms in North Indian children with epilepsy on valproate monotherapy attending the pediatric neurology clinic at a tertiary care center and how these polymorphisms affect serum valproate levels.
This cross-sectional study was carried out in the Department of Pediatrics in a tertiary care center in North India from March 2011 to July 2012. Children aged 3-12 years with a diagnosis of epilepsy  on valproate monotherapy (plain tablets) in a stable dose for at least 1 month were included. UGT1A6 enzyme activity may take 10 months to reach the adult level. Hence, the enrolled children had mature enzyme activity. Children receiving drugs which may interfere with valproate metabolism  or herbal or alternative medicine, with inadequate compliance or having chronic hepatic, renal, or cardiac disease were excluded. The compliance was evaluated by detailed questioning of the parents and, wherever possible, of the child and by checking for leftover sachets. After informed consent was obtained all enrolled children underwent detailed clinical evaluation and clinical screening for the adverse effects of valproate.
Five milliliters of venous blood was obtained immediately before the morning dose of valproate. The blood samples were separated into three portions, one of which was centrifuged immediately to obtain plasma and then stored at −80°C until used for valproate level estimation and the second sample was immediately stored at 4°C until used for deoxyribonucleic acid (DNA) extraction. The third portion was sent for renal and hepatic function tests along with the estimation of calcium, phosphorus, and alkaline phosphatase.
Steady-state trough plasma concentrations of valproate were determined by High-Performance Liquid Chromatography (HPLC) analysis. Trough plasma concentrations of valproate (VPA) were standardized by adjusting with patients' weight and dose and expressed as C S (C S = trough plasma concentration/[daily dose/weight]). 
Genomic DNA was extracted from peripheral blood using QIAamp DNA Mini Kits according to the manufacturer's recommendations. DNA concentrations were determined by measuring absorbance at 260 nm using AlphaSpec (Alpha Innotech, San Leandro, CA, USA). The UGT1A6 polymorphisms (S7A, T181A, and R184S) were studied by polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP). PCRs were performed in 25 μl volume using forward (5'- GAGCCCTGTGATTTGGA-3') and reverse (5'-CAGCTGATGCGAGTTCTT-3') primers of UGT1A6.  The reactants were subjected to 95°C for 5 min, and then 30 cycles of 94°C for 30 minutes, 56°C for 1 minute, and 72°C for 1 minute. The final extension was at 72°C for 10 minutes. A 768 bp region of the exon was amplified to supply substrate to screen for sensitivity to Hha I, Nsi I, and Bbv I endonuclease treatment separately according to the supplier's instructions. The reaction mixtures were subjected to electrophoresis through a 2% agarose gel with ethidium bromide. Random samples were selected for validation by DNA sequencing done on Applied Biosystems ® 3130 Genetic Analyzer according to the manufacturer's instructions.
Based on the genotype frequencies reported previously,  the calculated sample size was 185 (power 80%, alpha at 0.05).  However, due to limited resources, a smaller sample size could be studied.
The summary statistics were used to summarize the baseline characteristics. The prevalence was reported as frequencies (%) with 95% confidence intervals. Hardy-Weinberg equilibrium analysis of UGT1A6 single nucleotide polymorphisms (SNPs) was performed using the site, http://www.oege.org/software/hwe-mr-calc.shtml. The mean standardized serum valproate levels between different UGT1A6 functional genotypes were compared using analysis of variance (ANOVA) and Kruskal-Wallis test. P value less than 0.05 was considered as statistically significant. Statistical analysis was performed using STATA 11.0. The study was approved by the ethics committee of the Institute.
Ninety-four children were screened for enrollment during the study period. Four children were on anti-tubercular therapy, and 10 had inadequate drug compliance and hence were excluded from the study. Eighty children were subsequently analyzed. The baseline characteristics of the study population are tabulated in [Table 1].
The causes of epilepsy in this study population included neurocysticercosis (37.5%), associated cerebral palsy (30%), generalized epilepsy of unknown etiology (13.8%), epilepsy with febrile seizure plus (6.3%), focal epilepsy of unknown etiology (5%), associated intellectual disability (2.5%), and miscellaneous causes (5%; calcified tuberculoma, Sturge- Weber syndrome More Details, Rolandic epilepsy, and Landau-Kleffner syndrome).
The frequencies of each UGT1A6 genotype are shown in [Table 2]. The frequencies were consistent with the Hardy-Weinberg equilibrium. The mean serum valproate observed was 66.8 (standard deviation [SD]-35.3 μg/ml). The mean standardized serum valproate concentration was 3.51 (SD-2.50). No significant association was observed between valproate doses, serum valproate concentration or standardized valproate concentration, and UGT1A6 genotypes [Table 3].
The most common adverse effects associated with valproate were sedation (20%) and headache (6.3%). The other adverse effects reported by the parents were gastrointestinal complaints (3.8%), weight gain (3.8%), hair loss (3.8%), and hyperactivity (2.5%). None of these adverse effects were disabling, and they were well tolerated. All the children had normal calcium and phosphorus, but 13.8% children had elevated alkaline phosphatase levels. Only two children had significant asymptomatic elevation of transaminases (> 3 times normal).
Various genetic polymorphisms may contribute to the inter-individual variability in pharmacokinetics and pharmacodynamics of valproate. As previously noted, hepatic glucuronidation contributes to around 50% of valproate metabolism in humans. UGT1A6 is one of the major UGT isoforms involved in valproate metabolism. Thus, this study was done to find the prevalence of various UGT1A6 polymorphisms and how these affect the serum valproate levels in children with epilepsy on valproate monotherapy.
The genotype frequencies in the current study as compared to previous studies ,,,,, are shown in [Table 4]. Recently, Munisamy et al.  explored the prevalence of UGT1A6 polymorphisms (T181A and R184S) in adults with epilepsy on valproate monotherapy with and without valproate toxicity (n = 100). The overall genotype prevalence of T181A (P = 0.13) and R184S (P = 0.17) in both their groups were similar to our study. They also showed a significant genotypic as well as allelic association with valproic acid toxicity for UGT1A6 (T181A and R184S) polymorphic enzymes. The adverse effects were infrequent and mild in our study population.
The polymorphisms of UGT1A6 had no significant effect on the valproate dose requirements, serum valproate concentrations, and standardized serum valproate concentrations in the study population. Two previous studies had shown similar negative results. Wang et al.  studied UGT1A6 541A > G polymorphism in 147 children with epilepsy and found no effect of the polymorphism on the serum valproate levels. Chu et al.  studied UGT1A6 polymorphisms (541A > G and 552A > C) in 242 adult epilepsy patients and reported no significant effect of polymorphisms on the standardized serum valproate concentrations.
Few large studies have shown conflicting data. The carriers of the variant UGT1A6 tended to have lower serum valproate levels than the wild genotype. ,, Larger studies, studies in different ethnic backgrounds, and meta-analyses of current data will help clarify the functional impact of UGT1A6 polymorphisms.
The clinical use of the majority of the drugs still depends on empirical data and doctor's experience. The dosing, effectiveness, and adverse effects are not fully understood. Further, the metabolism of drugs is extremely complex and multifactorial. More studies are required to acquire better understanding of the drug metabolism and clinically implement an "individualized patient treatment." Although many polymorphisms have been discovered in genes responsible for valproate metabolism, their impact on valproate glucuronidation or its clinical significance is still unclear. Furthermore, there are significant ethnic variations in haplotype frequencies. Thus, conclusions made from one population cannot be safely applied on another.
However, this is the first study exploring the prevalence of UGT1A6 polymorphisms in Indian population and exclusively in pediatric patients with epilepsy. This study done in a small population reports no association between UGT1A6 polymorphisms and valproate dose requirements or serum valproate levels. This result is conflicting with the results of previous large Chinese studies. Further, in this study, stringent measures were taken to ensure steady-state valproate levels at the time of withdrawing blood. Other factors which may affect serum valproate levels were minimized.
However, a small sample size and unknown prevalence of UGT1A6 polymorphisms in normal Indian population were the limitations of this study. Only one UGT isoform could be studied in view of limited resources. Further, this study was underpowered to detect the differences in serum valproic acid levels or standardized concentrations in different UGT1A6 genotypes.
To summarize, the common heterozygotes of each of the three UGT1A6 polymorphisms were found in nearly half of the children with epilepsy from North Indian origin. The association between valproate doses, serum valproate concentration or standardized valproate concentration, and the various UGT1A6 genotypes could not be studied reliably as the study was underpowered. And finally, valproate was well-tolerated in this study population.
The authors are grateful to all the Ph. D students and the technicians in the genetics and pharmacology laboratories, especially Mr Pankaj Sharma (Genetics) at the All India Institute of Medical Sciences, New Delhi for their assistance provided during this study.
[Table 1], [Table 2], [Table 3], [Table 4]