This study found that within a certain dose range of 8–46 mg/kg, the MHD plasma concentration was positively correlated with the maintenance dose of OXC, which was consistent with previously reported literature [7, 8]. The Perucca’s study[7] showed that MHD plasma concentrations in children are considerably lower than those adults receiving comparable OXC dosages, possibly due to lower body fat/lean mass ratio, higher total body water, higher plasma protein binding rate, and higher renal blood flow in children.Other study has shown that children younger than 8 years have clearance rates 30–40% higher than those in older children[9]. As a result, children, especially younger children, need to be given more dosage to achieve the same plasma concentration. There were not many studies on the correlation between gender and age with MHD plasma concentration, and the conclusions were also different. In the previous study, the MHD plasma concentration increased with patient age[10]. Another study in the Korean population found that there are not associations between OXC concentration and patient age, weight, gender, or seizure type[11]. Partially consistent with previous studies, we only found that CDR increased with patient weight, and corresponding maintenance dose decreased with patient weight. This suggests that the pharmacokinetic behavior of MHD is weight-dependent, and its clearance rate tends to decrease with the increase of weight, which can be further explored. In terms of the duraton of epilepsy, it was found that the CDR in the ≥ 6 years group was significantly higher than that in the < 1 year group, 1–3 years group and 3–6 years group. The possible reason is that the age of the children in the disease course ≥ 6 years group is relatively older, and the renal clearance rate is relatively lower, resulting in higher concentration.
The UDP-glucuronosyltransferase (UGT) superfamily participates in the glucuronidation metabolism of various endogenous substances and exogenous compounds[12]. The elimination of OXC and MHD is mainly by binding with UGT enzyme to form glucuronic acid conjugate, which is then excreted in the urine. In our study, of the four SNPs (UGT1A4 rs2011425, UGT1A6 rs6759892, UGT1A9 rs2741049 and UGT2B15 rs1902023), we found UGT1A4 rs2011425 and UGT2B15 rs1902023 were significantly associated with MHD plasma concentretion, which was inconsistent with Lu et al’s study[13], as MHD plasma concentration of UGT1A9 rs2741049 mutant allele T carriers was significantly lower than those non-carriers. It may be explained by differences in the age of the population, because Lu et al included both children and adults in their study. In the pediatric population, smaller variation in microsomal glucuronidation activity was observed compared with that in adults, and the activities of UGT1A4, UGT1A6, UGT1A9 and UGT2B15 increased with age[14]. The human UGT1A4 gene is primarily expressed in the human liver, both in the biliary epithelium and the hepatocyte. In contrast to other UGT1A proteins, the catalytic activity of UGT1A4 is most specific for primary and secondary amines, sapogenins, as well as for steroid hormones, which are commonly present in therapeutic drugs[15]. At present, a large number of studies have reported that the variation of UGT1A4 rs2011425 is related to lamotrigine metabolism[16–18]. Reimers et al’s study[18] found that the influence trend of UGT1A4 rs2011425 on enzyme activity depended on substrate type. For LTG glucuronidation, UGT1A4 rs2011425 variant showed reduction in activity as compared with the wild-type enzyme. Theoretically, UGT1A4 rs2011425 mutant LTG plasma concentration was higher than those wild-type. In this study, it was found that the UGT1A4 rs2011425 mutant allele carriers had higher plasma concentration, and whether OXC had the same influence trend on mutant enzyme activity as lamotrigine remains to be further studied. UGT2B15 was identified as important for the glucuronidation of androgenic steroids, drugs, drug metabolites and other xenobiotic[19]. The mutant allele of UGT2B15rs1902023 was found to reduce the systemic clearance of lorazepam, while also reducing the glucuronidation level of oxazepam in the liver, suggesting that the mutant allele carriers had a relatively higher plasma concentration[20]. Consistent with the above research, we found that the carriers of the variant allele of UGT2B15rs1902023 had higher MHD plasma concentration than homozygous wild-type. As for UGT2B7 rs7439366, the allele frequency of the nonsynonymous polymorphism in the Chinese population is 32.8%[21]. Our study demonstrated that UGT2B7 rs7439366 gene polymorphism was not correlated with MHD plasma concentration, which was in agreement with Shen et al’s study[22] and Ma et al’s study[23].
Cytochrome P450 (CYP450) system is a major group of biotransformational enzymes in many organisms, including humans, and is used for the metabolism of endogenous and exogenous compounds[24]. OXC metabolism is largely unaffected by CYP450 system. However, OXC can inhibit CYP2C19 and induce CYP3A4/5. At present, studies on the influence of CYP450 gene polymorphism on MHD plasma concentration are as follows: CYP3A4 rs4646440, CYP3A4 rs2242480, CYP3A5 rs15524, CYP3A5 rs776746, CYP2C19*2 rs4244285 and CYP2C19*3 rs4986893[25–27]. In this study, no correlation was found between the polymorphisms of the above genes and MHD plasma concentration, which was inconsistent with Zhou et al’s study[25], who considered that the plasma concentration of patients carrying CYP2C19*3 mutant allele was higher. The most characterized alleles of the CYP2C19 are CYP2C19*2 and CYP2C19*3, and CYP2C19*2 is the most common allele among Asians[28]. CYP2C19 is responsible for catalytic oxidation and metabolic clearance of up to 20% of clinically important antiepileptic drugs, such as phenobarbital and phenytoin[29]. However, previous study suggested that there was no statistical difference in the antiepileptic drug inhibition effect between CYP2C19*2 and CYP2C19*3 genotypes[30].
In addition to the effects of the above metabolic enzymes, new progress has been made in recent years on the effect of transporter (ABCB1, ABCC2 coding) related gene polymorphism on the plasma concentration of MHD. The ABCB1 gene encodes a highly polymorphic P-gp. ABCB1 rs1045642 in exon 26 is the most widely studied SNP. According to the current research, we have concluded that P-gp is involved in the transport of OXC and MHD at the blood-brain barrier, which means that they are the substrates of P-gp[31, 32]. A study found[22] that ABCB1 rs1045642 gene polymorphism was significantly associated with CDR, and CDR of CC genotype was higher than that of CT and TT genotype, which was consistent with our results. However, another study[26] involving 40 Chinese patients with epilepsy found no correlation between ABCB1 (rs1045642, rs2032582, rs1128503 and rs10234411) and the CDR. In addition, ABCC2 is also a member of the ATP binding cassette transporter superfamily. The overexpression of ABCC2 transporters on the blood-brain barrier cells will increase the outflow of AEDs into the capillary cavity, which ultimately leads to a reduction of AEDs concentration in the brain that is insufficient to control seizures[33]. In vitro studies indicated that CBZ is a substrate of ABCC2[34], and the change of ABCC2 function will affect the absorption and excretion of its substrate. OXC is a 10-ketone analogue of CBZ, and whether ABCC2 affects the maintenance dose and plasma concentration of OXC needs to be further studied. Our study failed to observe a significant correlation among ABCC2 rs3740066 and ABCC2 rs2273697 and MHD plasma concentration, which was consistent with previous studies[22, 23]. However, Ma et al’s study[23] found that carriers of the ABCC2 rs2273697 variant alleles required significantly higher OXC maintenance doses than non-carriers.
OXC plays an antiepileptic role by the inhibition of sodium channel activity. Sodium channels consist of one large α-subunit and two small β-subunits[35]. Variants in four isoforms of the α-subunit, Nav1.1, Nav1.2, Nav1.3,and Nav1.6 (encoded by the SCN1A, SCN2A, SCN3A and SCN8A genes, respectively), have been associated with epilepsy[36]. The association of SCN1A and SCN2A polymorphisms with MHD plasma concentration and OXC maintenance dose has been discussed in several studies. In a study on Chinese Han patients with epilepsy[23], of the four candidate alleles (SCN1A rs3812718, SCN1A rs2298771, SCN2A rs17183814 and SCN2A rs2304016), only SCN1A rs3812718 was significantly associated with OXC maintenance dose and CDR, which was inconsistent with the Italian populations[37]. It may be explained by differences in ethnic inclusion. Another study[38] found that the SCN2A rs17183814 mutant required a lower OXC maintenance dose and higher CDR in the over-weight group, while higher OXC maintenance dose in the low-weight group. In this study, we failed to observe a significant correlation among SCN1A/SCN2A genotypes and OXC maintenance dose and MHD plasma concentration.
The voltage-dependent calcium channel is composed of "α1", "β", "α2δ" and "γ" subunits, in which "α1" is the functional subunit and "β", "α2δ" and "γ" are the auxiliary subunits[39]. According to different pharmacological characteristics, voltage-dependent calcium channels can be divided into P/Q-type (conducted by Cav2.1 channel, which is encoded by the CACNA1A genes), N-type (conducted by Cav2.2 channel, which is encoded by the CACNA1B genes), R-type (conducted by Cav2.3 channel, which is encoded by the CACNA1E genes), L-type (conducted by Cav1.1-1.4 channels, which is encoded by the CACNA1S, CACNA1C, CACNA1D and CACNA1F genes, respectively) and T-type (conducted by Cav3.1-3.3 channels, which is encoded by the CACNA1G, CACNA1H and CACNA1I genes, respectively)[40]. In a previous study[41], minor allele of two polymorphisms (CACNA1H rs61734410, CACNA1I rs3747178) in the ethosuximide subjects were more common in not seizure free participants, and two CACNA1H polymorphisms (rs2753326, rs2753325) in the lamotrigine subjects appeared more commonly among seizure free participants, and one polymorphism of CACNA1H rs2235634 in the valproic acid subjects was associated with increased rates of not seizure free outcomes. Whether this correlation also exists in OXC requires further study. In our study, we discussed five polymorphisms of CACNA1H and CACNA1I, and it was found that the CDR of CACNA1H rs2753325 heterozygous AG type was significantly higher than mutant homozygous GG type, suggesting that the gene polymorphism of calcium channel may be related to OXC maintenance dose and MHD plasma concentration.