4 Discussion
As a clinical first-line antiepileptic drug, CBZ has a relatively narrow therapeutic range (4–12μg/mL) and wide inter-individual variability [34]. Genetic variants in ABCB1, EPHX1, SCN1A genes are wellknown to influence the individual CBZ daily maintenance doses, metabolism and resistance for epileptic patients. Thus, clarify the role of ABCB1, EPHX1, SCN1A genetic polymorphisms in pharmacokinetics and pharmacodynamics of CBZ will be helpful for the effective clinical individualized treatment of epilepsy.
The ABCB1 (c.3435C>T, c.2677G>T/A and c.1236C>T) in exon 26 closely related with the altered expression levels of P-gp, range of CBZ resistance in epileptics. In the meta-analysis, c.3435C>T andc.2677G>T/A polymorphisms were found to be associated with adjusted concentrations of CBZ. For the high heterogeneity in c.2677G>T/A , we did not conduct subgroup analysis or sensitivity analysis limited by only 3 articles included. In order to seek the mechanism ofc.3435C>T on the clinical effect of CBZ in epilepsy patients, we further analyzed whether c.3435C>Tpolymorphism change the clinical efficacy by influencing CBZ resistance, and found that patients with the c.3435C>T TTgenotype exhibited lower CDZ resistance compared with CT or(CT + CC) genotype, Hongmei Meng’s study [29] also supported it. This might attributable to the CC genotype ofc.3435C>T enhancing the expression of P-gp and excessive CBZ efflux across the blood-brain barrier (BBB) resulting in drug resistance in epileptics. However, the included studies of Armond Daci [8] and Gulay Oner Ozgon’s [30] results unsupported it in Kosovar Albanian and Turkey epileptic patients. This phenomenon may be attribute to ethnic differences, but we did not conduct racial analysis limited number of literatures included. Overall, our results confirmedc.3435C>T polymorphism may influence the clinical efficacy via regulating CBZ resistance followed by adjusting serum concentration of CBZ although more high-quality original studies are still required.
As an active metabolites of CBZ, CBZ-10,11-epoxide (CBZE) is further transformed to an inactive metabolite, CBZ-10,11-diol (CBZD) via microsomal epoxide hydrolase (mEH) which encoded by EPHX1 [35, 36]. Thus, EPHX1 (c.337T>C and c.416A>G)mutation may affect CBZ pharmocokinetics and pharmacodynamics by changing the function of mEH [36, 37]. Our study found that the patients with AA genotype in c.416A>G has lower level of CDRCBZD, prompting that CBZ metabolism may be inhibited by the c.416A>G variant [37], which has been demonstrated to decrease mEH activity in Yukiko Nakajima’s study [38]. However, no association between any genetic models of EPHX1 c.416A>G polymorphism and CBZ resistance was found, which supported by Wenting Yun et al. study [25]. In addition, we abandoned high heterogeneity data of CDRCBZE (for c.337T>C ) and the instability data of CDRCBZD : CDRCBZEratio (for c.416A>G ) limited by 2 articles with small samples. Furthermore, no association was found between EPHX1 c.416A>G and CBZ resistance in our present study. Therefore, EPHX1 c.416A>G associated with CBZ metabolism and resistance need to be further studied in the larger cohorts of CBZ for epilepsy.
IVS5-91G>A located at 5 splice donor site ofSCN1A exon 5N, plays a crucial role in epilepsy by encoding the S4 transmembrane segment or voltage sensor in the first domain of the Nav1.1 protein and initiating action potentials in neurons in different parts of the mammalian brain [39, 40]. Whilec.3184A>G polymorphism in SCN1A may influence the structural and functional properties of sodium channels and further impact treatment of CBZ for epilepsy via converting threonine to alanine [24]. It was noteworthy that, no association was found based on the present included studies containing Chinese, Japanese, Serbia, and Pakistan epilepsy patients among SCN1A (c.3184A>G and IVS5-91G>A) and CBZ resistance, which cannot be further supported by CBZ metabolism for small numbers of literatures and samples included. This may be due to the lack of data for further evaluation differences in the geographical distribution of the subjects or environmental factors [41]. Consequently, further investigations based on more well-designed experiments and larger sample size should be performed to verify the effects of SCN1A (c.3184A>G and IVS5-91G>A) on CBZ resistance and metabolism.
This meta-analysis first comprehensively collected the existing studies and reviewed the relationships among ABCB1 (c.3435C>T) , EPHX1 (c.337T>C and c.416A>G) , SCN1A (c.3184A>G and IVS5-91G>A) polymorphisms and CBZ metabolisms and resistance. Compared with precious meta-analysis [15], the present study not only evaluated the associations between the ABCB1 (c.3435C>T) , EPHX1 (c.337T>C and c.416A>G) , SCN1A (c.3184A>G and IVS5-91G>A) polymorphisms and drug-resistance risk, but also clarified the effect of corresponding genes on CBZ metabolism in epileptics. Consequently, our data suggest that ABCB1 c.3435C>T and EPHX1 c.416A>Gpolymorphisms could influence inter-individual variability of CBZ metabolism and resistance.
The inconsistencies in the studies may be explained by some phenomena. Firstly, only English and Chinese studies were included, many literatures in other languages were ignored which may lead to the existence of bias. Secondly, the small sample size may result in false-positive problem and weaken the statistical power. Thirdly, the optimal dosage of CBZ may be influenced by a variety of confounding factors such as age, sex, ethnicity, and the interaction between gene mutations, but the lack of original data limited our ability to further assess these factors. Fourthly, lack of some important raw data limits the further evaluation of other indicators, such as the therapeutic effect of concentration-time curves on metabolic clearance and adverse events. Furthermore, based on Egger’s test, we detected a small publication bias which might be caused by unpublished data, because negative studies were less likely to be published and included in journals and computerized database, and only studies indexed by the selected databases were included, this results in a potential overestimation of effect sizes. Finally, different doses and treatment durations were used in the included studies, which may affect the observable effect of pharmacokinetic genes. These caveats should be considered and overcome in future large sample and multiracial studies. In spite of above potential limitations, our meta-analysis collected the most recently published data on the association between ABCB1, EPHX1 gene polymorphisms and CBZ metabolism and resistance.