| 摘要: | Carbamazepine是一種被廣泛使用且治療濃度範圍狹窄的抗癲癇藥，而基因對 carbamazepine 的個體間劑量差異之影響，目前仍待了解。本研究將探討與 carbamazepine 藥物動力學及藥物效力學相關之基因多型性與其劑量、穩定狀態血中濃度與濃度與劑量之比值 (CDR) 的相關性。
本研究於國立台灣大學附設醫院神經科共收納了 189 位健康受試者與 234 位服用 carbamazepine 之癲癇症病人，並記錄病人服用劑量、血中濃度與 CDR 等臨床相關資訊。而這些病患皆服用carbamazepine超過六個月，已達維持劑量。在病患簽署受試者同意書後，採集10毫升的血液樣本，並以 PCR-RFLP 與 Real-Time PCR 的方法，分別對包含SCN1A，SCN2A，EPHX1，UGT2B7，ABCB1 與 ABCC2在內的共14個基因多型性位點進行基因型的檢測分析。
在相關性分析方面，我們分析各基因多型性位點與三個劑量分組 (< 800 mg，800-1199 mg 與 ≧1200 mg) 、穩定狀態血中濃度與兩個CDR分組 (CDR≦10 ?尳-1 與 CDR >10 ?尳-1) 之間的相關性。結果顯示，SCN1A IVS5-91G>A 與 EPHX1 c.337C>T 之基因多型性與carbamazepine的劑量與CDR之間，在統計上皆具有顯著的相關性。在SCN1A IVS5-91G>A基因多型性位點方面，帶有變異基因型或基因座者會較野生型者有較高的比例服用高劑量之 carbamazepine 與低CDR值 (p<0.0001)；在EPHX1 c.337T>C基因多型性位點方面，帶有變異基因型或基因座者會較野生型者有較高的比例服用高劑量之 carbamazepine 與低 CDR 值 (p<0.0001)。我們進一步在 SCN1Ac.3184A>G – IVS5-91G>A的單套型與單套型組合相關性分析中，發現帶有AA與GA單套型者會較帶有AG者有較高的比例服用高劑量之carbamazepine (p<0.0001與p=0.0030)，而帶有AA單套型者會較帶有AG者有較高的比例為低CDR值 (p=0.0003)。此外，帶有AG/AA, AA/AA, AA/GG 與 AA/GA單套型組合者會較帶有AG/AG單套型組合者有較高的比例服用高劑量之carbamazepine (p=0.0441, <0.0001, 0.0214與0.0021)。上述統計相關性經過Bonferroni correction後仍然存在顯著影響。在比例勝算羅吉斯回歸分析中，我們納入了病人的癲癇症類型、病因及併用抗癲癇藥等干擾因子於模式中。模式化分析結果顯示，併服levetiracetam、SCN1A IVS5-91G>A、EPHX1 c.337T>C，及ABCB1 c.1236C>T 與 c.2677G>T/A 之交互作用與個體間劑量差異具有顯著的相關性 (adjusted r2 = 0.71)；另一方面，併服 phenytoin、 lamotrigine、SCN1A IVS5-91G>A、EPHX1 c.337T>C與UGT2B7 c.802T>C 則會對個體間 CDR 差異造成顯著影響 (adjusted r2 = 0.37)。
本研究發現，癲癇症病人之基因多型性位點 SCN1A IVS5-91G>A 及 EPHX1 c.337T>C 會影響carbamazepine之劑量與 CDR。而本研究結果所建立的藥物基因體學模式，將能對患者投予較適合的 carbamazepine 劑量以治療其癲癇症。
Carbamazepine is a widely prescribed antiepileptic drug (AED) with narrow therapeutic index. The genetic effect on interindividual variations of dosage of carbamazepine was not well understood. The aim of the present study was to investigate the association between the variants in pharmacokinetics and pharmacodynamics related genes and the carbamazepine dosage requirement, serum concentration and concentration to dose ratio (CDR).
189 Healthy control subjects and 234 patients treated with carbamazepine were recruited from the department of Neurology of National Taiwan University Hospital (NTUH), and the daily dosage, serum concentration and CDR of carbamazepine was recorded. These patients were treated with carbamazepine over six months to achieve the maintenance dose. After signed informed consent, 10 ml blood was drawn to extract genomic DNA. The polymerase chain reaction (PCR)-restriction fragment length polymorphism (RFLP) and Real-Time PCR were applied to analyze the 14 single nucleotide polymorphisms (SNPs) in the candidate genes: SCN1A, SCN2A, EPHX1, UGT2B7, ABCB1 and ABCC2.
Association tests were conducted among the genotypes and alleles of each polymorphism and the three dosage groups (<800 mg, 800-1199 mg and ≧1200 mg), mean serum concentration and two CDR groups (CDR≦10 ?尳-1and CDR >10 ?尳-1). The results demonstrated that SCN1A IVS5-91G>A and EPHX1 c.337C>T polymorphisms were significantly associated with the dose of carbamazepine and CDR. For SCN1A IVS5-91G>A, the patients with the variants genotypes and alleles were more likely to take a higher dose of carbamazepine and lower CDR compared with wild-type (all p<0.0001). For EPHX1 c.337T>C, the patients with variants genotypes and alleles were more likely to take a higher dose of carbamazepine and lower CDR compared with wild-type (all p<0.0001). Further haplotype and haplotype combination analyses also revealed that the haplotypes composed of SCN1A c.3184A>G – IVS5-91G>A were significantly associated with carbamazepine dose and CDR. Patients with the AA and GA haplotypes were more likely to take a higher dose of carbamazepine compared with AG haplotype (p<0.0001 and p=0.0030, respectively). Moreover, the patients with AA haplotype were more likely to have lower CDR compared with AG haplotype (p=0.0003). Besides, patients with AG/AA, AA/AA, AA/GG and AA/GA haplotype combinations were more likely to take the higher doses of carbamazepine compared with AG/AG haplotype combination (p=0.0441, <0.0001, 0.0214 and 0.0021, respectively). These associations remained significant after Bonferroni’s corrections. The proportional odds regression analyses were conducted under adjustment of confounding factors, such as epilepsy syndromes, etiology classifications and concomitant AEDs. The results indicated that factors co-medication of levetiracetam, SCN1A IVS5-91G>A, EPHX1 c.337T>C and the interaction terms of ABCB1 c.1236C>T and c.2677G>T/A significantly affected the dosage of carbamazepine (adjusted r2 = 0.71). On the other hand, the factors co-medication of phenytoin, lamotrigine, SCN1A IVS5-91G>A, EPHX1 c.337T>C and UGT2B7 c.802T>C significantly affected the CDR of carbamazepine (adjusted r2 = 0.37).
Our findings demonstrated that genetic variants in SCN1A IVS5-91G>A and EPHX1c.337T>C genetic polymorphisms may be associated with the dose and CDR of carbamazepine in treatment of epilepsy patients. The present study set the stage for a prospective evaluation of how pharmacogenomic results may be used to optimize the dose of carbamazepine in epileptic patients. |
