The commonly used antiseizure medication lamotrigine is a substrate to ATP binding cassette subfamily G member 2 (ABCG2) transporter. The objective of this study was to evaluate the effect of the common loss-of-function polymorphism ABCG2 c.421C>A (rs2231142) on the lamotrigine trough concentrations at steady state in adults with epilepsy. In two consecutive studies (Study 1, Study 2) in patients on lamotrigine monotherapy, carriers of the variant ABCG2 c.421C>A allele (CA/AA) were considered exposed, and wild-type homozygotes (CC) were considered controls. They were mutually balanced on covariates (age, sex, body weight, several polymorphisms in genes encoding other transporter proteins and lamotrigine-metabolizing enzymes that have been suggested to affect exposure to lamotrigine) to estimate the exposure effect (geometric means ratios, GMRs) in each study separately and overall (individual patient data meta-analysis). The overall estimate was evaluated for sensitivity to residual confounding. In both studies (exposed n = 28 vs. controls n = 103; exposed n = 44 vs. controls n = 153, in Study 1 and Study 2, respectively) and overall (exposed n = 72 vs. controls n = 256), dose-corrected lamotrigine trough concentrations were moderately lower in the exposed patients: frequentist GMR [95% CI] = 0.82 [0.63-1.08]; GMR = 0.69 [0.60-0.81] and GMR = 0.72 [0.63-0.83] in Study 1, Study 2 and overall, respectively; Bayes GMR [95% CrI] = 0.83 [0.68-1.00]; GMR = 0.69 [0.58-0.83] and GMR = 0.75 [0.65-0.86] in Study 1, Study 2 and overall, respectively. Estimates appeared resistant to unmeasured confounding-the E-values for the pooled point estimates were high, and estimates corrected for a strong hypothetical bias were GMR = 0.78 [0.68-0.90] frequentist and GMR = 0.81 [0.70-0.93] Bayes. Polymorphism ABCG2 c.421C>A moderately reduces lamotrigine concentrations in adults with epilepsy.
Read full abstract