To test the hypothesis that the effect of valproate on dose-adjusted lamotrigine troughs was moderated by the ABCG2 c.421C>A (rs2231142) polymorphism, we conducted two studies in adults with epilepsy. Since Study 2 replicated the findings of Study 1, we analysed combined data [total N=471; 140 exposed to valproate (treated), 331 not exposed (controls)]. With adjustment for cotreatments and comorbidities, age, sex, body weight and polymorphisms (and linked polymorphisms) suggested to affect exposure to lamotrigine (UGT1A4*3 c.142T>G, rs2011425; UGT2B7 -161C>T, rs7668258; ABCB1 1236C>T, rs1128503) (by entropy balancing), primary analysis indicated: in variant carriers, geometric means ratio (GMR) [treated (n=21) vs. controls (n=72)] was around 60% higher than in wild-type subjects [treated (n=119) vs. controls (n=259)] – ratio of GMRs 1.61 (95%CI 1.23-2.11) and 1.63 (95%CrI 1.26-2.10), in the frequentist and Bayesian analysis, respectively. Similar differences in valproate effects between ABCG2 c.421C>A variant carriers and wild-type subjects were found in secondary analysis (adjustment by exact matching) when exposure to valproate was defined as valproate troughs up to 364 mol/L or ≥364 mol/L (vs. no exposure to valproate). Susceptibility of the estimates to (hypothetical) unmeasured confounding was low. Data suggests that polymorphism rs2231142 moderates the effect of valproate on exposure to lamotrigine.