META-ANALYSIS OF CYP2C19 ASSOCIATION WITH EFFICACY AND SIDE EFFECTS OF CITALOPRAM AND ESCITALOPRAM USING DATA FROM GENOME-WIDE ASSOCIATION STUDIES

Background Major Depressive Disorder (MDD) is a leading cause of disability-adjusted life years worldwide. Genetic polymorphisms may have the potential to guide antidepressant treatment to increase current remission rates. Cytochrome P450 (CYP450) enzymes are responsible for antidepressant metabolism, with CYP2C19 playing a major role in citalopram/escitalopram metabolism. This study is the first meta-analysis to investigate how CYP2C19 polymorphisms predict citalopram/escitalopram efficacy and side effects. Methods CYP2C19 phenotypes comprise poor metabolizers (PM), intermediate and intermediate+ metabolizers (IM; IM+), extensive and extensive+ metabolizers (EM [wild type]; EM+) and ultra-rapid metabolizers (UM) defined by the two most common CYP2C19 functional polymorphisms (rs4244285 and rs12248560). These polymorphisms were directly genotyped or imputed from genome-wide data in four MDD pharmacogenetic samples treated with citalopram or escitalopram (GENDEP, STAR*D, GenPod and PGRN-AMPS). Treatment efficacy was (1) percentage symptom improvement (corrected for covariates and standardized) and (2) remission at endpoint. Side effect data were available only in GENDEP and STAR*D, where side effect from shared categories were analysed (gastro-intestinal, cardiovascular, CNS, sleep and sexual side effects). Presence of side effects were considered at weeks 2, 6 and 9. A fixed-effects meta-analysis was performed, all phenotypic groups were compared among each other using EM (wild type) as the reference group. Standardized mean difference (SMD) or odds ratio (OR) with 95% confidence intervals (CI) were calculated. Results Meta-analyses included 2558 patients for efficacy endpoints and 1644 patients for side effects. Imputed genotypes showed good quality. Compared to EMs, PMs had higher symptom improvement scores (SMD=0.43, CI=0.19–0.66, I2=11.5%) and higher remission rates (OR=1.55, CI=1.23–1.96, I2=0%). At week 2, PMs showed higher risk of gastro-intestinal side effects (OR=1.25, CI=1.06–1.48; week 6 values similar), of CNS side effects (OR=1.29, CI=1.06–1.58) and of sexual side effects (OR=1.56, CI=1.19–2.05; week 6 values similar) with low heterogeneity (I2 range 0% -37%). No difference was seen at week 9 for any side effect; no difference in total side effects burden was observed at any time point. Discussion This meta-analysis supports an association between CYP2C19 metabolizing level and citalopram/escitalopram efficacy and side effects. CYP2C19 polymorphisms may provide helpful information for guiding treatment in PMs.