CYP2D6 GENOTYPES AND TREATMENT FAILURE WITH RISPERIDONE OR ARIPIPRAZOLE: RESULTS FROM A ONE-YEAR LONGITUDINAL STUDY

Abstract Background The atypical antipsychotics risperidone and aripiprazole are metabolized by the cytochrome P450 2D6 (CYP2D6), as are ~20% of currently prescribed medications. According to the enzyme level of activity, individuals can be categorized as ultrarapid, normal, intermediate, or poor (UM, NM, IM and PM, respectively) CYP2D6 metabolizers. Greater odds of switching from risperidone within one year after the last therapeutic drug monitoring were previously reported among CYP2D6 PM and UM while no differences were observed for aripiprazole between metabolizer groups (Jukic et al., 2019). Importantly, duration of treatment up to the switch was not known. Aims & Objectives We aimed to investigate treatment failure to risperidone and aripiprazole, defined as treatment duration up to switching medication, in function of CYP2D6 activity. Method Risperidone treatment duration up to switching was analyzed using Kaplan-Meier estimates with log-rank tests and Cox regression among 515 patients. Aripiprazole survival analysis followed by log- rank tests and logistic regressions were performed to assess the probability of discontinuation within the first 3, 6 and 12 months of treatment among 467 patients. CYP2D6 *3, *4, *5, *6 null activity alleles, *9, *10, *41 reduced activity alleles and *XN allele of increased activity were analyzed to classify patients as CYP2D6 UM, NM, IM, and PM. The co-prescription of drugs which are CYP2D6 strong inhibitors, transforming UM, EM and IM into PM (phenoconversion) was also taken into account. Results The overall probability of switching from risperidone was 44% after one year of treatment. Considering PM and other CYP2D6 phenotypes (i.e. IM + NM + UM) separately, 70% and 42% of switching probability was found, respectively (p=0.026). The risk of switching increased over time for PM (interaction term: 1.01; p=0.011), becoming statistically significant after three months of treatment (Hazard Ratio: 1.79 at three months, p=0.028). Similar results were obtained when considering phenoconversion, although phenoconverted PM risk of switch was constant over time (HR: 1.77, p=0.007). For aripiprazole, the incidence of treatment discontinuation from aripiprazole was significantly increased at 3 months (OR: 2.57: CI: 1.15-5.58) in CYP2D6 extreme metabolizers (UM or PM) when compared to other metabolizers (NM or IM). Similar results were obtained when considering phenoconversion. Discussion & Conclusion Over one year, CYP2D6 PM presented an increasing risk to switch from risperidone over time, the risk becoming statistically significant after the first three months of treatment. A CYP2D6 extreme phenotype (PM or UM), with or without phenoconversion, is associated with increased discontinuation rate at 3 months of aripiprazole treatment. This supports pre-emptive genotyping for CYP2D6 before risperidone and aripiprazole, taking also into account CYP2D6 strong inhibitors. References Jukic, M.M. et al. (2019) ‘Effect of CYP2D6 genotype on exposure and efficacy of risperidone and aripiprazole: a retrospective, cohort study’, The Lancet Psychiatry, 6(5), p418-426.