Association between CYP2D6 and CYP2C19 genotypes and venlafaxine treatment failure: a retrospective study on two cohorts from Norway and the UK

Abstract

BACKGROUND: Venlafaxine shows superior efficacy over SSRIs in treating major depression. Metabolism of venlafaxine is mediated by CYP2D6 with a secondary role of CYP2C19. The activity of both enzymes is determined by pharmacogenetic variability, which might affect therapeutic response to venlafaxine. We aimed to investigate associations between CYP2D6 and CYP2C19 genotypes and venlafaxine treatment failure. METHODS: This retrospective, observational study included two independent, naturalistic cohorts of individuals treated with venlafaxine from the Center for Psychopharmacology (Oslo, Norway) and the UK Biobank. Participants were excluded if they were comedicated with strong CYP2D6 or CYP2C19 inhibitors (ie, bupropion, fluoxetine, paroxetine, fluvoxamine, or levomepromazine) or CYP inducers (ie, carbamazepine, phenobarbital, or phenytoin) at their last venlafaxine therapeutic drug monitoring measurement. Participants were stratified into genotype-predicted poor metabolisers (PMs), intermediate metabolisers (IMs), normal metabolisers (NMs), and ultra-rapid metabolisers (UMs). The primary outcome was treatment failure defined as switching away from venlafaxine to another antidepressant within 1 year after last exposure. People with lived experience did not contribute to the design and writing of this study. FINDINGS: Between Jan 1, 2005, and Dec 15, 2025, 5443 participants genotyped for CYP2D6 and CYP2C19 (3315 [60·9%] women, 2128 [39·1%] men; mean age 48·9 years [SD 18·6]) were identified from the Center for Psychopharmacology. 3624 participants (2401 [66·3%] women, 1223 [33·7%] men; mean age 56·5 years [SD 9·8]) with prescription data were identified from the UK Biobank between 2006 and 2010. In the Norwegian cohort, odds of switching were 2·05 times higher in CYP2D6 PMs (95% CI 1·53-2·71, p<0·0001), 1·25 times higher in IMs (1·05-1·50, p=0·014), and 1·78 times higher in UMs (1·06-2·83, p=0·021) compared with NMs. Increased odds of switching in CYP2D6 PMs were also shown in the UK Biobank cohort (odds ratio 1·58 [95% CI 1·14-2·19], p=0·0060). CYP2C19 PM status amplified risk of failure in the Norwegian cohort with 6·11 times higher odds of switching in dual PMs versus NMs (95% CI 1·81-18·77, p=0·0019). INTERPRETATION: CYP2D6 PM status was consistently associated with increased risk of venlafaxine treatment failure in two large, real-life cohorts, suggesting that pre-emptive genotyping could facilitate personalised venlafaxine therapy. FUNDING: European Research Council, Swedish Research Council, Novo Nordisk Foundation, Robert Bosch Foundation, National Research Foundation of Korea, and the South-Eastern Norway Regional Health Authority.

The authors' abstract, as published at the source. The Lancet Psychiatry, 2026 · DOI ↗