Real-world effectiveness of clozapine augmentation with antidepressants: within-subject data evidence from two European nationwide cohorts

• Published in: The Lancet. Psychiatry Vol. 12; no. 8; p. 568
• Main Authors: Taipale, Heidi, Siskind, Dan, Tanskanen, Antti, Mittendorfer-Rutz, Ellenor, Correll, Christoph U, Tiihonen, Jari
• Format: Journal Article
• Language: English
• Published: England 01.08.2025
• Subjects: Adult; Antidepressive Agents - administration & dosage; Antidepressive Agents - adverse effects; Antidepressive Agents - therapeutic use; Antipsychotic Agents - administration & dosage; Antipsychotic Agents - therapeutic use; Clozapine - administration & dosage; Clozapine - adverse effects; Clozapine - therapeutic use; Cohort Studies; Drug Therapy, Combination; Female; Finland; Humans; Male; Middle Aged; Recurrence; Schizophrenia - drug therapy; Schizophrenia, Treatment-Resistant - drug therapy; Sweden; Treatment Outcome; Sweden; Finland
• ISSN: 2215-0374
• DOI: 10.1016/S2215-0366(25)00163-4

Only 40% of people with treatment-resistant schizophrenia will have an adequate response to clozapine. Data on the effect of antidepressant augmentation on these patients are scarce. We aimed to study the effectiveness and safety of antidepressant augmentations to clozapine in two nationwide cohorts. For this within-subject data evidence study, data from two cohort studies were analysed then meta-analysed, including patients with schizophrenia using clozapine identified from Finnish (1996-2017) and Swedish (2006-23) nationwide registers, followed up from first clozapine use until death or end of data linkage. Clozapine augmentation with specific antidepressants was compared with clozapine use only. The primary outcome was schizophrenia relapse (hospital admission with ICD-10 F20-F29 diagnosis), and the secondary outcome was somatic hospital admission. The risk of outcomes (expressed as adjusted hazard ratio [aHR]) was assessed by a within-individual cohort design, using each individual as their own control. Ethnicity data were not available. People with lived experience were not involved. The analysis included 23 206 clozapine users (9531 women and 13 675 men), with a mean age of 41·3 years (SD 13·5). The lowest risk of relapse was observed for sertraline (meta-analysis aHR 0·76 [95% CI 0·69-0·83]), duloxetine (aHR 0·78 [0·68-0·89]), and escitalopram (0·85 [0·79-0·92]). When antidepressants were recategorised by dose, the lowest risk of relapse was observed for standard doses of sertraline at 30-54 mg/day (aHR 0·49 [0·27-0·89]), escitalopram 6-10 mg/day (aHR 0·57 [0·37-0·88]), and duloxetine 18-32 mg/day (aHR 0·59 [0·46-0·75]). No antidepressant was associated with an increased risk of somatic hospital admission at low or standard dose, whereas high-dose use was associated with increased risk, although most of the risk estimates were not significant. This meta-analysis from two nationwide cohorts indicates that SSRI and SNRI augmentation therapies for people with treatment-resistant schizophrenia are associated with a substantially decreased risk of relapse at standard dose, unlike other antidepressants. Potential benefits associated with augmentation must be balanced against risks of possible deleterious somatic outcomes, which were observed only when antidepressants were used at high doses. Sigrid Jusélius Foundation.