Efficacy and safety of the novel glycine transporter inhibitor BI 425809 once daily in patients with schizophrenia: a double-blind, randomised, placebo-controlled phase 2 study

Lancet Psychiatry. 2021 Mar;8(3):191-201. doi: 10.1016/S2215-0366(20)30513-7.


Background: Cognitive impairment associated with schizophrenia predicts poor functional outcomes, but currently no approved pharmacotherapy is available. This study investigated whether the glycine transporter-1 inhibitor BI 425809 improves cognition in patients with schizophrenia.

Methods: This phase 2, randomised, double-blind, placebo-controlled, parallel-group trial (81 centres, 11 countries), randomly assigned outpatients (aged 18-50 years) with schizophrenia on stable treatment to add-on once-daily oral BI 425809 2 mg, 5 mg, 10 mg, or 25 mg or placebo (1:1:1:1:2) for 12 weeks. Treatment was assigned in blocks using interactive response technology; patients, investigators, and all trial personnel were masked to group assignment. The primary endpoint was change from baseline in MATRICS Consensus Cognitive Battery (MCCB) overall composite T-score at week 12. Six predefined dose-response models were evaluated using a multiple comparison procedure and modelling approach with mixed model repeated measures to assess evidence for a non-flat dose-response relationship for cognitive improvements with BI 425809. Adverse events were monitored. Safety analyses included all randomly allocated patients who received one or more doses of trial medication; efficacy analyses included patients from this set who also had available baseline data and at least one post-baseline on-treatment measurement for the primary or secondary endpoint. This study is registered with ClinicalTrials.gov, number NCT02832037.

Findings: 509 patients were randomly assigned between April 25, 2018, and Oct 4, 2019 (BI 425809 2 mg, n=85; 5 mg, n=84; 10 mg, n=85; 25 mg, n=85; placebo, n=170 444 (87%) completed the 12-week treatment. Five of six dose-response models showed a statistically significant benefit of BI 425809 over placebo (linear [t=2·55, p=0·015], linear in log [t=2·56, p=0·015]; Emax [t=2·75, p=0·0089], sigmoid Emax [t=2·98, p=0·0038], logistic [t=2·77, p=0·0085]). Pairwise comparisons showed greater mean improvement from baseline in MCCB overall composite T-score at week 12 with BI 425809 10 mg and 25 mg versus placebo (adjusted mean difference 1·98 [95% CI 0·43-3·53] for 10 mg and 1·73 [0·18-3·28] for 25 mg; standardised effect size 0·34 for 10 mg and 0·30 for 25 mg). Adverse events were balanced across groups, reported in 50 (59%) of 85 patients on BI 425809 2 mg, 44 (52%) of 84 on 5 mg, 35 (41%) of 85 on 10 mg, 36 (42%) of 85 on 25 mg, and 74 (44%) of 170 on placebo.

Interpretation: BI 425809 improved cognition after 12 weeks in patients with schizophrenia; doses of 10 mg and 25 mg showed the largest separation from placebo. If these encouraging results are confirmed in phase 3 trials, BI 425809 could provide an effective treatment for cognitive impairment associated with schizophrenia.

Funding: Boehringer Ingelheim.

Publication types

  • Clinical Trial, Phase II
  • Multicenter Study
  • Randomized Controlled Trial
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Adult
  • Cognition / drug effects*
  • Cognitive Dysfunction / drug therapy
  • Dose-Response Relationship, Drug
  • Double-Blind Method
  • Female
  • Glycine Plasma Membrane Transport Proteins / antagonists & inhibitors*
  • Humans
  • Likelihood Functions
  • Male
  • Middle Aged
  • Organic Chemicals / administration & dosage*
  • Organic Chemicals / adverse effects
  • Schizophrenia / drug therapy*
  • Treatment Outcome
  • Young Adult


  • BI 425809
  • Glycine Plasma Membrane Transport Proteins
  • Organic Chemicals

Associated data

  • ClinicalTrials.gov/NCT02832037