Summary Background Although treatments for Alzheimer's disease sometimes improve cognition, functional ability, or behaviour compared with baseline levels, such improvements are inconsistent across studies and measures, and effects diminish over time. More effective treatments are needed. We assessed the safety, tolerability, and efficacy of dimebon in the treatment of patients with mild-to-moderate Alzheimer's disease. Methods We enrolled 183 patients with mild-to-moderate Alzheimer's disease (mini-mental state examination MMSE scores 10–24) at 11 sites in Russia. Patients were randomly assigned by a computer-generated randomisation scheme to receive oral dimebon, 20 mg three times a day (60 mg/day n=89), or matched placebo (n=94). Other antidementia drugs were not allowed. The primary outcome measure assessed cognition, the difference in mean change from baseline to week 26, or last completed observation on the cognitive subscale of the Alzheimer's disease assessment scale (ADAS-cog). All patients and study personnel were blinded throughout the study. We compared dimebon with placebo with an intention-to-treat analysis, with last observation carried forward (ITT-LOCF) imputation. Analyses were repeated on the fully evaluable population, defined as all patients in the intention-to-treat population who had an ADAS-cog at week 26 and at least 80% compliance. 134 patients (68 in dimebon group, 66 in placebo group) enrolled in the 6-month blinded extension phase of the study. This trial is registered with Clinicaltrials.gov , number NCT00377715. Findings 155 (85%) patients completed the trial (78 88% in dimebon group, 77 82% in placebo group). Treatment with dimebon resulted in significant benefits in ADAS-cog compared with placebo (ITT-LOCF) at week 26 (mean drug-placebo difference −4·0 95% CI −5·73 to −2·28; p<0·0001). Results of the ITT-LOCF and the evaluable population analyses were much the same for all measures. Patients given dimebon were significantly improved over baseline for ADAS-cog (mean difference −1·9 −2·92 to −0·85; p=0·0005). Dimebon was well tolerated: dry mouth and depressed mood or depression were the most common adverse events associated with dimebon (12 14% patients for each symptom by week 26). The percentage of patients who had adverse events in the two groups did not differ. Interpretation Dimebon was safe, well tolerated, and significantly improved the clinical course of patients with mild-to-moderate Alzheimer's disease. Funding Medivation (USA).