The paper gives data on the clinical efficiency and safety profile of long-term use of the infliximab (INF) biosimilar BCD-055 versus the reference drug Remicade® (REM) in a population of patients with active ankylosing spondylitis (AS).Subjects and methods. An international multicenter randomized double-blind Phase III clinical trial was conducted in 199 patients who were randomized into two groups in a 2:1 ratio and who received BCD-055 or REM at a dose of 5 mg/kg at 0, 2, and 6 weeks, then every 8 weeks. Efficiency assessment was made at 14, 30 and 54 weeks in patients who received at least one dose of INF [intent-to-treat (ITT)], as well as at 54 weeks in those who completed the study according to the Protocol (PP) (per protocol). The efficiency endpoints were the proportion of patients who had achieved ASAS20/ASAS40 responses; changes in BASDAI, BASMI, BASFI, MASES, and SF-36 scores. Immunogenicity was assessed by the proportion of patients in each group with identified binding and neutralizing antibodies (BAbs and NAbs) against INF. The safety analysis included the overall rate of adverse events (AEs), including those that met the respective criteria for serous AEs (SAEs), and grade 3–4 toxicity, as well as the number of cases of early termination of the study because of AEs and SAEs.Results and discussion. The ITT population included 199 patients and the PP one consisted of 161 people. The groups were not statistically different in the rate of and reasons for patient withdrawal from the study. A comparable number of patients achieved ASAS20/ASAS40 responses at 14, 30, 54 weeks (р ≥ 0.05). At 54 week, the proportion of patients who received BCD-055 and REM therapy and achieved an ASAS20 response was 67.42 and 52.24% in the ITT population (p = 0.053) and 80.91 and 68.63% in the PP population (p = 0.128). The BCD-055 and drug comparison groups achieved an ASAS40 response in 53.03 and 38.81% in the ITT population (p = 0.081) and in 63.64 and 50.98% in the PP one (p = 0.177). The proportion of persons with identified BAbs and NAbs was comparable: 21.26 and 3.15% in the BCD-055 group (p = 0.920) and 20.63 and 6.35% in the group REM (p = 0.443), respectively. It was found that the presence of NAbs did not affect the therapeutic response. Both groups did not differ in the detection rate and profile of AEs and SAEs or in the rate of patient withdrawal due to AEs. Most identified AEs were mild to moderate.Conclusion. The efficacy of the INF biosimilar BCD-055 used long in patients with AS did not significantly differ from that of the original drug REM; the safety profile of both drugs was comparable.