METHODS: This multicentre, randomised, open-label phase 3 study was done at 26 sites (primarily secondary or tertiary centres) in 12 countries. Men, boys, and young adults aged 12 years or older with severe haemophilia A or haemophilia B with inhibitors previously treated with on-demand bypassing agents were randomly assigned (2:1) to receive once-a-month 80 mg subcutaneous fitusiran prophylaxis (fitusiran prophylaxis group) or to continue with bypassing agents on-demand (bypassing agents on-demand group) for 9 months. The primary endpoint was mean annualised bleeding rate during the efficacy period in the intention-to-treat population estimated by negative binomial model. Safety was assessed as a secondary endpoint in the safety population. This trial is complete and is registered with ClinicalTrials.gov, NCT03417102.

FINDINGS: Between Feb 14, 2018, and June 23, 2021, 85 participants were screened for inclusion, of whom 57 (67%; 57 [100%] men; median age 27·0 years [IQR 19·5-33·5]) were randomly assigned: 19 (33%) participants to the bypassing agent on-demand group and 38 (67%) participants to the fitusiran prophylaxis. Negative binomial model-based mean annualised bleeding rate was significantly lower in the fitusiran prophylaxis group (1·7 [95% CI 1·0-2·7]) than in the bypassing agents on-demand group (18·1 [10·6-30·8]), corresponding to a 90·8% (95% CI 80·8-95·6) reduction in annualised bleeding rate in favour of fitusiran prophylaxis (p<0·0001). 25 (66%) participants had zero treated bleeds in the fitusiran prophylaxis group versus one (5%) in the bypassing agents on-demand group. The most frequent treatment-emergent adverse event in the fitusiran prophylaxis group was increased alanine aminotransferase in 13 (32%) of 41 participants in the safety population; there were no increased alanine aminotransferase treatment-emergent adverse events in the bypassing agents on-demand group. Suspected or confirmed thromboembolic events were reported in two (5%) participants in the fitusiran prophylaxis group. No deaths were reported.

INTERPRETATION: Subcutaneous fitusiran prophylaxis resulted in statistically significant reductions in annualised bleeding rate in participants with haemophilia A or haemophilia B with inhibitors, with two-thirds of participants having zero bleeds. Fitusiran prophylaxis might show haemostatic efficacy in participants with haemophilia A or haemophilia B with inhibitors; therefore, the therapeutic might have the potential to improve the management of people with haemophilia.