The International Immune Tolerance Study was a multicenter, prospective, randomized comparison of high-dose (HD; 200 IU/kg/d) and low-dose (LD; 50 IU/kg 3 times/week) factor VIII regimens in 115 "good-risk," severe high-titer inhibitor hemophilia A subjects. Sixty-six of 115 subjects reached the defined study end points: success, n = 46 (69.7%); partial response, n = 3 (4.5%); and failure, n = 17 (25.8%). Successes did not differ between treatment arms (24 of 58 LD vs 22/57 HD, P = .909). The times taken to achieve a negative titer (P = .027), a normal recovery (P = .002), and tolerance (P = .116, nonsignificant) were shorter with the HD immune tolerance induction (ITI). Peak historical (P = .026) and on-ITI (P = .002) titers were correlated inversely with success, but only peak titer on ITI predicted outcome in a multivariate analysis (P = .002). LD subjects bled more often (odds ratio, 2.2; P = .0019). The early bleed rate/month was 0.62 (LD) and 0.28 (HD; P = .000 24), decreasing by 90% once negative titers were achieved. Bleeding was absent in 8 of 58 LD versus 21 of 57 HD subjects (P = .0085). One hundred twenty-four central catheter infections were reported in 41 subjects (19 LD); infection frequency did not differ between the treatment arms. Neither bleeding nor infection influenced outcome. Although it was stopped early for futility and safety considerations, this trial contributed valuable data toward evidence-based ITI practice.