Introduction Turoctocog alfa is a recombinant, B domain-truncated factor VIII (FVIII) approved for patients with haemophilia A. Aim To evaluate the safety and efficacy of turoctocog alfa in previously untreated patients (PUPs) with severe haemophilia A. Methods Guardian 4 was a multicentre, multinational, non-randomized, open-label phase 3 trial comprising a main and extension phase. The former concluded once >= 50 patients had received treatment for >= 50 exposure days (EDs) or developed inhibitors. Patients received turoctocog alfa intravenously for prevention and treatment of bleeds. The primary endpoint was the incidence rate of FVIII inhibitors (>= 0.6 Bethesda Units) reported during the first 50 EDs. Results Of the 58 patients who completed the main phase, 25 (43.1%) patients developed inhibitors (detected within 6-24 [mean: 14.2] EDs from treatment start). High-risk mutations were identified in 60% of patients who developed inhibitors in the main phase and were a significant predictor of inhibitor development (P = .003). Of the 21 patients who started immune tolerance induction therapy, 85.7% completed treatment with a negative inhibitor test (note that data on the last 3 patients completing ITI are based on information collated from sites prior to the final database lock). Haemostatic response (including missing values as failure) was rated as 'excellent' or 'good' for 86.1% of bleeds occurring during prophylaxis. The estimated mean annualized bleeding rate for patients on prophylaxis was 4.26 bleeds/patient/year (95% CI: 3.34 - 5.44). Conclusions Turoctocog alfa was effective at preventing and stopping bleeds and was well tolerated. Inhibitor development was within the expected range for this PUP population.