LEOPOLD II 2015.

Study characteristics
Methods	Multicentre RCT. Open label. Cross‐over trial (see 'Notes'). Trial period:12 months. Conducted at 30 centres in 11 countries in Europe, South Africa, North America, South America, and Asia.
Participants	Males aged 12 – 65 years with severe haemophilia A who had not received regular prophylaxis treatment for > 6 consecutive months in the previous 5 years. Number randomised: 83; number included in the analysis: 80.
Interventions	1. Twice‐weekly prophylaxis (20 – 30 IU/kg). 2. Thrice‐weekly prophylaxis (30 – 40 IU kg). 3. On‐demand treatment with BAY 81‐8973: a recombinant factor VIII product. Participants were randomised to 1 of 6 treatment arms (2 low‐dose prophylaxis groups, 2 high‐dose prophylaxis groups, and 2 on‐demand treatment groups; participants received treatment based on CS/EP or CS/ADJ for 6 months each with an intra‐individual cross‐over after 6 months. Follow‐up duration: 3 to 8 weeks of screening and 52 weeks of follow‐up on either high‐ or low‐dose prophylaxis.
Outcomes	Annualised number of all bleeding events and adverse events.
Notes	ClinicalTrials.gov identifier: NCT01233258. This is not a traditional cross‐over trial. Participants were randomised to receive on demand or prophylactic therapy with FVIII (2 different regimen); patients were crossed‐over within their treatment groups but only with respect to the methods for measuring the content of FVIII in the vials (CS/EP or CS/ADJ) therefore this study has been treated as a parallel trial for the analysis, "Study drug was labeled using the chromogenic substrate assay per European Pharmacopoeia (CS/EP) or adjusted by a predefined factor to mimic results obtained with the one‐stage assay (CS/ADJ). Because of differences in the detection of FVIII activity between the two potency assays, the difference in the actual amount of FVIII received for prophylaxis injections in the CS/EP and CS/ADJ periods was ~20–25%, with higher amounts received during the CS/ADJ period. Patients received treatment based on CS/EP or CS/ADJ for 6 months each with an intraindividual crossover after 6 months (Fig. 1)." Date of trial: Between January 2011 and December 2012. Source of funding: Bayer HealthCare AG, Leverkusen, Germany.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	"...system generated by the sponsor’s randomization management."
Allocation concealment (selection bias)	Low risk	"Patient assignment was performed using a centralized telephone interactive voice response system or interactive web response system".
Blinding (performance bias and detection bias) Clinical Joint Function	Unclear risk	Not assessed.
Blinding (performance bias and detection bias) Bleeding	High risk	An open‐label trial where participants and outcome assessors were aware of the allocation.
Blinding (performance bias and detection bias) Radiologic Joint Score	Unclear risk	Not assessed.
Blinding of outcome assessment (detection bias) Bleeding	High risk	An open‐label trial. It was unclear how the primary end‐point of bleeding events was assessed.
Blinding of outcome assessment (detection bias) Clinical Joint Function	Unclear risk	Not assessed.
Blinding of outcome assessment (detection bias) Radiologic Joint Score	Unclear risk	Not assessed.
Incomplete outcome data (attrition bias) All outcomes	Low risk	There were 3 participants who were randomised but did not complete the study. Reasons for dropout were given for all of these participants.
Selective reporting (reporting bias)	Low risk	All outcomes reported in the protocol were explored in the final study report.
Other bias	Low risk	3 of the trial authors are employees of the funding body, Bayer Healthcare AG.