Levy 2007.

Methods	Randomized, double‐blind placebo‐controlled trial. Multicentre, North America and Europe.
Participants	89 children and adults with PKU, over 8 years of age, with blood phe ≥450μmol/L; individuals who had a reduction of 30% or more in blood phe concentration after 8 days of treatment with sapropterin at a dose of 10 mg/kg in a previous screening test (PKU 001) were eligible for the trial. Participants had been involved in a phase 1 screening study.
Interventions	Sapropterin 10 mg/kg/day versus placebo; treatment was for a period of 6 weeks.
Outcomes	Change in blood phe concentration.
Notes	Data published only as a six‐week trial; further details requested from the author.
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Treatment allocations were made centrally from randomisation lists generated by a computer program. Each randomisation list started with a block of two, followed by blocks of four.
Allocation concealment (selection bias)	Low risk	Described that a central interactive voice response system was used; investigators, participants and sponsors were kept unaware of treatment allocation until database was locked; block size was not divulged to sponsors or investigators until the trial was completed.
Blinding (performance bias and detection bias) All outcomes	Low risk	Stated as double blind; sapropterin and placebo tablets were identical in taste and appearance.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Two withdrawals (one from each group) described. One withdrawal from the trial in the sapropterin group before closing due to an inability to comply with the trial course. The data were not included in the analysis as the participant did not receive even one dose of trial drug. There was another withdrawal from the control group due to non‐compliance with specified dosing, but the data was included in the analysis.
Selective reporting (reporting bias)	High risk	Comparison of the trial protocol available at the ClinicalTrials.gov web site and also the 'Methods' section with the results reported in the final paper showed all the detailed outcomes in the protocol were reported in the published reports. However, in the Levy trial they measured data at several time points (weeks 0, 1, 2, 4,and 6) but only reported them at 6 weeks.
Other bias	Unclear risk	Sponsored by BioMarin Pharmaceutical Inc.