In phenylketonuria (PKU), there are no data available for children with respect to evaluating casein glycomacropeptide (CGMP) as an alternative to phenylalanine-free protein substitutes [Phe-free L-amino acid (AA)]. CGMP contains a residual amount of phenylalanine, which may alter blood phenylalanine control. In a prospective 6-month pilot study, we investigated the effect on blood phenylalanine control of CGMP-amino acid (CGMP-AA) protein substitute in 22 PKU subjects (13 boys, nine girls), median age (range) 11years (6-16years). Twelve received CGMP-AA and nine received Phe-free L-AA, (1 CGMP-AA withdrawal). Subjects partially or wholly replaced Phe-free L-AA with CGMP-AA. If blood phenylalanine exceeded the target range, the CGMP-AA dose was reduced and replaced with Phe-free L-amino acids. The control group remained on Phe-free L-AAs. Phenylalanine, tyrosine and Phe:Tyr ratio concentrations were compared with the results for the previous year. In the CGMP-AA group, there was a significant increase in blood phenylalanine concentrations (pre-study, 275μmolL-1 ; CGMP-AA, 317μmol L-1 ; P=0.02), a decrease in tyrosine concentrations (pre-study, 50μmolL-1 ; CGMP-AA, 40μmolL-1 ; P=0.03) and an increase in Phe: Tyr ratios (pre-study, Phe:Tyr 4.9:1; CGMP-AA, Phe:Tyr 8:1; P=0.02). In the control group there was a non-significant fall in phenylalanine concentrations (pre-study 325μmol/L: study 280μmol/L [p = 0.9], and no significant changes for tyrosine or phe/tyr ratios [p = 0.9]. Children taking the CGMP-AA found it more acceptable to L-AA. Blood phenylalanine control declined with CGMP-AA but, by titrating the dose of CGMP-AA, blood phenylalanine control remained within target range. The additional intake of phenylalanine may have contributed to the change in blood phenylalanine concentration. CGMP-AA use requires careful monitoring in children.