Bone mineral density in participants with phenylalanine hydroxylase (PAH) deficiency: a report from the PHEFREE rare disorders consortium.

Low bone mineral density (BMD) in individuals with phenylketonuria (PKU) due to phenylalanine hydroxylase (PAH) deficiency has been variably reported in the medical literature, but the clinical significance of this observation has been uncertain. We measured BMD in a cross-sectional cohort of 117 individuals with PAH deficiency between 6.2 and 56.8 years age who participate in the PHEFREE Consortium Longitudinal Natural History Study. 49% of children under age 18 years and 26% of adults were consuming a phenylalanine (Phe)-restricted diet, supplemented with low Phe or Phe-free medical foods and treated with sapropterin dihydrochloride. 22% of adults and 32% of children were not responsive to sapropterin and were treated with dietary Phe restriction and supplementary medical foods alone. 20 adults (37%) reported receiving enzyme replacement therapy with pegvaliase. Phenotype predictions from the genotypes of the participants in these three treatment groups were predominantly classical PKU. Participants with mild PKU or mild hyperphenylalaninemia (HPA) phenotypes treatable with sapropterin alone without the need for medical foods made up 9% of adults and 14% of children in the study cohort. Mean (± SD) blood phenylalanine (Phe) concentrations over the year prior to bone densitometry studies measured 449 ± 336 μM in the entire study cohort. Mean BMD z-scores for both adults and children in all body regions were negative with the difference between study participants and the comparator non-PKU population reaching statistical significance (p < 0.05). At least one z-score was <-2 in 15% of children and 10% of adults with PKU (2.3% expected). These differences were apparent in some children already at 6 years age, the earliest time point measured. Mean BMD was more negative in individuals with history of fracture than in those without, but this difference did not reach statistical significance. Multivariate analysis revealed a positive correlation between BMD and body mass index (BMI) in children, with minor contributions from age and natural protein intake. There was no correlation between BMD and blood Phe concentration measured in the year prior to BMD measurement. The pathophysiologic mechanism causing and the clinical significance of these differences remain uncertain, but concern persists that low BMD in this relatively well-treated cohort could portend frank osteopenia and increased fracture risk as these individuals reach advanced age. Whether novel therapeutics, that both correct hyperphenylalaninemia and allow increased dietary intact protein intake, will impact bone mineralization is yet unknown. Longitudinal monitoring of BMD throughout the lifespan is warranted.