Long-term treatment with tetrahydrobiopterin in phenylketonuria: treatment strategies and prediction of long-term responders.

Tetrahydrobiopterin (BH4) responsive phenylketonuria has been described more than 10 years ago. However, criteria for the identification of long-term BH4 responsive patients are not yet established. 116 patients with phenylketonuria, aged 4-18 years, were screened for potential long-term BH4 responsiveness by at least two of the following criteria: positive neonatal BH4 loading test, putative BH4 responsive genotype, and/or milder phenotype. Patients had to be on permanent dietary treatment. 23 patients fulfilled these criteria and were tested for long-term BH4 responsiveness: 18/23 were long-term BH4 responsive, 5/23 were not. On long-term BH4 treatment over a period of 48 ± 27 months in a dose of 14.9 ± 3.3mg/kg/day phenylalanine tolerance was increased from 452 ± 201 mg/day to 1593 ± 647 mg/day, corresponding to a mean increase of 1141 ± 528 mg/day. Dietary phenylalanine intake was increased stepwise according to a clear defined protocol. In 8/18 patients, diet was completely liberalized; 10/18 patients still received phenylalanine-free amino acid formula with 0.63 ± 0.23 g/kg/day. The most predictive value for long-term BH4 responsiveness was the combination of pretreatment phenylalanine of < 1200 μmol/L, pretreatment phenylalanine/tyrosine ratio of <15, phenylalanine/tyrosine ratio of <15 on treatment, phenylalanine tolerance of >20mg/kg/day at age 3 years, positive neonatal BH4 loading, and at least one putative BH4 responsive mutation (p = 0.00024). Our data show that long-term BH4 responsiveness may be predicted already during neonatal period by determining maximum pretreatment phenylalanine and phenylalanine/tyrosine concentrations, neonatal BH4 loading and PAH genotype. A clear defined protocol is necessary to install long-term BH4 treatment.