Urinary Phenylacetylglutamine as Dosing Biomarker for Patients with Urea Cycle Disorders

Abstract We have analyzed pharmacokinetic data for glycerol phenylbutyrate(also GT4P or HPN-100) and sodium phenylbutyratewith respect to possible dosing biomarkers in patients with urea cycle disorders(UCD). Study design These analyses are based on over 3000 urine and plasma data points from 54 adult and 11 pediatric UCD patients (ages 6–17) who participated in three clinical studies comparing ammonia control and pharmacokinetics during steady state treatment with glycerol phenylbutyrateor sodium phenylbutyrate. All patients received phenylbutyricacid equivalent dosesof glycerol phenylbutyrateor sodium phenylbutyratein a cross over fashion and underwent 24-hour blood samples and urine sampling for phenylbutyricacid, phenylacetic acidand phenylacetylglutamine. Results Patients received phenylbutyricacid equivalent dosesof glycerol phenylbutyrateranging from 1.5 to 31.8 g/day and of sodium phenylbutyrateranging from 1.3 to 31.7 g/day. Plasma metabolite levels varied widely, with average fluctuation indices ranging from 1979% to 5690% for phenylbutyricacid, 843% to 3931% for phenylacetic acid, and 881% to 1434% for phenylacetylglutamine. Mean percent recovery of phenylbutyricacid as urinary phenylacetylglutaminewas 66.4 and 69.0 for pediatric patients and 68.7 and 71.4 for adult patients on glycerol phenylbutyrateand sodium phenylbutyrate, respectively. The correlation with dose was strongest for urinary phenylacetylglutamineexcretion, either as morning spot urine (r = 0.730, p 24-hour , plasma phenylacetic acidAUC 24-hour and phenylbutyricacid AUC 24-hour . Plasma phenylacetic acidlevels in adult and pediatric patients did not show a consistent relationship with either urinary phenylacetylglutamineor ammonia control. Conclusion The findings are collectively consistent with substantial yet variable pre-systemic (1st pass) conversion of phenylbutyricacid to phenylacetic acidand/or phenylacetylglutamine. The variability of blood metabolite levels during the day, their weaker correlation with dose, the need for multiple blood samples to capture trough and peak, and the inconsistency between phenylacetic acidand urinary phenylacetylglutamineas a marker of waste nitrogen scavenging limit the utility of plasma levels for therapeutic monitoring. By contrast, 24-hour urinary phenylacetylglutamineand morning spot urine phenylacetylglutaminecorrelate strongly with dose and appear to be clinically useful non-invasive biomarkers for compliance and therapeutic monitoring.