Study Design

Study design and characteristics of the three short-term, non-randomized trials are summarized in Table 14.

Table 14

Summary Design and Characters of Short-Term, Non-Randomized Studies.

Patients who completed the switch-over phase of studies HPN-100-005 and HPN-100-012 were offered the opportunity to continue in the safety-extension phase to receive open-label GPB for up to 12 months. Results of the safety-extension phase are presented in APPENDIX 7. Because of the small sample size, no adjustment for covariates was considered.

UP 1204-003: All efficacy analyses were performed using the intention-to-treat (ITT) population, which consisted of all patients who received any amount of both study drugs. For the primary analysis of ammonia levels using time-normalized area under the curve (TNAUC), patients with missing TNAUC at visit 2-1 (steady-state NaPBA) or with missing TNAUC at visit 11-1 (steady-state GPB) were excluded. The last-observation-carried-forward method was planned for imputation of data, but was not used (all 10 patients had available data to calculate TNAUC). Peak venous ammonia was also imputed using the last-observation-carried-forward method if there was at least one post-first GPB dose peak venous ammonia level. A formal testing of hypotheses was not performed for this study. The study was not powered to demonstrate equivalence between GPB and NaPBA. Subgroup analysis was not conducted due to the small sample size.

HPN-100-005: Descriptive statistics were used to summarize study data. An analysis of variance (ANOVA) model was constructed with factors for treatment and patient to assess the noninferiority of GPB to NaPBA in ammonia control (as blood ammonia AUC_0–24 levels) in the ITT and per-protocol (PP) populations. Noninferiority was concluded when the upper bound of the 90% confidence interval (CI) was less than and equal to 1.25. Missing ammonia data were imputed for patients who did not have a calculable venous ammonia AUC_0–24 value. The relative effect of GPB and NaPBA on glutamine levels was evaluated post hoc. A post-hoc evaluation of blood ammonia AUC_0–24 and C_max levels was conducted in children (aged six years to 11 years) and adolescents (aged 12 years to 17 years).

HPN-100-012: The study was not powered to detect a statistically significant difference between GPB and NaPBA. Given the small number of patients, all hypothesis tests were interpreted as exploratory in nature. An ANOVA model was constructed with factors for treatment and patient to assess the noninferiority of GPB to NaPBA in ammonia control (as blood ammonia AUC_0–24 levels) in the ITT and PP populations. Noninferiority was concluded when the upper bound of the 90% CI was less than and equal to 1.25.

Disposition

The disposition of patients across the three switch-over studies (UP 1204-003, HPN-100-005, and HPN-100-012) is summarized in Table 15.

Table 15

Disposition of Patients in Studies UP1204-003, HPN-100-005, and HPN-100-012.

Results

The main demographic and baseline characteristics of patients were similar across treatment arms within studies, but were variable across studies. Patient characteristics are summarized in Table 16.

Table 16

Demographics and Main Baseline Characteristics in Studies UP1204-003, HPN-100-005, and HPN-100-012 (Intention-to-Treat Population).

In UP 1204-003, all patients reported high compliance with the study medication (93% to 100% of planned doses were taken). Four patients reported missing doses and no patient reported missing more than two doses of study drug. Compliance with the dietary regimen was variable and ranged from 1% of prescribed protein/kg per day to 215% of prescribed protein/kg per day, and from 3% of calories/kg per day to 127% of calories/kg per day.

In HPN-100-005, among the 11 patients enrolled in the switch-over phase, 10 were 100% compliant with their prescribed NaPBA treatment and nine were 100% compliant with their GPB treatment.

Details of treatment compliance are presented in Table 17.

Table 17

Compliance to Study Drugs During Treatment (%).

A summary of the results of blood ammonia levels after treatment with GPB and NaPBA in the adult or pediatric patients with UCDs is presented in Table 18.

Table 18

Change in Blood Ammonia and Glutamine Levels in Studies UP 1204-003, HPN-100-005, and HPN-100-012 (Intention-To-Treat Population).

In UP 1204-003, the TNAUC for blood ammonia after seven days of treatment with NaPBA (at steady state) was higher (mean value was 38.40 μmol/L) compared with after seven days of treatment with GPB (26.5 μmol/L). The overall difference in ammonia values (TNAUC) measured after NaPBA treatment and after GPB treatment did not reach statistical significance. The mean peak ammonia level was higher with steady-state NaPBA (79.14 μmol/L) compared with the mean peak ammonia level with GPB treatment (56.31 μmol/L) at steady state. Glutamine decreased after switch to GPB in eight patients who had values at both visits 2-1 and 11-1. On average, there was a mean decrease from visit 2-1 to visit 11-1 (−87 μmol/dL) in these eight patients.

In HPN-100-005, the difference in blood ammonia AUC_0–24 levels was not statistically different between the GPB and NaPBA treatments (P = 0.1028). However, in the PP population, blood ammonia AUC_0–24 levels were statistically lower with GPB treatment than with NaPBA treatment (P = 0.0304). In the ITT and PP populations, blood ammonia AUC_0–24 levels with GPB treatment compared with NaPBA treatment met the predefined noninferiority criteria. There were no statistically significant differences between the treatments in mean C_max concentrations in either the ITT (P = 0.2481) or PP (P = 0.1441) populations.

In HPN-100-012, blood ammonia, assessed as mean AUC_0–24, was lower after GPB treatment than after NaPBA treatment (mean between-group difference −237.46 μmol·h/L, P > 0.05). C_max for ammonia was lower after GPB treatment compared with after NaPBA treatment.

Safety of the study drugs was evaluated in these three studies. A summary of the adverse events (AEs) during the studies is presented in Table 19. There were no deaths reported in the three studies. In general, the risks of AEs were similar between GPB and NaPBA.

Table 19

Adverse Events Reported in Studies UP 1204-003, HPN-100-005, and HPN-100-012 (Safety Population).

In UP1204-003, the occurrence of AEs was similar between 100% NaPBA treatment and 100% GPB treatment. The most frequently reported (AEs) were gastrointestinal (GI) disorders (i.e., nausea, dyspepsia, and abdominal pain), which were reported in 35.7% of patients during the study, including 21.4% during 100% NaPBA and 20% during 100% GPB. Two patients had serious adverse events (SAEs), both during NaPBA treatment, one before and one after the GPB treatment period. There were no deaths during the study.

In HPN-100-005, all but one of the AEs were considered mild, and for one patient the investigator considered the event (vomiting while on GPB) to be moderate-intensity.

In HPN-100-012, six patients reported AEs during the GPB treatment period. All AEs were mild. There were no SAEs or discontinuations due to AEs.