- Email: [email protected]
Pharmacokinetics of glycerol phenylbutyrate in pediatric patients 2 months to 2 years of age with urea cycle disorders
Accepted Manuscript Pharmacokinetics of glycerol phenylbutyrate in pediatric patients 2 months to 2 years of age with urea cycle disorders
S.A. Berry, J. Vockley, A.A. Vinks, M. Dong, G.A. Diaz, S.E. McCandless, W.E. Smith, C.O. Harding, R. Zori, C. Ficicioglu, U. Lichter-Konecki, R. Perdok, B. Robinson, R.J. Holt, N. Longo PII: DOI: Reference:
S1096-7192(18)30379-2 doi:10.1016/j.ymgme.2018.09.001 YMGME 6402
To appear in:
Molecular Genetics and Metabolism
Received date: Revised date: Accepted date:
23 June 2018 27 July 2018 2 September 2018
Please cite this article as: S.A. Berry, J. Vockley, A.A. Vinks, M. Dong, G.A. Diaz, S.E. McCandless, W.E. Smith, C.O. Harding, R. Zori, C. Ficicioglu, U. Lichter-Konecki, R. Perdok, B. Robinson, R.J. Holt, N. Longo , Pharmacokinetics of glycerol phenylbutyrate in pediatric patients 2 months to 2 years of age with urea cycle disorders. Ymgme (2018), doi:10.1016/j.ymgme.2018.09.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Title: Pharmacokinetics of glycerol phenylbutyrate in pediatric patients 2 months to 2 years of age with Urea Cycle Disorders Authors: Berry SAa, Vockley Jb , Vinks AAc, Dong, Mc, Diaz GAd , McCandless SEe, Smith WEf, Harding COg , Zori Rh , Ficicioglu Ci , Lichter-Konecki Uj , Perdok Rk , Robinson Bk , Holt RJk,l , Longo Nm.
T
a
University of Minnesota Department of Pediatrics, Minneapolis, MN, USA University of Pittsburgh School of Medicine and the Children’s Hospital of Pittsburgh, Pittsburgh, PA, USA c Division of Clinical Pharmacology, Cincinnati Children’s Hospital Medical Center and Department of Pediatrics, University of Cincinnati, Cincinnati, OH, USA d Icahn School of Medicine at Mount Sinai, New York, NY, USA e University of Colorado Denver School of Medicine and Children’s Hospital Colorado, Aurora, CO, USA f Maine Medical Center, Portland, ME, USA g Oregon Health & Science University, Portland, OR, USA h University of Florida, Gainesville, FL, USA i Children’s Hospital of Philadelphia, Philadelphia, PA, USA j Children’s Hospital of Pittsburgh, Pittsburgh, PA, USA k Horizon Pharma USA, Inc, Lake Forest, IL, USA l University of Illinois-Chicago, Chicago, IL USA m University of Utah, Salt Lake City, UT, USA
M
AN
US
CR
IP
b
ED
Abstract
CE
PT
Introduction: Glycerol phenylbutyrate (GPB) is approved in the US and EU for the chronic management of patients ≥2 months of age with urea cycle disorders (UCDs) who cannot be managed by dietary protein restriction and/or amino acid supplementation alone. GPB is a pre-prodrug, hydrolyzed by lipases to phenylbutyric acid (PBA) that upon absorption is beta-oxidized to the active nitrogen scavenger phenylacetic acid (PAA), which is conjugated to glutamine (PAGN) and excreted as urinary PAGN (UPAGN). Pharmacokinetics (PK) of GPB were examined to see if hydrolysis is impaired in very young patients who may lack lipase activity.
AC
Methods: Patients 2 months to <2 years of age with UCDs from two open label studies (n=17, median age 10 months) predominantly on stable doses of nitrogen scavengers (n = 14) were switched to GPB. Primary assessments included traditional plasma PK analyses of PBA, PAA, and PAGN, using noncompartmental methods with WinNonlin™. UPAGN was collected periodically throughout the study up to 12 months. Results: PBA, PAA and PAGN rapidly appeared in plasma after GPB dosing, demonstrating evidence of GPB cleavage with subsequent PBA absorption. Median concentrations of PBA, PAA and PAGN did not increase over time and were similar to or lower than the values observed in older UCD patients. The median PAA/PAGN ratio was well below one over time, demonstrating that conjugation of PAA with glutamine to form PAGN did not reach saturation. Covariate analyses indicated that age did not influence the PK parameters, with body surface area (BSA) being the most significant covariate, reinforcing current BSA based dosing recommendations as seen in older patients.
ACCEPTED MANUSCRIPT Conclusion: These observations demonstrate that UCD patients aged 2 months to <2 years have sufficient lipase activity to adequately convert the pre-prodrug GPB to PBA. PBA is then converted to its active moiety (PAA) providing successful nitrogen scavenging even in very young children. Keywords: urea cycle disorders, glycerol phenylbutyrate, infants, children, pharmacokinetics 1.
Introduction
Glycerol phenylbutyrate (GPB) is approved in the US and EU for the management of patients ≥2 months
T
of age with urea cycle disorders (UCDs) who cannot be managed by dietary protein restriction and/or
IP
amino acid supplementation alone. GPB is hydrolyzed by pancreatic lipases to yield glycerol and phenylbutyric acid (PBA), which undergoes β-oxidation in the liver to phenylacetic acid (PAA). PAA is
CR
conjugated to glutamine in the liver and the kidney by glutamine-N-phenylacetyltransferase to form phenylacetylglutamine (PAGN), which is subsequently excreted in urine (UPAGN). (Figure 1) On a
US
molar basis, PAGN, like urea, contains 2 moles of nitrogen and provides an alternate mechanism for
AC
CE
PT
ED
M
AN
waste nitrogen excretion [1].
Figure 1. Metabolizing pathway and mechanism of action of GPB GPB (glycerol phenylbutyrate); PAA (phenylacetic acid); PBA (phenylbutyric acid); PAGN (phenylacetylglutamine)
ACCEPTED MANUSCRIPT Clinical trials in adults and children > 2 years of age indicate that GPB is at least as effective as sodium phenylbutyrate (NaPBA) without the associated bad taste, odor, sodium content, and burden of high pill number administration [2-6]. In addition, analysis of pooled data from these trials suggest that GPB might confer better control of ammonia due to slower absorption and reduced fluctuations in plasma metabolite concentrations, as demonstrated by previous pharmacokinetic (PK) modeling [7, 8]. In this modeling, body size was found to significantly affect clearance, volume, and presystemic conversion,
T
resulting in smaller individuals with smaller body surface area (BSA) having smaller PK values for these
IP
parameters [7]. The lower clearance for both PBA and PAGN coupled with the saturable conversion of PAA to PAGN may result in greater PAA exposure in younger patients. Greater PAA exposure in
CR
smaller patients was more pronounced with NaPBA than with GPB, despite equivalent dosing; PAA levels reached steady state in approximately 3 days without further accumulation [7]. Notably, the median
US
PAA levels were well below 500 µg/ml (the theoretical toxic limit based on studies in cancer patients using intravenous PAA) [9, 10]. In addition, neurotoxic adverse events have not been associated with
AN
either NaPBA or GPB [3].
Recently published data extended and confirmed the safety and efficacy of GPB in children 2 months to
M
<2 years of age [11]. However, limited data exist on the PK of GPB (or NaPBA) in this patient population. Here we examine the short- and long-term PK of GPB in UCD patients 2 months to <2 years
ED
of age whose pancreatic and liver function are immature to determine if there are notable differences compared to that previously reported in older UCD patients. Materials and Methods
2.1
Study design and treatments
PT
2.
CE
The PK of seventeen subjects 2 months to 2 years of age were evaluated from two clinical trials (HPN100-012 switch over and extension study and HPN100-009). Studies HPN100-012 and HPN-100-
AC
009 will be referred to as Study 1 and 2, respectively. The study design, patient disposition, and treatments have been previously described [2, 11]. As in previous studies, patients on stable doses of NaPBA were switched to an equimolar dose of GPB, however, if they were naïve to phenylbutyrate their initial doses were based on BSA (8.5 mL/m2 /day) [11].
For those receiving GPB orally, it was
administered via an oral syringe, just prior to breastfeeding or intake of formula or food. If needed, GPB was added to a small amount of formula in a small syringe and administered orally. A G tube was utilized in patients who were unable to tolerate oral dosing. 2.1.1
Study 1 (NCT01347073)
ACCEPTED MANUSCRIPT The switch-over phase of Study 1 was designed as a fixed-sequence, open-label, switch to GPB from chronic treatment with NaPBA. On Day 1, subjects received their prescribed dose of NaPBA with feedings. Subjects were observed in a monitored clinical setting for 24 hours while undergoing 24-hour blood sampling on their prescribed dose of NaPBA (Day 1) and then all subjects were switched from NaPBA to GPB in a single transition step. Subjects returned to the clinic after four to ten days on GPB for an additional visit (day 10) and 24-hour blood sampling, after which they continued in the long-term
T
extension phase. During the switch-over phase, plasma PK samples were collected at pre-dose and post-
IP
dose at 8, 12 and 24 hours and urine samples for PK analysis were collected on day 1 and day 10 at the two time periods relative to the first daily dose: 0-12 hours and 12-24 hours following dosing. Additional
CR
subjects were able to enter the extension phase without undergoing the switch-over phase of the study. In the extension phase, blood and urine samples for PK analyses were collected at months 0 (baseline), 1, 2,
US
3, 6, 9, and 12, and at week 1 for patients who did not participate in the switch over. There was no prespecified timing defined in the protocol for collection of these samples. Study 2 (NCT02246218)
AN
2.1.2
Study 2 was an open-label study where patients could enter naïve to nitrogen-scavenging medication or
M
switch from current nitrogen-scavenging treatment to GPB. Plasma PK samples were collected at predose and post-dose at 4-6 hours, 8 hours and between 12 and 24 hours on the first day of GPB dosing
ED
only. Urine PK samples were collected pre-dose and post-dose at 1, 2, 4-6, 8, and 12-24 hours. Additional plasma and urine sampling (at time 2 – 12 hours post first dose of the day) was performed on day 7,
2.2
PT
months 1-6, and every 3 months until subjects completed or prematurely terminated the study. Pharmacokinetic analysis
CE
Plasma and urine samples were analyzed via a validated method. Liquid chromatography/mass spectroscopy/mass spectroscopy (LC/MS/MS) analysis was performed for the quantitation of PAGN in
AC
urine and PBA, PAA, and PAGN in plasma. The individual plasma concentration-time data of PBA, PAA, and PAGN following oral administration of GPB were analyzed using non-compartmental methods. For short-term (24-hour) PK measurements of AUC, Cmax, Cmin , and Tmax, the results of Study 1 and Study 2 were analyzed separately as Study 1 PK samples for GPB were collected after at least 4 days of GPB and Study 2 PK samples were collected on day 1 of GPB only dosing. Initial efforts were made to develop a population PK model using PK data collected from infants 2 to 24 months old. Due to the small sample size and the sparse sampling schedule, the data did not fully support a similar PK model as described in the literature for adults and older children [7]. Therefore, empirical
ACCEPTED MANUSCRIPT Bayes estimates of the individual PK parameters were generated based on individual plasma concentration measurements using the previously described PK model [7]. Individual PK parameter estimates of these infants were then compared to those reported in older patients. The plasma concentrations of the three analytes from the 10 patients in Study 2 were plotted against the reported concentrations in older children and adults for comparison. 2.3
Software and Statistical Methods
T
Pharmacokinetic parameters of PBA, PAA and PAGN in plasma and UPAGN were calculated using
IP
Phoenix WinNonlin™ (Version 6.3; Certara USA Inc. NJ). The software package NONMEM™, version
CR
7.2 (ICON, Hanover, MD, USA), was used for the empirical Bayesian estimation. Individual plasma and urine data and PK parameters of PBA, PAA and/or PAGN were summarized with descriptive statistics by treatment and analyte (i.e., N, mean, standard deviation (SD), Coefficient of variation (CV%), Median,
US
Min, Max). Naïve subjects with below quantification limit (BLQ) at time 0 were set to 0. All other BLQ were set to 0.5 µg/mL for this summary. Previous covariate analysis identified that body surface area
AN
(BSA) was a significant covariate accounting for age-related PK differences [7]. To test if this is also true in infant UCD patients, linear regression analysis was conducted to evaluate the correlation between body
M
mass metrics including BSA and total body weight (WT) with PK parameters. Results
3.1
Patient disposition and demographics
ED
3.
PT
Four UCD patients underwent the switch-over phase of Study 1. These patients continued into a longterm extension study along with 3 additional patients added to the long-term analysis (for a total of 7 patients) who were analyzed for up to 1 year on GPB with only 6 of the them contributing to PK data at 1
CE
year. Ten patients entered Study 2 and contributed to 24-hour PK data. Three patients were dosed with GPB for ≥6 months with 1 patient contributing PK data at 6 months.
AC
In Study 1, all subjects were transitioned to GPB from NaPBA. In Study 2, 7 of the 10 subjects were already stable on other nitrogen scavengers and were transitioned to GPB. Two subjects were newly diagnosed at the time of enrollment and one subject entered just after hyperammonemic crisis (HAC) and transitioned from Ammonul® (sodium phenylacetate and sodium benzoate) for injection, to GPB. Overall, 11 subjects transitioned from NaPBA in a single step with 24 hour monitoring, 1 patient transitioned from Ammonul in 3 steps (100% Ammonul dose and 50% GPB dose for 4 – 8 hours followed by 50% Ammonul dose and 100% GPB dose for 4 – 8 hours followed by discontinuation of Ammonul), 2 patients were naïve to nitrogen scavenging therapy and data on transition for 3 patients was not available due to them transitioning directly into the long-term follow-up study. PK
ACCEPTED MANUSCRIPT samples were taken on day 1 in all Study 2 subjects except for the one entering in hyperammonemic crisis (PK on day 2). Collectively, data from subjects with five UCD subtypes were analyzed. The median age of subjects was 10 months (range 2 to 21 months) with an average body surface area (BSA) of 0.425 m2 (SD: 0.076) (Table 1). As previously described by Berry et al., the mean total daily dose and dose range of GPB during the study were 3.9 grams (1.65, 7.86), providing 8.6 g/m2 (3.68, 13.5), 3.5 mL (1.50, 7.15) or 398
T
mg/kg (164, 554) administered per day.[11] GPB was given in 3, 4 and 6 doses per day in 9, 6 and 2 Forty one percent (n=7) of subjects had a G tube at some point during the
IP
subjects, respectively.[11]
CR
study. Due to the small number of patients and sparsity of data collection, PK analytes were not available at all time points for all patients, however, ammonia levels were not widely different among those who had differing administration schedules, likely because daily doses were standardized by BSA and amongst
US
the individual feeding schedules.
AN
Table 1. Patient Demographics and Characteristics
Study 2
Total
(N=7)
(N=10)
(N=17)
5 (71.4%)
5 (50.0%)
10 (58.8%)
2 (28.6%)
5 (50.0%)
7 (41.2%)
11.1 (7.5)
9.9 (5.5)
10.4 (6.2)
11
9
10
2, 21
4, 21
2, 21
17.6 (1.4)
18.2 (3.8)
17.9 (3.0)
17.3
17.8
17.3
16.1, 19.7
14.1, 27.0
14.1, 27.0
0.443 (0.090)
0.412 (0.067)
0.425 (0.076)
M
Gender
ED
Female
PT
Male Age (months)
Study 1
CE
Mean (SD) Median
AC
Min, Max
Body Mass Index (kg/m2) Mean (SD) Median Min, Max Body Surface Area (m2 ) Mean (SD)
Study 1
Study 2
Total
(N=7)
(N=10)
(N=17)
0.46
0.41
0.43
0.32, 0.56
0.32, 0.52
0.32, 0.56
ASL deficiency
2 (28.6%)
3 (30.0%)
5 (29.4%)
ASS deficiency
3 (42.9%)
2 (20.0%)
5 (29.4%)
OTC deficiency
1 (14.3%)
2 (20.0%)
3 (17.6%)
ARG deficiency
1 (14.3%)
CR
ACCEPTED MANUSCRIPT
1 (10.0%)
2 (11.8%)
CPS I deficiency
0
2 (20.0%)
2 (11.8%)
Median Min, Max
US
IP
T
UCD Diagnosis
3.2.1
Study 1 (Steady State)
M
Short Term Pharmacokinetics
ED
3.2
AN
ARG (arginase); ASL (argininosuccinate lyase); ASS (argininosuccinate synthetase); CPS I (carbamyl phosphate synthetase I); DNA (deoxyribonucleic acid); max (maximum); min (minimum); OTC (ornithine transcarbamylase); SD (standard deviation); UCD (urea cycle disorder)
In a post-hoc model of Study 1 results, slower absorption was demonstrated with GPB in comparison to
PT
NaPBA and this was consistent with what was seen and reported in older children and adults.[7] PAA exposure (AUC0-24 ) with GPB and NaPBA was 3,322 µg*hr/mL (n=3) and 4,138 µg*hr/mL (n=4),
CE
respectively (Table 2). The PK profile for one subject on GPB could not be calculated due to an inadequate number of samples. One of the three subjects experienced high PAA AUC 0-24 levels while on
AC
both GPB and NaPBA.
Table 2. Plasma Pharmacokinetic Parameters at Steady State Following Multiple Administrations of GPB and NaPBA in Study 1
Pharmacokinetic Descriptor PBA in Plasma N AUC0-24 (µg•h/mL) Cmaxss (µg/mL) Cminss (µg/mL)
Arithmetic Mean (CV %) GPB a
NaPBA b
3 242 (77.4) 45.2 (72.1) 0.00 (NC)
4 265 (90.9) 39.7 (97.9) 2.44 (167.9)
ACCEPTED MANUSCRIPT
4 4138 (123.7) 227 (102.9) 171 (142.1)
3 1775 (76.3) 104 (54.6) 54.8 (128.3)
4 1463 (49.1) 93.6 (34.2) 52.7 (71.2)
20.1 (30.9) 21.0 (80.7)c 17.5 (44.0)d
6.6 (48.7)d 27.1 (109.5)c 17.9 (74.3)d
IP
T
3 3322 (148.0) 196 (128.0) 118 (172.2)
CR
PAA in Plasma N AUC0-24 (µg•h/mL) Cmaxss (µg/mL) Cminss (µg/mL) PAGN in Plasma N AUC0-24 (µg•h/mL) Cmaxss (µg/mL) Cminss (µg/mL) PAGN in urine (mg/mL) Concentration 0 h Concentration 12 h Concentration 24 h
NC (not calculated) 24 hour PK after at least 4 days on GPB b 24 hour PK at start of study while on NaPBA c N=2 d N=3
Study 2 (Day 1 of GPB dosing)
AN
3.2.2
US
a
Mean Cmax for PBA, PAA, and PAGN were 42.4, 36.5, and 62.5 µg/mL respectively on the first day of
M
GPB-only dosing (Table 3). PBA, PAA, and PAGN concentrations in plasma were characterized by high
ED
variability given study constraints in this patient population. This variability was primarily due to the sampling time intervals being wide and not adjusted for actual dosing time. A regimented sampling protocol over 24 hours was not possible in this infant patient population given the medical management
PT
challenges of these UCD infants and their small sizes. Median Tmax occurred at approximately 10 hrs for PBA, 6.4 hrs for PAA, and 4.4 hrs for PAGN (Table 3). An accurate assessment of Tmax was difficult to
CE
obtain from these study data due to limited sampling times. Though limited samples were available, and variability was high for the 10 subjects in Study 2, BSA and weight significantly correlated with PBA
AC
clearance (P<0.001). BSA correlated slightly better than weight with PBA clearance, with a correlation coefficient of 0.737 versus 0.702. A similar phenomenon was also observed for PAGN clearance. Both BSA and weight were significant predictors of the apparent clearance of PAGN (P<0.001) with a higher correlation coefficient for BSA than for weight, which is similar to reports in older UCD patients [7]. Table 3. Plasma Pharmacokinetic Results for the First Full Day of GPB Dosing in Study 2 Parameter Cmax (µg/mL), n
PBA
PAA
PAGN
10
10
10
62.5 (27.3)
26.2 (10, 120)
31.0 (4.6, 101)
56.4 (34.8, 127)
86.5
87.0
43.7
10
10
10
1.8 (2.1)
4.3 (6.3)
20.6 (14.5)
0.8 (0, 6.3)
0.8 (0, 18.3)
115.8
145.6
70.5
10
10
10
18.2 (16.5)
42.0 (17.9)
12.9 (2.4, 48.2)
39.7 (13.7, 78.9)
100.9
90.8
42.5
10
10
10
6.4 (0, 22.3)
4.4 (0, 22.3)
10
10
10
286.2 (289.8)
249.4 (235.2)
583.8 (285.2)
175.2 (37.2, 832.3)
176.3 (26.4, 669.1)
530.5 (177.5, 1029.8)
101.2
94.3
48.9
Cmin (µg/mL), n Mean (SD) Median (min, max) CV% Cavg (µg/mL), n Mean (SD)
19.0 (19.1) 12.6 (3.7, 65.8)
CV%
PT
AUC0-last (µg*hr/mL), n Mean (SD)
CE
Median (min, max) CV%
10.0 (0, 12.6)
ED
Median (min, max)
M
Tmax (hr), n
AN
Median (min, max)
IP
CV%
T
36.5 (31.8)
US
Mean (SD) Median (min, max)
42.4 (36.7)
CR
ACCEPTED MANUSCRIPT
20.4 (0, 43.9)
AC
Cmax (maximum observed plasma concentration), C min (minimum observed plasma concentration), Cavg (average plasma concentration), T max (time at which the maximum plasma concentration was observed), AUC 0-last (area under the plasma concentration-time curve from time 0 to time of last measurable plasma concentration) Note: Naïve subjects with below quantification limit (BLQ) at time 0 were set to 0. All other BLQ were set to 0.5 µg/mL for this summary.
3.2.3
Integrated Findings from the Studies
First 24-hour PK data while on GPB was available for 13 subjects participating in Studies 1 or 2. Plasma concentration-time profiles of PBA, PAA and PAGN for individual patients are displayed in Figure 2. These data demonstrate that the initial dosing recommendations present some directional PK variability in a few patients but are similar over all. The PK profiles of PBA, PAA and PAGN observed in these young patients were compared to adult patients by plotting the concentration-time data observed in patients
ACCEPTED MANUSCRIPT against the PK profile observed in adults. As shown in Figure 3, most of the observations in the infants are within or below the 90% prediction interval in previous studies, suggesting a similar or lower overall exposure of these three analytes. Individual post-hoc Bayes estimates for PBA apparent clearance ranged from 9.7 to 14.3 L/h/70kg; these values were within the range observed in the studies in older patients (7.6 to 14.4 L/h/70kg). As reported previously, PAA levels were found to be relatively consistent across the previously recommended dosing range based on BSA in UCD patients, but variability has been
T
reported [7]. To examine this further we completed a correlation/regression among the 13 patients who
IP
had PAA levels, observing a statistically significant correlation between dose/BSA and PAA concentration (p = 0.048), but the coefficient of determination was low (r2 = 0.31). This was due to an
CR
outlier subject that received the highest daily dose/BSA (12.86 mL/m2 /day) and had the highest mean daily PAA level (376 µg/mL). The doses in the other 12 patients ranged from 4.41 to 11.67 mL/m2 /day
US
and reported reduced PAA variability as seen in Figure 2.
In both studies, PAGN appeared in the urine on day 1. Urine sampling times differed between the two
AN
studies. In Study 1, the mean U-PAGN concentration with GPB was highest at the 12-hour sampling time (20.96 mg/mL, range of 9.00 to 32.92). In Study 2, the mean U-PAGN concentration was highest at the
AC
CE
PT
ED
M
sampling time of 12-24 hours (7.56 mg/mL, range of 2.61 to 24.60).
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
ED
Figure 2: PK profiles of GPB analytes
AC
CE
PT
Note: The different colored lines represent individual patient observations.
AC
CE
PT
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
Figure 3. Short-term plasma PK results for pediatric subjects aged 2 months to < 2 years compared with observations in previous adult analyses . This is a model qualification plot using a visual predictive check (VPC) for PBA, PAA and PAGN for 2m to <2 years compared to the adult population. The VPC used the original dataset from the actual studies, enriched with plasma sampling that was
ACCEPTED MANUSCRIPT simulated every hour for 24 hours instead of the actual sampling times. This dataset was used to calculate a 90% prediction interval, which is shown by the shaded area. The numbers in the plots indicate the study number from which the data originated [3 = UP-1204-003 (NCT00551200); 6=HPN-100-006 (NCT00992459)] and serve as a symbol of the data point. The solid blue line is the VPC median profile of the simulated PK profile and the solid black line is the median profile of the observed PK profile from previously published studies.
3.3
Long Term Pharmacokinetics
Mean targeted ammonia values were maintained at scheduled visits throughout the studies. Mean
T
normalized ammonia values (± SD) were 89.2 µmol/L (± 63.1) at baseline (n=17) and 35.7 µmol/L (±
IP
15.7) at the end of study (n=6). Median plasma PBA/PAA/PAGN values did not change over time.
CR
Overall median plasma PAA was 1.0 µg/mL in Study 1 and 13.3 µg/mL in Study 2, resulting in an overall concentration of 3.4 µg/mL for all treated patients. The short term variability in PK analytes, as seen in Figure 3, are not notably different from that found in older UCD subjects. The longer term variability
US
could have been affected by sparse sampling, but importantly, the data in aggregate point to safe and effective nitrogen scavenging. Median plasma PAA:PAGN ratios were <1 at each time point (Table 4).
AN
All patients with the exception of 2 had a plasma PAA:PAGN ratio <2 at each time point over the long term. One patient had a PAA value at month 1 which exceeded 500 µg/mL, and the other patient had a PAA value exceeding 200 µg/mL at month 3. These values subsequently decreased at the next measure
M
(9 months) at similar GPB doses with one patient’s value rising again at 12 months.[11] Though analytes
ED
decreased slightly over time as seen in Table 4, the mean PAA/PAGN ratios were remarkably consistent and mean UPAGN increased slightly or remained the same (Table 5), indicating efficient nitrogen
PT
scavenging over 12 months.
CE
Table 4. Long-term plasma concentrations in Study 1 and Study 2
AC
Timepoint (N)
PBA (µg/mL)
Median (Min, Max) PAA (µg/mL)
PAGN (µg/mL)
PAA:PAGN ratio
Month 1
5.63
4.87
33.35
0.149
(N=14)
(0.5, 97.2)
(0.5, 1214.5)
(5.5, 111.2)
(0.03, 10.92)
Month 3
1.38
3.62
30.41
0.119
(N=9)
(0.5, 29.7)
(0.5, 291.2)
(7.1, 95.4)
(0.04, 3.05)
Month 6
2.25
1.05
15.56
0.077
(N=5)
(0.5, 53.0)
(0.5, 13.3)
(6.5, 49.0)
(0.03, 0.27)
ACCEPTED MANUSCRIPT Month 9
1.22
0.50
11.27
0.146
(N=6)
(0.5, 2.1)
(0.5, 91.7)
(2.5, 64.6)
(0.04, 1.42)
Month 12
2.55
1.23
14.21
0.127
(N=6)
(0.5, 6.8)
(0.5, 175.2)
(3.1, 52.0)
(0.07, 3.37)
2.45
3.41
31.26
0.153
(0.5, 97.2)
(0.5, 1214.5)
(0.5, 111.2)
(0.03, 10.92)
IP
T
Overall*
CR
Below quantification limit values were set to 0.5 µg/mL for this summary *Overall values calculated using all subjects at all visits including months 2, 4, and 5 were not included here
A summary of the urinary concentrations for PAGN is presented in Table 5. Due to the difficulty in
US
getting urine volumes from this infant population, total urine volume was not collected and only ‘spot’ urine concentrations collected at pre-specified times are reported. Concentrations ranged from 0.18
AN
(month 2, not shown in table) to 42.30 mg/mL for urinary PAGN demonstrating intestinal hydrolysis of GPB and subsequent absorption of PBA with metabolism to PAA and conjugation with glutamine.
M
Table 5. Long-term Urine Concentrations of PAGN (mg/mL) in Study 1 and Study 2
AC
Month 3, n
Mean (SD) Median Min, Max Month 6, n
Sample Time Point 24 hours
2-12 hours
4
4
3
7
9.92 (11.47)
9.73 (5.27)
30.75 (7.75)
5.80 (4.98)
7.16
7.96
28.30
4.81
0.35, 25.02
5.64, 17.36
24.52, 39.43
0.41, 15.30
3
3
3
3
13.28 (7.73)
11.01 (11.32)
27.49 (12.57)
20.88 (15.94)
11.64
7.28
28.01
20.30
6.50, 21.70
2.02, 23.73
14.67, 39.79
5.24, 37.10
2
3
4
CE
Month 1, n
Min, Max
Sample Time Point 12 hours
PT
0 hours
Median
Study 2
ED
Timepoint
Mean (SD)
Study 1
ACCEPTED MANUSCRIPT
Mean (SD)
22.13
16.86
26.84
19.29, 24.98
7.52, 26.28
25.47, 44.30
1
2
2
13.40
11.70 (12.64)
16.40 (3.11)
11.70
16.40
2.76, 20.64
14.20, 18.61
Median Min, Max Month 12, n Mean (SD) Median
2
21.75 (6.32)
13.08 (2.53)
21.75
13.08
17.29, 26.22
11.29, 14.86
20.12 (4.45) 20.12
16.97, 23.26
Discussion
M
4.
2
AN
Min, Max
2
T
Month 9, n
30.86 (8.99)
IP
Min, Max
16.89 (9.38)
CR
Median
22.13 (4.02)
US
Mean (SD)
ED
This PK analysis indicates that pediatric UCD patients aged 2 months–2 years are able to effectively hydrolyze GPB and that this treatment, combined with diet and amino acid supplementation, can result in maintenance of blood ammonia within normal limits. There was no evidence of decreased metabolism of
PT
GPB in infants aged 2 months to 2 years as dosing was followed by prompt appearance of metabolites (PBA, PAA) in blood and PAGN in urine.
CE
Hydrolysis of GPB has been demonstrated in in vitro studies to occur via pancreatic lipases: human pancreatic triglyceride lipase (PTL), pancreatic lipase-related protein 2 (PLRP2), and carboxyl-ester
AC
lipase (CEL) [12]. PLRP2 and CEL, unlike PTL, are expressed at birth, suggesting that GPB may be digested by newborns, which is of particular relevance as UCDs can present shortly after birth.
While
recent data has demonstrated the safety and efficacy of GPB in children 2 months to 2 years, thereby suggesting that GPB is digestible in this population, limited data was available regarding the PK profile in these patients [11]. Monteleone, et al. previously developed an integrated pharmacokinetic model where plasma PBA, PAA, PAGN concentration time and urinary PAGN amount-time data following treatment were analyzed simultaneously using nonlinear mixed effects modelling [7]. An interesting finding from this analysis was that the pharmacokinetic behavior of GPB was similar across all age groups, including patients aged 2 months to 2 years (data from Study 1) [7]. Due to the limited number of patients in Study
ACCEPTED MANUSCRIPT 1, the data from Study 2 allows further understanding of how GPB is absorbed, metabolized and excreted in this patient population. The presence of plasma concentrations of PBA, PAA and PAGN as well as urinary concentrations of PAGN in the 24-hour analyte analysis adds further evidence of enzymatic cleavage of GPB in this population to yield the active moiety, PAA. Overall, concentrations of PBA, PAA and PAGN overlaid with the observations in previous studies in the adult population found similar exposure to these analytes, further supporting the similarity in PK behavior of GPB in patients aged 2
T
months to 2 years.
IP
Using the PAA:PAGN ratio as a proxy for the efficiency with which an individual converts PAA to
CR
PAGN and a predictor of patients at risk of having an elevated PAA level, a median ratio of <1 (as found in the long term PK analyses in this population) indicate a low risk of high PAA levels (>500 µg/mL) [13]. Previous analyses showed considerable variability in PAA levels over a 24-hr period in patients > 2
US
years regardless of the dose and enzyme deficiency and this also holds true in this patient population. BSA was found to be the most significant covariate for GPB analytes in infants 2 to <24 months old
AN
(similar to the adult population), suggesting that the current dosing regimen based on BSA is reasonable. Based on the 10 patients in Study 2, age does not appear to influence PK analytes except for the single
M
subject older than 14 months for whom the values appear consistently higher than the rest of the group and may be driven by this subject’s relatively high BSA. The doses of GPB (absolute or by BSA) do not
ED
appear to correlate with the main PK parameters for any analyte. The patients reported here had large inter-individual variability in their PK analytes as demonstrated by
PT
the inter-subject coefficient of variation (CV%), which has been previously demonstrated in older UCD populations. The variability in this analysis was primarily due to the sampling time intervals being wide
CE
and lack of accurate accounting for actual dosing time, which is a management challenge in infants with UCD. As previously alluded to by Berry, et al., there are physiologic limitations on blood draws in
AC
infants; thus data was not available at some time points [11]. In addition to the small number of patients, this constitutes one of the limitations of this analysis.
Another contributing factor to the variability
between subjects is that the conjugating enzyme (glutamine-N-phenylacetyl transferase) has never been purified, and it is not known how much inter-individual variability there is in the kinetics of the enzymatic reaction. These data support that, within the dosing range used here, GPB could be safely used by all individuals tested. Another limitation of this analysis was the sampling time in Study 2, which was the first 24 hours of GPB treatment. This does not allow sufficient time for steady state levels of GPB to be achieved, which may result in inconsistencies in the PK parameters such as T max. Another limitation of this analysis was the sampling time in Study 2, which was the first 24 hours of GPB treatment. This does not allow sufficient time for steady state levels of GPB to be achieved, which may result in
ACCEPTED MANUSCRIPT inconsistencies in the PK parameters such as T max. However, PK parameters were collected out to 12 months in many patients and detailed PK samples were taken at Day 10 in Study 1, at steady state. Though not all patients were transitioned from NaPBA 14 of 17 were, moderating the variability that could be observed as patients entered the study with differing levels of plasma PBA, PAA and PAGN. These limitations are, however, exceeded by the overall consistent direction of the results, which indicate that glycerol phenylbutyrate use in UCD patients 2 to < 24 months of age is effective and behaves much
Conclusions
IP
5.
T
the same as that in older patients.
CR
There was no evidence for decreased metabolism of GPB in these infants as dosing was followed by prompt appearance (first day) of metabolites (PBA, PAA) in blood and PAGN in urine. Elevated PAA levels were infrequent and transient; the median PAA:PAGN ratio of <1 in the long-term PK analyses
US
suggests that PAA to PAGN conversion was not saturated. Analyte values were generally within the range of that seen in older patients and, as observed in older patients, BSA was found to be the most
AN
significant covariate, reinforcing that dosing should be based on body mass. These results collectively indicate that pediatric UCD patients aged 2 months–2 years effectively hydrolyze GPB to release PBA
M
and convert PAA to UPAGN so as to maintain blood ammonia within normal limits. The observation that the conjugating enzyme was not saturated at any dosing level used in the study suggests that the safe
ED
dosing range may actually be higher than currently reported, and that the PK parameters used here could be of benefit in assessment of patients in whom higher doses might be considered.
PT
Conflicts of interest statement
Horizon Pharma funded the development, conduct, and analysis of the studies. B. Robinson, R. Perdok,
CE
and R Holt employees of, and have stock in, Horizon Pharma USA, Inc. None of the other authors has a financial interest in Horizon Pharma USA, Inc.
AC
For the following authors, payments were made by Horizon Pharma to their institutions for services provided in the conduct of the clinical studies upon which this report is based: George Diaz (Icahn School of Medicine at Mount Sinai), Uta Lichter-Konecki (Children's National Medical Center and Children's Hospital of Pittsburgh), Shawn McCandless (University Hospitals Cleveland Medical Center), Cary O. Harding (Oregon Health & Science University), Nicola Longo (University of Utah), Roberto Zori (University of Florida), Can Ficicioglu (Children's Hospital of Philadelphia), Wendy E. Smith (Maine Medical Center), Jerry Vockley (Children's Hospital of Pittsburgh) and Susan Berry (University of Minnesota).
ACCEPTED MANUSCRIPT Can Ficicioglu has provided consulting support to and received grant support and honoraria for speaking engagements from BioMarin, Abbott Laboratories, Alexion, Shire, Pfizer, Swedish Orphan (Sobi), Sanofi-Genzyme, and Horizon Pharma (previously Hyperion Therapeutics, Inc). For M. Dong, the Cincinnati Children’s Hospital Medical Center has received research funds from Horizon Pharma to conduct this research. A.A. Vinks is a consultant for the sponsor through an agreement between Horizon Pharma and Cincinnati Children’s Hospital Medical Center.
T
Susan Berry, Nicola Longo, George Diaz, Shawn McCandless, Wendy E. Smith, Roberto Zori, Can
IP
Ficicioglu and Jerry Vockley have served as Horizon Pharma Advisory Board Members.
CR
Acknowledgments
Research at Oregon Health & Science University (Cary Harding) was made possible with support from
US
the Oregon Clinical and Translational Research Institute (OCTRI), grant number UL1TR000128 from the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes
AN
of Health (NIH), and NIH Roadmap for Medical Research.
The authors would like to acknowledge the following study coordinators: Carrie Bailey BS, CCRC,
M
Michele Bergman RN, BSN, Jenny Billy, Sarah Couchon, Sara Elsbecker MS, APRN, CNP, Luca Fierro MS, CGC, Christel Gross RN, Michelle Hunter, Deborah Kawchak MS, RD,
ED
CCRP, Margaret Kuenzi, Angela Leshinski MBA, RD, LDN, Jessica Lindenberger RN, Susan Mortenson RN, CCRC, Hadley Morotti, Audrey Lynn, and Kim Wallis MS, LGC. In addition, the authors would like
PT
to thank Megan Francis-Sedlak PhD, and Teresa M.Y. Kok RPh, MBA from Horizon Pharma for data
References
CE
analysis, writing and editorial support.
Batshaw ML. Hyperammonemia. Current problems in pediatrics 1984 Nov;14(11):1-69.
2.
Berry SA, Lichter-Konecki U, Diaz GA, McCandless SE, Rhead W, Smith W, et al. Glycerol
AC
1.
phenylbutyrate treatment in children with urea cycle disorders: pooled analysis of short and long-term ammonia control and outcomes. Molecular genetics and metabolism 2014 May;112(1):17-24. 3.
Diaz GA, Krivitzky LS, Mokhtarani M, Rhead W, Bartley J, Feigenbaum A, et al. Ammonia
control and neurocognitive outcome among urea cycle disorder patients treated with glycerol phenylbutyrate. Hepatology (Baltimore, Md) 2013 Jun;57(6):2171-9. 4.
Lee B, Rhead W, Diaz GA, Scharschmidt BF, Mian A, Shchelochkov O, et al. Phase 2
comparison of a novel ammonia scavenging agent with sodium phenylbutyrate in patients with urea cycle
ACCEPTED MANUSCRIPT disorders: safety, pharmacokinetics and ammonia control. Molecular genetics and metabolism 2010 Jul;100(3):221-8. 5.
Nagamani SC, Diaz GA, Rhead W, Berry SA, Le Mons C, Lichter-Konecki U, et al. Self-reported
treatment-associated symptoms among patients with urea cycle disorders participating in glycerol phenylbutyrate clinical trials. Molecular genetics and metabolism 2015 Sep-Oct;116(1-2):29-34. 6.
Smith W, Diaz GA, Lichter-Konecki U, Berry SA, Harding CO, McCandless SE, et al. Ammonia
T
control in children ages 2 months through 5 years with urea cycle disorders: comparison of sodium
Monteleone JP, Mokhtarani M, Diaz GA, Rhead W, Lichter-Konecki U, Berry SA, et al.
CR
7.
IP
phenylbutyrate and glycerol phenylbutyrate. The Journal of pediatrics 2013 Jun;162(6):1228-34, 34.e1.
Population pharmacokinetic modeling and dosing simulations of nitrogen-scavenging compounds:
US
disposition of glycerol phenylbutyrate and sodium phenylbutyrate in adult and pediatric patients with urea cycle disorders. Journal of clinical pharmacology 2013 Jul;53(7):699-710. Longo N, Holt RJ. Glycerol phenylbutyrate for the maintenance treatment of patients with
AN
8.
deficiencies in enzymes of the urea cycle. Expert Opinion on Orphan Drugs 2017;5(12):999-1010. Thibault A, Cooper MR, Figg WD, Venzon DJ, Sartor AO, Tompkins AC, et al. A phase I and
M
9.
pharmacokinetic study of intravenous phenylacetate in patients with cancer. Cancer research 1994 Apr
10.
ED
01;54(7):1690-4.
Thibault A, Samid D, Cooper MR, Figg WD, Tompkins AC, Patronas N, et al. Phase I study of
11.
PT
phenylacetate administered twice daily to patients with cancer. Cancer 1995 Jun 15;75(12):2932-8. Berry SA, Longo N, Diaz GA, McCandless SE, Smith WE, Harding CO, et al. Safety and
CE
efficacy of glycerol phenylbutyrate for management of urea cycle disorders in patients aged 2 months to 2 years. Molecular genetics and metabolism 2017;122:46-53. McGuire BM, Zupanets IA, Lowe ME, Xiao X, Syplyviy VA, Monteleone J, et al. Pharmacology
AC
12.
and safety of glycerol phenylbutyrate in healthy adults and adults with cirrhosis. Hepatology (Baltimore, Md) 2010 Jun;51(6):2077-85. 13.
Mokhtarani M, Diaz GA, Rhead W, Berry SA, Lichter-Konecki U, Feigenbaum A, et al. Elevated
phenylacetic acid levels do not correlate with adverse events in patients with urea cycle disorders or hepatic encephalopathy and can be predicted based on the plasma PAA to PAGN ratio. Molecular genetics and metabolism 2013 Dec;110(4):446-53.
Figure 1
Figure 2
Figure 3
S.A. Berry, J. Vockley, A.A. Vinks, M. Dong, G.A. Diaz, S.E. McCandless, W.E. Smith, C.O. Harding, R. Zori, C. Ficicioglu, U. Lichter-Konecki, R. Perdok, B. Robinson, R.J. Holt, N. Longo PII: DOI: Reference:
S1096-7192(18)30379-2 doi:10.1016/j.ymgme.2018.09.001 YMGME 6402
To appear in:
Molecular Genetics and Metabolism
Received date: Revised date: Accepted date:
23 June 2018 27 July 2018 2 September 2018
Please cite this article as: S.A. Berry, J. Vockley, A.A. Vinks, M. Dong, G.A. Diaz, S.E. McCandless, W.E. Smith, C.O. Harding, R. Zori, C. Ficicioglu, U. Lichter-Konecki, R. Perdok, B. Robinson, R.J. Holt, N. Longo , Pharmacokinetics of glycerol phenylbutyrate in pediatric patients 2 months to 2 years of age with urea cycle disorders. Ymgme (2018), doi:10.1016/j.ymgme.2018.09.001
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Title: Pharmacokinetics of glycerol phenylbutyrate in pediatric patients 2 months to 2 years of age with Urea Cycle Disorders Authors: Berry SAa, Vockley Jb , Vinks AAc, Dong, Mc, Diaz GAd , McCandless SEe, Smith WEf, Harding COg , Zori Rh , Ficicioglu Ci , Lichter-Konecki Uj , Perdok Rk , Robinson Bk , Holt RJk,l , Longo Nm.
T
a
University of Minnesota Department of Pediatrics, Minneapolis, MN, USA University of Pittsburgh School of Medicine and the Children’s Hospital of Pittsburgh, Pittsburgh, PA, USA c Division of Clinical Pharmacology, Cincinnati Children’s Hospital Medical Center and Department of Pediatrics, University of Cincinnati, Cincinnati, OH, USA d Icahn School of Medicine at Mount Sinai, New York, NY, USA e University of Colorado Denver School of Medicine and Children’s Hospital Colorado, Aurora, CO, USA f Maine Medical Center, Portland, ME, USA g Oregon Health & Science University, Portland, OR, USA h University of Florida, Gainesville, FL, USA i Children’s Hospital of Philadelphia, Philadelphia, PA, USA j Children’s Hospital of Pittsburgh, Pittsburgh, PA, USA k Horizon Pharma USA, Inc, Lake Forest, IL, USA l University of Illinois-Chicago, Chicago, IL USA m University of Utah, Salt Lake City, UT, USA
M
AN
US
CR
IP
b
ED
Abstract
CE
PT
Introduction: Glycerol phenylbutyrate (GPB) is approved in the US and EU for the chronic management of patients ≥2 months of age with urea cycle disorders (UCDs) who cannot be managed by dietary protein restriction and/or amino acid supplementation alone. GPB is a pre-prodrug, hydrolyzed by lipases to phenylbutyric acid (PBA) that upon absorption is beta-oxidized to the active nitrogen scavenger phenylacetic acid (PAA), which is conjugated to glutamine (PAGN) and excreted as urinary PAGN (UPAGN). Pharmacokinetics (PK) of GPB were examined to see if hydrolysis is impaired in very young patients who may lack lipase activity.
AC
Methods: Patients 2 months to <2 years of age with UCDs from two open label studies (n=17, median age 10 months) predominantly on stable doses of nitrogen scavengers (n = 14) were switched to GPB. Primary assessments included traditional plasma PK analyses of PBA, PAA, and PAGN, using noncompartmental methods with WinNonlin™. UPAGN was collected periodically throughout the study up to 12 months. Results: PBA, PAA and PAGN rapidly appeared in plasma after GPB dosing, demonstrating evidence of GPB cleavage with subsequent PBA absorption. Median concentrations of PBA, PAA and PAGN did not increase over time and were similar to or lower than the values observed in older UCD patients. The median PAA/PAGN ratio was well below one over time, demonstrating that conjugation of PAA with glutamine to form PAGN did not reach saturation. Covariate analyses indicated that age did not influence the PK parameters, with body surface area (BSA) being the most significant covariate, reinforcing current BSA based dosing recommendations as seen in older patients.
ACCEPTED MANUSCRIPT Conclusion: These observations demonstrate that UCD patients aged 2 months to <2 years have sufficient lipase activity to adequately convert the pre-prodrug GPB to PBA. PBA is then converted to its active moiety (PAA) providing successful nitrogen scavenging even in very young children. Keywords: urea cycle disorders, glycerol phenylbutyrate, infants, children, pharmacokinetics 1.
Introduction
Glycerol phenylbutyrate (GPB) is approved in the US and EU for the management of patients ≥2 months
T
of age with urea cycle disorders (UCDs) who cannot be managed by dietary protein restriction and/or
IP
amino acid supplementation alone. GPB is hydrolyzed by pancreatic lipases to yield glycerol and phenylbutyric acid (PBA), which undergoes β-oxidation in the liver to phenylacetic acid (PAA). PAA is
CR
conjugated to glutamine in the liver and the kidney by glutamine-N-phenylacetyltransferase to form phenylacetylglutamine (PAGN), which is subsequently excreted in urine (UPAGN). (Figure 1) On a
US
molar basis, PAGN, like urea, contains 2 moles of nitrogen and provides an alternate mechanism for
AC
CE
PT
ED
M
AN
waste nitrogen excretion [1].
Figure 1. Metabolizing pathway and mechanism of action of GPB GPB (glycerol phenylbutyrate); PAA (phenylacetic acid); PBA (phenylbutyric acid); PAGN (phenylacetylglutamine)
ACCEPTED MANUSCRIPT Clinical trials in adults and children > 2 years of age indicate that GPB is at least as effective as sodium phenylbutyrate (NaPBA) without the associated bad taste, odor, sodium content, and burden of high pill number administration [2-6]. In addition, analysis of pooled data from these trials suggest that GPB might confer better control of ammonia due to slower absorption and reduced fluctuations in plasma metabolite concentrations, as demonstrated by previous pharmacokinetic (PK) modeling [7, 8]. In this modeling, body size was found to significantly affect clearance, volume, and presystemic conversion,
T
resulting in smaller individuals with smaller body surface area (BSA) having smaller PK values for these
IP
parameters [7]. The lower clearance for both PBA and PAGN coupled with the saturable conversion of PAA to PAGN may result in greater PAA exposure in younger patients. Greater PAA exposure in
CR
smaller patients was more pronounced with NaPBA than with GPB, despite equivalent dosing; PAA levels reached steady state in approximately 3 days without further accumulation [7]. Notably, the median
US
PAA levels were well below 500 µg/ml (the theoretical toxic limit based on studies in cancer patients using intravenous PAA) [9, 10]. In addition, neurotoxic adverse events have not been associated with
AN
either NaPBA or GPB [3].
Recently published data extended and confirmed the safety and efficacy of GPB in children 2 months to
M
<2 years of age [11]. However, limited data exist on the PK of GPB (or NaPBA) in this patient population. Here we examine the short- and long-term PK of GPB in UCD patients 2 months to <2 years
ED
of age whose pancreatic and liver function are immature to determine if there are notable differences compared to that previously reported in older UCD patients. Materials and Methods
2.1
Study design and treatments
PT
2.
CE
The PK of seventeen subjects 2 months to 2 years of age were evaluated from two clinical trials (HPN100-012 switch over and extension study and HPN100-009). Studies HPN100-012 and HPN-100-
AC
009 will be referred to as Study 1 and 2, respectively. The study design, patient disposition, and treatments have been previously described [2, 11]. As in previous studies, patients on stable doses of NaPBA were switched to an equimolar dose of GPB, however, if they were naïve to phenylbutyrate their initial doses were based on BSA (8.5 mL/m2 /day) [11].
For those receiving GPB orally, it was
administered via an oral syringe, just prior to breastfeeding or intake of formula or food. If needed, GPB was added to a small amount of formula in a small syringe and administered orally. A G tube was utilized in patients who were unable to tolerate oral dosing. 2.1.1
Study 1 (NCT01347073)
ACCEPTED MANUSCRIPT The switch-over phase of Study 1 was designed as a fixed-sequence, open-label, switch to GPB from chronic treatment with NaPBA. On Day 1, subjects received their prescribed dose of NaPBA with feedings. Subjects were observed in a monitored clinical setting for 24 hours while undergoing 24-hour blood sampling on their prescribed dose of NaPBA (Day 1) and then all subjects were switched from NaPBA to GPB in a single transition step. Subjects returned to the clinic after four to ten days on GPB for an additional visit (day 10) and 24-hour blood sampling, after which they continued in the long-term
T
extension phase. During the switch-over phase, plasma PK samples were collected at pre-dose and post-
IP
dose at 8, 12 and 24 hours and urine samples for PK analysis were collected on day 1 and day 10 at the two time periods relative to the first daily dose: 0-12 hours and 12-24 hours following dosing. Additional
CR
subjects were able to enter the extension phase without undergoing the switch-over phase of the study. In the extension phase, blood and urine samples for PK analyses were collected at months 0 (baseline), 1, 2,
US
3, 6, 9, and 12, and at week 1 for patients who did not participate in the switch over. There was no prespecified timing defined in the protocol for collection of these samples. Study 2 (NCT02246218)
AN
2.1.2
Study 2 was an open-label study where patients could enter naïve to nitrogen-scavenging medication or
M
switch from current nitrogen-scavenging treatment to GPB. Plasma PK samples were collected at predose and post-dose at 4-6 hours, 8 hours and between 12 and 24 hours on the first day of GPB dosing
ED
only. Urine PK samples were collected pre-dose and post-dose at 1, 2, 4-6, 8, and 12-24 hours. Additional plasma and urine sampling (at time 2 – 12 hours post first dose of the day) was performed on day 7,
2.2
PT
months 1-6, and every 3 months until subjects completed or prematurely terminated the study. Pharmacokinetic analysis
CE
Plasma and urine samples were analyzed via a validated method. Liquid chromatography/mass spectroscopy/mass spectroscopy (LC/MS/MS) analysis was performed for the quantitation of PAGN in
AC
urine and PBA, PAA, and PAGN in plasma. The individual plasma concentration-time data of PBA, PAA, and PAGN following oral administration of GPB were analyzed using non-compartmental methods. For short-term (24-hour) PK measurements of AUC, Cmax, Cmin , and Tmax, the results of Study 1 and Study 2 were analyzed separately as Study 1 PK samples for GPB were collected after at least 4 days of GPB and Study 2 PK samples were collected on day 1 of GPB only dosing. Initial efforts were made to develop a population PK model using PK data collected from infants 2 to 24 months old. Due to the small sample size and the sparse sampling schedule, the data did not fully support a similar PK model as described in the literature for adults and older children [7]. Therefore, empirical
ACCEPTED MANUSCRIPT Bayes estimates of the individual PK parameters were generated based on individual plasma concentration measurements using the previously described PK model [7]. Individual PK parameter estimates of these infants were then compared to those reported in older patients. The plasma concentrations of the three analytes from the 10 patients in Study 2 were plotted against the reported concentrations in older children and adults for comparison. 2.3
Software and Statistical Methods
T
Pharmacokinetic parameters of PBA, PAA and PAGN in plasma and UPAGN were calculated using
IP
Phoenix WinNonlin™ (Version 6.3; Certara USA Inc. NJ). The software package NONMEM™, version
CR
7.2 (ICON, Hanover, MD, USA), was used for the empirical Bayesian estimation. Individual plasma and urine data and PK parameters of PBA, PAA and/or PAGN were summarized with descriptive statistics by treatment and analyte (i.e., N, mean, standard deviation (SD), Coefficient of variation (CV%), Median,
US
Min, Max). Naïve subjects with below quantification limit (BLQ) at time 0 were set to 0. All other BLQ were set to 0.5 µg/mL for this summary. Previous covariate analysis identified that body surface area
AN
(BSA) was a significant covariate accounting for age-related PK differences [7]. To test if this is also true in infant UCD patients, linear regression analysis was conducted to evaluate the correlation between body
M
mass metrics including BSA and total body weight (WT) with PK parameters. Results
3.1
Patient disposition and demographics
ED
3.
PT
Four UCD patients underwent the switch-over phase of Study 1. These patients continued into a longterm extension study along with 3 additional patients added to the long-term analysis (for a total of 7 patients) who were analyzed for up to 1 year on GPB with only 6 of the them contributing to PK data at 1
CE
year. Ten patients entered Study 2 and contributed to 24-hour PK data. Three patients were dosed with GPB for ≥6 months with 1 patient contributing PK data at 6 months.
AC
In Study 1, all subjects were transitioned to GPB from NaPBA. In Study 2, 7 of the 10 subjects were already stable on other nitrogen scavengers and were transitioned to GPB. Two subjects were newly diagnosed at the time of enrollment and one subject entered just after hyperammonemic crisis (HAC) and transitioned from Ammonul® (sodium phenylacetate and sodium benzoate) for injection, to GPB. Overall, 11 subjects transitioned from NaPBA in a single step with 24 hour monitoring, 1 patient transitioned from Ammonul in 3 steps (100% Ammonul dose and 50% GPB dose for 4 – 8 hours followed by 50% Ammonul dose and 100% GPB dose for 4 – 8 hours followed by discontinuation of Ammonul), 2 patients were naïve to nitrogen scavenging therapy and data on transition for 3 patients was not available due to them transitioning directly into the long-term follow-up study. PK
ACCEPTED MANUSCRIPT samples were taken on day 1 in all Study 2 subjects except for the one entering in hyperammonemic crisis (PK on day 2). Collectively, data from subjects with five UCD subtypes were analyzed. The median age of subjects was 10 months (range 2 to 21 months) with an average body surface area (BSA) of 0.425 m2 (SD: 0.076) (Table 1). As previously described by Berry et al., the mean total daily dose and dose range of GPB during the study were 3.9 grams (1.65, 7.86), providing 8.6 g/m2 (3.68, 13.5), 3.5 mL (1.50, 7.15) or 398
T
mg/kg (164, 554) administered per day.[11] GPB was given in 3, 4 and 6 doses per day in 9, 6 and 2 Forty one percent (n=7) of subjects had a G tube at some point during the
IP
subjects, respectively.[11]
CR
study. Due to the small number of patients and sparsity of data collection, PK analytes were not available at all time points for all patients, however, ammonia levels were not widely different among those who had differing administration schedules, likely because daily doses were standardized by BSA and amongst
US
the individual feeding schedules.
AN
Table 1. Patient Demographics and Characteristics
Study 2
Total
(N=7)
(N=10)
(N=17)
5 (71.4%)
5 (50.0%)
10 (58.8%)
2 (28.6%)
5 (50.0%)
7 (41.2%)
11.1 (7.5)
9.9 (5.5)
10.4 (6.2)
11
9
10
2, 21
4, 21
2, 21
17.6 (1.4)
18.2 (3.8)
17.9 (3.0)
17.3
17.8
17.3
16.1, 19.7
14.1, 27.0
14.1, 27.0
0.443 (0.090)
0.412 (0.067)
0.425 (0.076)
M
Gender
ED
Female
PT
Male Age (months)
Study 1
CE
Mean (SD) Median
AC
Min, Max
Body Mass Index (kg/m2) Mean (SD) Median Min, Max Body Surface Area (m2 ) Mean (SD)
Study 1
Study 2
Total
(N=7)
(N=10)
(N=17)
0.46
0.41
0.43
0.32, 0.56
0.32, 0.52
0.32, 0.56
ASL deficiency
2 (28.6%)
3 (30.0%)
5 (29.4%)
ASS deficiency
3 (42.9%)
2 (20.0%)
5 (29.4%)
OTC deficiency
1 (14.3%)
2 (20.0%)
3 (17.6%)
ARG deficiency
1 (14.3%)
CR
ACCEPTED MANUSCRIPT
1 (10.0%)
2 (11.8%)
CPS I deficiency
0
2 (20.0%)
2 (11.8%)
Median Min, Max
US
IP
T
UCD Diagnosis
3.2.1
Study 1 (Steady State)
M
Short Term Pharmacokinetics
ED
3.2
AN
ARG (arginase); ASL (argininosuccinate lyase); ASS (argininosuccinate synthetase); CPS I (carbamyl phosphate synthetase I); DNA (deoxyribonucleic acid); max (maximum); min (minimum); OTC (ornithine transcarbamylase); SD (standard deviation); UCD (urea cycle disorder)
In a post-hoc model of Study 1 results, slower absorption was demonstrated with GPB in comparison to
PT
NaPBA and this was consistent with what was seen and reported in older children and adults.[7] PAA exposure (AUC0-24 ) with GPB and NaPBA was 3,322 µg*hr/mL (n=3) and 4,138 µg*hr/mL (n=4),
CE
respectively (Table 2). The PK profile for one subject on GPB could not be calculated due to an inadequate number of samples. One of the three subjects experienced high PAA AUC 0-24 levels while on
AC
both GPB and NaPBA.
Table 2. Plasma Pharmacokinetic Parameters at Steady State Following Multiple Administrations of GPB and NaPBA in Study 1
Pharmacokinetic Descriptor PBA in Plasma N AUC0-24 (µg•h/mL) Cmaxss (µg/mL) Cminss (µg/mL)
Arithmetic Mean (CV %) GPB a
NaPBA b
3 242 (77.4) 45.2 (72.1) 0.00 (NC)
4 265 (90.9) 39.7 (97.9) 2.44 (167.9)
ACCEPTED MANUSCRIPT
4 4138 (123.7) 227 (102.9) 171 (142.1)
3 1775 (76.3) 104 (54.6) 54.8 (128.3)
4 1463 (49.1) 93.6 (34.2) 52.7 (71.2)
20.1 (30.9) 21.0 (80.7)c 17.5 (44.0)d
6.6 (48.7)d 27.1 (109.5)c 17.9 (74.3)d
IP
T
3 3322 (148.0) 196 (128.0) 118 (172.2)
CR
PAA in Plasma N AUC0-24 (µg•h/mL) Cmaxss (µg/mL) Cminss (µg/mL) PAGN in Plasma N AUC0-24 (µg•h/mL) Cmaxss (µg/mL) Cminss (µg/mL) PAGN in urine (mg/mL) Concentration 0 h Concentration 12 h Concentration 24 h
NC (not calculated) 24 hour PK after at least 4 days on GPB b 24 hour PK at start of study while on NaPBA c N=2 d N=3
Study 2 (Day 1 of GPB dosing)
AN
3.2.2
US
a
Mean Cmax for PBA, PAA, and PAGN were 42.4, 36.5, and 62.5 µg/mL respectively on the first day of
M
GPB-only dosing (Table 3). PBA, PAA, and PAGN concentrations in plasma were characterized by high
ED
variability given study constraints in this patient population. This variability was primarily due to the sampling time intervals being wide and not adjusted for actual dosing time. A regimented sampling protocol over 24 hours was not possible in this infant patient population given the medical management
PT
challenges of these UCD infants and their small sizes. Median Tmax occurred at approximately 10 hrs for PBA, 6.4 hrs for PAA, and 4.4 hrs for PAGN (Table 3). An accurate assessment of Tmax was difficult to
CE
obtain from these study data due to limited sampling times. Though limited samples were available, and variability was high for the 10 subjects in Study 2, BSA and weight significantly correlated with PBA
AC
clearance (P<0.001). BSA correlated slightly better than weight with PBA clearance, with a correlation coefficient of 0.737 versus 0.702. A similar phenomenon was also observed for PAGN clearance. Both BSA and weight were significant predictors of the apparent clearance of PAGN (P<0.001) with a higher correlation coefficient for BSA than for weight, which is similar to reports in older UCD patients [7]. Table 3. Plasma Pharmacokinetic Results for the First Full Day of GPB Dosing in Study 2 Parameter Cmax (µg/mL), n
PBA
PAA
PAGN
10
10
10
62.5 (27.3)
26.2 (10, 120)
31.0 (4.6, 101)
56.4 (34.8, 127)
86.5
87.0
43.7
10
10
10
1.8 (2.1)
4.3 (6.3)
20.6 (14.5)
0.8 (0, 6.3)
0.8 (0, 18.3)
115.8
145.6
70.5
10
10
10
18.2 (16.5)
42.0 (17.9)
12.9 (2.4, 48.2)
39.7 (13.7, 78.9)
100.9
90.8
42.5
10
10
10
6.4 (0, 22.3)
4.4 (0, 22.3)
10
10
10
286.2 (289.8)
249.4 (235.2)
583.8 (285.2)
175.2 (37.2, 832.3)
176.3 (26.4, 669.1)
530.5 (177.5, 1029.8)
101.2
94.3
48.9
Cmin (µg/mL), n Mean (SD) Median (min, max) CV% Cavg (µg/mL), n Mean (SD)
19.0 (19.1) 12.6 (3.7, 65.8)
CV%
PT
AUC0-last (µg*hr/mL), n Mean (SD)
CE
Median (min, max) CV%
10.0 (0, 12.6)
ED
Median (min, max)
M
Tmax (hr), n
AN
Median (min, max)
IP
CV%
T
36.5 (31.8)
US
Mean (SD) Median (min, max)
42.4 (36.7)
CR
ACCEPTED MANUSCRIPT
20.4 (0, 43.9)
AC
Cmax (maximum observed plasma concentration), C min (minimum observed plasma concentration), Cavg (average plasma concentration), T max (time at which the maximum plasma concentration was observed), AUC 0-last (area under the plasma concentration-time curve from time 0 to time of last measurable plasma concentration) Note: Naïve subjects with below quantification limit (BLQ) at time 0 were set to 0. All other BLQ were set to 0.5 µg/mL for this summary.
3.2.3
Integrated Findings from the Studies
First 24-hour PK data while on GPB was available for 13 subjects participating in Studies 1 or 2. Plasma concentration-time profiles of PBA, PAA and PAGN for individual patients are displayed in Figure 2. These data demonstrate that the initial dosing recommendations present some directional PK variability in a few patients but are similar over all. The PK profiles of PBA, PAA and PAGN observed in these young patients were compared to adult patients by plotting the concentration-time data observed in patients
ACCEPTED MANUSCRIPT against the PK profile observed in adults. As shown in Figure 3, most of the observations in the infants are within or below the 90% prediction interval in previous studies, suggesting a similar or lower overall exposure of these three analytes. Individual post-hoc Bayes estimates for PBA apparent clearance ranged from 9.7 to 14.3 L/h/70kg; these values were within the range observed in the studies in older patients (7.6 to 14.4 L/h/70kg). As reported previously, PAA levels were found to be relatively consistent across the previously recommended dosing range based on BSA in UCD patients, but variability has been
T
reported [7]. To examine this further we completed a correlation/regression among the 13 patients who
IP
had PAA levels, observing a statistically significant correlation between dose/BSA and PAA concentration (p = 0.048), but the coefficient of determination was low (r2 = 0.31). This was due to an
CR
outlier subject that received the highest daily dose/BSA (12.86 mL/m2 /day) and had the highest mean daily PAA level (376 µg/mL). The doses in the other 12 patients ranged from 4.41 to 11.67 mL/m2 /day
US
and reported reduced PAA variability as seen in Figure 2.
In both studies, PAGN appeared in the urine on day 1. Urine sampling times differed between the two
AN
studies. In Study 1, the mean U-PAGN concentration with GPB was highest at the 12-hour sampling time (20.96 mg/mL, range of 9.00 to 32.92). In Study 2, the mean U-PAGN concentration was highest at the
AC
CE
PT
ED
M
sampling time of 12-24 hours (7.56 mg/mL, range of 2.61 to 24.60).
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
ED
Figure 2: PK profiles of GPB analytes
AC
CE
PT
Note: The different colored lines represent individual patient observations.
AC
CE
PT
ED
M
AN
US
CR
IP
T
ACCEPTED MANUSCRIPT
Figure 3. Short-term plasma PK results for pediatric subjects aged 2 months to < 2 years compared with observations in previous adult analyses . This is a model qualification plot using a visual predictive check (VPC) for PBA, PAA and PAGN for 2m to <2 years compared to the adult population. The VPC used the original dataset from the actual studies, enriched with plasma sampling that was
ACCEPTED MANUSCRIPT simulated every hour for 24 hours instead of the actual sampling times. This dataset was used to calculate a 90% prediction interval, which is shown by the shaded area. The numbers in the plots indicate the study number from which the data originated [3 = UP-1204-003 (NCT00551200); 6=HPN-100-006 (NCT00992459)] and serve as a symbol of the data point. The solid blue line is the VPC median profile of the simulated PK profile and the solid black line is the median profile of the observed PK profile from previously published studies.
3.3
Long Term Pharmacokinetics
Mean targeted ammonia values were maintained at scheduled visits throughout the studies. Mean
T
normalized ammonia values (± SD) were 89.2 µmol/L (± 63.1) at baseline (n=17) and 35.7 µmol/L (±
IP
15.7) at the end of study (n=6). Median plasma PBA/PAA/PAGN values did not change over time.
CR
Overall median plasma PAA was 1.0 µg/mL in Study 1 and 13.3 µg/mL in Study 2, resulting in an overall concentration of 3.4 µg/mL for all treated patients. The short term variability in PK analytes, as seen in Figure 3, are not notably different from that found in older UCD subjects. The longer term variability
US
could have been affected by sparse sampling, but importantly, the data in aggregate point to safe and effective nitrogen scavenging. Median plasma PAA:PAGN ratios were <1 at each time point (Table 4).
AN
All patients with the exception of 2 had a plasma PAA:PAGN ratio <2 at each time point over the long term. One patient had a PAA value at month 1 which exceeded 500 µg/mL, and the other patient had a PAA value exceeding 200 µg/mL at month 3. These values subsequently decreased at the next measure
M
(9 months) at similar GPB doses with one patient’s value rising again at 12 months.[11] Though analytes
ED
decreased slightly over time as seen in Table 4, the mean PAA/PAGN ratios were remarkably consistent and mean UPAGN increased slightly or remained the same (Table 5), indicating efficient nitrogen
PT
scavenging over 12 months.
CE
Table 4. Long-term plasma concentrations in Study 1 and Study 2
AC
Timepoint (N)
PBA (µg/mL)
Median (Min, Max) PAA (µg/mL)
PAGN (µg/mL)
PAA:PAGN ratio
Month 1
5.63
4.87
33.35
0.149
(N=14)
(0.5, 97.2)
(0.5, 1214.5)
(5.5, 111.2)
(0.03, 10.92)
Month 3
1.38
3.62
30.41
0.119
(N=9)
(0.5, 29.7)
(0.5, 291.2)
(7.1, 95.4)
(0.04, 3.05)
Month 6
2.25
1.05
15.56
0.077
(N=5)
(0.5, 53.0)
(0.5, 13.3)
(6.5, 49.0)
(0.03, 0.27)
ACCEPTED MANUSCRIPT Month 9
1.22
0.50
11.27
0.146
(N=6)
(0.5, 2.1)
(0.5, 91.7)
(2.5, 64.6)
(0.04, 1.42)
Month 12
2.55
1.23
14.21
0.127
(N=6)
(0.5, 6.8)
(0.5, 175.2)
(3.1, 52.0)
(0.07, 3.37)
2.45
3.41
31.26
0.153
(0.5, 97.2)
(0.5, 1214.5)
(0.5, 111.2)
(0.03, 10.92)
IP
T
Overall*
CR
Below quantification limit values were set to 0.5 µg/mL for this summary *Overall values calculated using all subjects at all visits including months 2, 4, and 5 were not included here
A summary of the urinary concentrations for PAGN is presented in Table 5. Due to the difficulty in
US
getting urine volumes from this infant population, total urine volume was not collected and only ‘spot’ urine concentrations collected at pre-specified times are reported. Concentrations ranged from 0.18
AN
(month 2, not shown in table) to 42.30 mg/mL for urinary PAGN demonstrating intestinal hydrolysis of GPB and subsequent absorption of PBA with metabolism to PAA and conjugation with glutamine.
M
Table 5. Long-term Urine Concentrations of PAGN (mg/mL) in Study 1 and Study 2
AC
Month 3, n
Mean (SD) Median Min, Max Month 6, n
Sample Time Point 24 hours
2-12 hours
4
4
3
7
9.92 (11.47)
9.73 (5.27)
30.75 (7.75)
5.80 (4.98)
7.16
7.96
28.30
4.81
0.35, 25.02
5.64, 17.36
24.52, 39.43
0.41, 15.30
3
3
3
3
13.28 (7.73)
11.01 (11.32)
27.49 (12.57)
20.88 (15.94)
11.64
7.28
28.01
20.30
6.50, 21.70
2.02, 23.73
14.67, 39.79
5.24, 37.10
2
3
4
CE
Month 1, n
Min, Max
Sample Time Point 12 hours
PT
0 hours
Median
Study 2
ED
Timepoint
Mean (SD)
Study 1
ACCEPTED MANUSCRIPT
Mean (SD)
22.13
16.86
26.84
19.29, 24.98
7.52, 26.28
25.47, 44.30
1
2
2
13.40
11.70 (12.64)
16.40 (3.11)
11.70
16.40
2.76, 20.64
14.20, 18.61
Median Min, Max Month 12, n Mean (SD) Median
2
21.75 (6.32)
13.08 (2.53)
21.75
13.08
17.29, 26.22
11.29, 14.86
20.12 (4.45) 20.12
16.97, 23.26
Discussion
M
4.
2
AN
Min, Max
2
T
Month 9, n
30.86 (8.99)
IP
Min, Max
16.89 (9.38)
CR
Median
22.13 (4.02)
US
Mean (SD)
ED
This PK analysis indicates that pediatric UCD patients aged 2 months–2 years are able to effectively hydrolyze GPB and that this treatment, combined with diet and amino acid supplementation, can result in maintenance of blood ammonia within normal limits. There was no evidence of decreased metabolism of
PT
GPB in infants aged 2 months to 2 years as dosing was followed by prompt appearance of metabolites (PBA, PAA) in blood and PAGN in urine.
CE
Hydrolysis of GPB has been demonstrated in in vitro studies to occur via pancreatic lipases: human pancreatic triglyceride lipase (PTL), pancreatic lipase-related protein 2 (PLRP2), and carboxyl-ester
AC
lipase (CEL) [12]. PLRP2 and CEL, unlike PTL, are expressed at birth, suggesting that GPB may be digested by newborns, which is of particular relevance as UCDs can present shortly after birth.
While
recent data has demonstrated the safety and efficacy of GPB in children 2 months to 2 years, thereby suggesting that GPB is digestible in this population, limited data was available regarding the PK profile in these patients [11]. Monteleone, et al. previously developed an integrated pharmacokinetic model where plasma PBA, PAA, PAGN concentration time and urinary PAGN amount-time data following treatment were analyzed simultaneously using nonlinear mixed effects modelling [7]. An interesting finding from this analysis was that the pharmacokinetic behavior of GPB was similar across all age groups, including patients aged 2 months to 2 years (data from Study 1) [7]. Due to the limited number of patients in Study
ACCEPTED MANUSCRIPT 1, the data from Study 2 allows further understanding of how GPB is absorbed, metabolized and excreted in this patient population. The presence of plasma concentrations of PBA, PAA and PAGN as well as urinary concentrations of PAGN in the 24-hour analyte analysis adds further evidence of enzymatic cleavage of GPB in this population to yield the active moiety, PAA. Overall, concentrations of PBA, PAA and PAGN overlaid with the observations in previous studies in the adult population found similar exposure to these analytes, further supporting the similarity in PK behavior of GPB in patients aged 2
T
months to 2 years.
IP
Using the PAA:PAGN ratio as a proxy for the efficiency with which an individual converts PAA to
CR
PAGN and a predictor of patients at risk of having an elevated PAA level, a median ratio of <1 (as found in the long term PK analyses in this population) indicate a low risk of high PAA levels (>500 µg/mL) [13]. Previous analyses showed considerable variability in PAA levels over a 24-hr period in patients > 2
US
years regardless of the dose and enzyme deficiency and this also holds true in this patient population. BSA was found to be the most significant covariate for GPB analytes in infants 2 to <24 months old
AN
(similar to the adult population), suggesting that the current dosing regimen based on BSA is reasonable. Based on the 10 patients in Study 2, age does not appear to influence PK analytes except for the single
M
subject older than 14 months for whom the values appear consistently higher than the rest of the group and may be driven by this subject’s relatively high BSA. The doses of GPB (absolute or by BSA) do not
ED
appear to correlate with the main PK parameters for any analyte. The patients reported here had large inter-individual variability in their PK analytes as demonstrated by
PT
the inter-subject coefficient of variation (CV%), which has been previously demonstrated in older UCD populations. The variability in this analysis was primarily due to the sampling time intervals being wide
CE
and lack of accurate accounting for actual dosing time, which is a management challenge in infants with UCD. As previously alluded to by Berry, et al., there are physiologic limitations on blood draws in
AC
infants; thus data was not available at some time points [11]. In addition to the small number of patients, this constitutes one of the limitations of this analysis.
Another contributing factor to the variability
between subjects is that the conjugating enzyme (glutamine-N-phenylacetyl transferase) has never been purified, and it is not known how much inter-individual variability there is in the kinetics of the enzymatic reaction. These data support that, within the dosing range used here, GPB could be safely used by all individuals tested. Another limitation of this analysis was the sampling time in Study 2, which was the first 24 hours of GPB treatment. This does not allow sufficient time for steady state levels of GPB to be achieved, which may result in inconsistencies in the PK parameters such as T max. Another limitation of this analysis was the sampling time in Study 2, which was the first 24 hours of GPB treatment. This does not allow sufficient time for steady state levels of GPB to be achieved, which may result in
ACCEPTED MANUSCRIPT inconsistencies in the PK parameters such as T max. However, PK parameters were collected out to 12 months in many patients and detailed PK samples were taken at Day 10 in Study 1, at steady state. Though not all patients were transitioned from NaPBA 14 of 17 were, moderating the variability that could be observed as patients entered the study with differing levels of plasma PBA, PAA and PAGN. These limitations are, however, exceeded by the overall consistent direction of the results, which indicate that glycerol phenylbutyrate use in UCD patients 2 to < 24 months of age is effective and behaves much
Conclusions
IP
5.
T
the same as that in older patients.
CR
There was no evidence for decreased metabolism of GPB in these infants as dosing was followed by prompt appearance (first day) of metabolites (PBA, PAA) in blood and PAGN in urine. Elevated PAA levels were infrequent and transient; the median PAA:PAGN ratio of <1 in the long-term PK analyses
US
suggests that PAA to PAGN conversion was not saturated. Analyte values were generally within the range of that seen in older patients and, as observed in older patients, BSA was found to be the most
AN
significant covariate, reinforcing that dosing should be based on body mass. These results collectively indicate that pediatric UCD patients aged 2 months–2 years effectively hydrolyze GPB to release PBA
M
and convert PAA to UPAGN so as to maintain blood ammonia within normal limits. The observation that the conjugating enzyme was not saturated at any dosing level used in the study suggests that the safe
ED
dosing range may actually be higher than currently reported, and that the PK parameters used here could be of benefit in assessment of patients in whom higher doses might be considered.
PT
Conflicts of interest statement
Horizon Pharma funded the development, conduct, and analysis of the studies. B. Robinson, R. Perdok,
CE
and R Holt employees of, and have stock in, Horizon Pharma USA, Inc. None of the other authors has a financial interest in Horizon Pharma USA, Inc.
AC
For the following authors, payments were made by Horizon Pharma to their institutions for services provided in the conduct of the clinical studies upon which this report is based: George Diaz (Icahn School of Medicine at Mount Sinai), Uta Lichter-Konecki (Children's National Medical Center and Children's Hospital of Pittsburgh), Shawn McCandless (University Hospitals Cleveland Medical Center), Cary O. Harding (Oregon Health & Science University), Nicola Longo (University of Utah), Roberto Zori (University of Florida), Can Ficicioglu (Children's Hospital of Philadelphia), Wendy E. Smith (Maine Medical Center), Jerry Vockley (Children's Hospital of Pittsburgh) and Susan Berry (University of Minnesota).
ACCEPTED MANUSCRIPT Can Ficicioglu has provided consulting support to and received grant support and honoraria for speaking engagements from BioMarin, Abbott Laboratories, Alexion, Shire, Pfizer, Swedish Orphan (Sobi), Sanofi-Genzyme, and Horizon Pharma (previously Hyperion Therapeutics, Inc). For M. Dong, the Cincinnati Children’s Hospital Medical Center has received research funds from Horizon Pharma to conduct this research. A.A. Vinks is a consultant for the sponsor through an agreement between Horizon Pharma and Cincinnati Children’s Hospital Medical Center.
T
Susan Berry, Nicola Longo, George Diaz, Shawn McCandless, Wendy E. Smith, Roberto Zori, Can
IP
Ficicioglu and Jerry Vockley have served as Horizon Pharma Advisory Board Members.
CR
Acknowledgments
Research at Oregon Health & Science University (Cary Harding) was made possible with support from
US
the Oregon Clinical and Translational Research Institute (OCTRI), grant number UL1TR000128 from the National Center for Advancing Translational Sciences (NCATS), a component of the National Institutes
AN
of Health (NIH), and NIH Roadmap for Medical Research.
The authors would like to acknowledge the following study coordinators: Carrie Bailey BS, CCRC,
M
Michele Bergman RN, BSN, Jenny Billy, Sarah Couchon, Sara Elsbecker MS, APRN, CNP, Luca Fierro MS, CGC, Christel Gross RN, Michelle Hunter, Deborah Kawchak MS, RD,
ED
CCRP, Margaret Kuenzi, Angela Leshinski MBA, RD, LDN, Jessica Lindenberger RN, Susan Mortenson RN, CCRC, Hadley Morotti, Audrey Lynn, and Kim Wallis MS, LGC. In addition, the authors would like
PT
to thank Megan Francis-Sedlak PhD, and Teresa M.Y. Kok RPh, MBA from Horizon Pharma for data
References
CE
analysis, writing and editorial support.
Batshaw ML. Hyperammonemia. Current problems in pediatrics 1984 Nov;14(11):1-69.
2.
Berry SA, Lichter-Konecki U, Diaz GA, McCandless SE, Rhead W, Smith W, et al. Glycerol
AC
1.
phenylbutyrate treatment in children with urea cycle disorders: pooled analysis of short and long-term ammonia control and outcomes. Molecular genetics and metabolism 2014 May;112(1):17-24. 3.
Diaz GA, Krivitzky LS, Mokhtarani M, Rhead W, Bartley J, Feigenbaum A, et al. Ammonia
control and neurocognitive outcome among urea cycle disorder patients treated with glycerol phenylbutyrate. Hepatology (Baltimore, Md) 2013 Jun;57(6):2171-9. 4.
Lee B, Rhead W, Diaz GA, Scharschmidt BF, Mian A, Shchelochkov O, et al. Phase 2
comparison of a novel ammonia scavenging agent with sodium phenylbutyrate in patients with urea cycle
ACCEPTED MANUSCRIPT disorders: safety, pharmacokinetics and ammonia control. Molecular genetics and metabolism 2010 Jul;100(3):221-8. 5.
Nagamani SC, Diaz GA, Rhead W, Berry SA, Le Mons C, Lichter-Konecki U, et al. Self-reported
treatment-associated symptoms among patients with urea cycle disorders participating in glycerol phenylbutyrate clinical trials. Molecular genetics and metabolism 2015 Sep-Oct;116(1-2):29-34. 6.
Smith W, Diaz GA, Lichter-Konecki U, Berry SA, Harding CO, McCandless SE, et al. Ammonia
T
control in children ages 2 months through 5 years with urea cycle disorders: comparison of sodium
Monteleone JP, Mokhtarani M, Diaz GA, Rhead W, Lichter-Konecki U, Berry SA, et al.
CR
7.
IP
phenylbutyrate and glycerol phenylbutyrate. The Journal of pediatrics 2013 Jun;162(6):1228-34, 34.e1.
Population pharmacokinetic modeling and dosing simulations of nitrogen-scavenging compounds:
US
disposition of glycerol phenylbutyrate and sodium phenylbutyrate in adult and pediatric patients with urea cycle disorders. Journal of clinical pharmacology 2013 Jul;53(7):699-710. Longo N, Holt RJ. Glycerol phenylbutyrate for the maintenance treatment of patients with
AN
8.
deficiencies in enzymes of the urea cycle. Expert Opinion on Orphan Drugs 2017;5(12):999-1010. Thibault A, Cooper MR, Figg WD, Venzon DJ, Sartor AO, Tompkins AC, et al. A phase I and
M
9.
pharmacokinetic study of intravenous phenylacetate in patients with cancer. Cancer research 1994 Apr
10.
ED
01;54(7):1690-4.
Thibault A, Samid D, Cooper MR, Figg WD, Tompkins AC, Patronas N, et al. Phase I study of
11.
PT
phenylacetate administered twice daily to patients with cancer. Cancer 1995 Jun 15;75(12):2932-8. Berry SA, Longo N, Diaz GA, McCandless SE, Smith WE, Harding CO, et al. Safety and
CE
efficacy of glycerol phenylbutyrate for management of urea cycle disorders in patients aged 2 months to 2 years. Molecular genetics and metabolism 2017;122:46-53. McGuire BM, Zupanets IA, Lowe ME, Xiao X, Syplyviy VA, Monteleone J, et al. Pharmacology
AC
12.
and safety of glycerol phenylbutyrate in healthy adults and adults with cirrhosis. Hepatology (Baltimore, Md) 2010 Jun;51(6):2077-85. 13.
Mokhtarani M, Diaz GA, Rhead W, Berry SA, Lichter-Konecki U, Feigenbaum A, et al. Elevated
phenylacetic acid levels do not correlate with adverse events in patients with urea cycle disorders or hepatic encephalopathy and can be predicted based on the plasma PAA to PAGN ratio. Molecular genetics and metabolism 2013 Dec;110(4):446-53.
Figure 1
Figure 2
Figure 3