Available online at www.sciencedirect.com
`
`SCIENCE @DIRECT®
`
`ELSEVIER
`
`Molecular Genetics and Metabolism 81 (2004) S67-S73
`
`Molecular Genetic~
`and Metabolism
`
`www.elsevier.com/locate/ymgme
`
`Pharmacokinetics of sodium phenylacetate and sodium
`benzoate following intravenous administration as both a bolus
`and continuous infusion to healthy adult volunteers
`
`Robert B. MacArthur, a·* Arman Altincatal,a and Mendel Tuchmanb
`
`a CPMC Research Pharmacy, Columbia University and New York State Psychiatric Institute Pharmacy, New York, NY 10032, USA
`b Children's Research Institute, Children's National Medical Center, The George Washington University, Washington, DC, USA
`
`Received 7 October 2003; received in revised form 15 December 2003; accepted 19 December 2003
`
`Abstract
`
`Background: Ammunol® (sodium phenylacetate/sodium benzoate) is an intravenously administered, investigational drug used for
`the treatment of acute hyperammonemia in infants, children, and adults with urea cycle enzyme deficiencies. A pharmacokinetic
`study of sodium phenylacetate/sodium benzoate (NAP A/NABZ) was performed in two groups of normal healthy volunteers, follow(cid:173)
`ing the dosing regimen used to treat hyperammonemia.
`Methods: The first group of subjects (n = 3) received a bolus dose of 5.5 g/m2 of NAP A/NABZ, over a period of 1.5 h. Following a
`seven-day washout, subjects then received the same bolus dose, followed by a continuous infusion of 5.5 g/m2 over 24 h. A second
`group of different subjects (n = 17) received the same treatment regimen, but using doses of 3.75 g/m2
`. Phenylacetate (PA) and benzo(cid:173)
`ate (BZ), and their respective metabolites, phenylacetylglutamine (P AG), and hippurate (HIP) were measured over a 24-h period. An
`HPLC method was used for the measurement of all analyte concentrations. Non-compartmental analysis and modeling was per(cid:173)
`formed using WinNonlin Professional®.
`Results: Both BZ and PA displayed saturable, non-linear elimination, with a decrease in clearance with increased dose. During the
`bolus dose with continuous infusion regimen, plasma levels of both BZ and P A peaked at the end of the priming dose, and P A levels
`remained near peak for 5- 9 h. In contrast, BZ plasma levels immediately fell following the priming dose, and became undetectable at
`14.1 ±4.2 and 26.8 ± 2.3 h in the low- and high-dose group, respectively. The formation of HIP occurred more rapidly than that of
`PAG. For both PA and BZ, metabolite formation increased in a linear fashion with the dose.
`Conclusion: These data describe the pharmacokinetics of PA and BZ, and their respective metabolites, as observed in healthy
`adult volunteers, with the higher dose studied equivalent to that used to treat hyperammonemia. Dose optimization is required to
`maximize nitrogen removal, while minimizing the risk of toxicity, especially due to PA. Because of the slower elimination ofPA, and
`the non-linear pharmacokinetic behavior displayed by both P A and BZ, only investigational protocol-specific doses should be used,
`and higher doses should be avoided unless blood level monitoring can be done promptly and frequently.
`© 2004 Elsevier Inc. All rights reserved.
`
`Keywords: Ammonul®; Pharmacokinetics; Sodium phenylacetate; Sodium benzoate; Intravenous administration; Hyperammonemia; Alternative
`pathway therapy; Urea cycle; Adult volunteers
`
`Introduction
`
`investigational combination drug product
`The
`(Ammonul®) containing 10% sodium phenylacetate
`(NAPA) and 10% sodium benzoate (NABZ) has been
`
`*Corresponding author. Fax: 1-212-543-5651.
`E-mail address: rbml7@columbia.edu (R.B. MacArthur).
`
`1096-7192/$- see front matter© 2004 Elsevier Inc. All rights reserved.
`doi: 10.1016/j.ymgme.2003.12.011
`
`used for the intravenous treatment of acute hyperam(cid:173)
`monemia in patients with urea cycle enzyme deficiencies
`since the mid-1980s. The product is diluted in 10% dex(cid:173)
`trose and administered as a 90-min bolus dose infusion,
`followed by a 24-h maintenance dose. Both NAPA and
`NABZ remove nitrogen via alternate non-urea cycle
`enzymatic pathways, and are the only intravenous
`ammonia scavenger agents currently available for
`clinical use.
`
`Par Pharmaceutical, Inc. Ex. 1010
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 1 of 7
`
`

`

`S68
`
`R.B. MacArthur eta/. I Molecular Genetics and Metabolism 81 (2004) S67-S73
`
`In patients with limited or no urea synthesizing
`capacity, acute hyperammonemia is a medical emergency
`that can require multiple medical interventions depend(cid:173)
`ing on severity, and including, dialysis, administration of
`ammonia scavenger agents and urea cycle intermediates
`(arginine or citrulline), protein restriction and high calo(cid:173)
`ric intake to counteract catabolism. The specific regi(cid:173)
`mens used vary with patient presentation and specific
`urea cycle enzyme deficiency, and comprehensive
`reviews are available [1].
`The individual pharmacokinetics of both NAPA and
`NABZ have been described in the literature [2-6], and the
`published pharmacokinetic parameters are summarized in
`Tables 1 and 2. To date, there has been no published work
`describing the pharmacokinetics of intravenous NAPA
`and NABZ, when co-administered in the same manner as
`
`Table 1
`Published pharmacokinetic parameters of sodium phenylacetate
`
`is currently used to treat hyperammonemia. To better
`understand the pharmacokinetics of this combination
`drug product, a study was undertaken in normal healthy
`volunteers utilizing the dosage regimen recommended for
`treating hyperammonemia in urea cycle enzyme-deficient
`patients.
`
`Methods
`
`Study design
`
`This was a prospective, open-label pharmacokinetic
`study of intravenous NAPA and NABZ, administered to
`normal healthy volunteers. It was conducted at a single
`
`Population
`
`18 Adults with cancer
`
`17 Adults with cancer
`
`Route of administration and
`dose
`
`Model
`
`Intravenous NAPA at 125
`and 150 mg/kg, with some
`additional dose escalation
`
`Intravenous NAPA priming
`dose followed by prolonged,
`escalating continuous
`infusion
`
`One-compartment model
`with Michaelis-Menten
`elimination noted that
`NAPA may induce its own
`elimination comparing day 1
`vs day 12-14 AUC
`Non-linear kinetics
`
`Parameter estimates
`
`V max = 29 ± 6.3 mg/kg/h
`Km = 106 ± 22 Jlg/ml
`Vd=21±4.81L
`
`vmax = 24.1 ± 5.2 mg/kglh
`Km = 105.1 ± 44.5 Jlg/ml
`vd = 19.2 ± 3.3 L
`
`Ref.
`
`[2]
`
`[3]
`
`Table 2
`Published pharmacokinetic parameters of sodium benzoate
`
`Population
`
`Route of administration and
`dose
`
`Model
`
`2 Patients with
`hyperammonemia
`
`Oral NABZ 130 and
`150mg/kg
`
`6 Healthy adult volunteers
`
`Oral benzoic acid at 3 doses
`(40, 80, 160 mg/kg) 3 way
`cross-over
`
`4 Neonates with
`hyperammonemia
`
`Intravenous NABZ
`3.5 mmol/kg/day divided q6h
`
`Post absorption data
`One-compartment model
`with Michaelis-Menten
`elimination
`
`Benzoate
`One-compartment model
`with first-order rate
`absorption and Michaelis(cid:173)
`Menten elimination
`Hippurate
`One-compartment model
`with first-order elimination
`Assumed first-order, one
`compartment elimination
`model
`
`Ref.
`
`[4]
`
`[5]
`
`[6]
`
`Parameter estimates
`
`Period 1
`vmax = 152 Jlg/mllh
`Km 65.7 Jlg/ml
`Period2
`V max = 90 Jlg/ml/h
`Km 30 Jlg/ml
`vmax = 101.9 Jlg/mllh
`Km = 10.5 Jlg/ml
`Cmax
`40 mg/kg = 99.7 Jlg/ml
`80 mg/kg = 202.8 Jlg/ml
`160mg/kg = 336.5 Jlg/ml
`
`Large inter-patient variability,
`low benzoate clearance associated
`with impaired renal function and
`hypotension
`vd = o.14 ± o.o7 Llkg
`(0.086-0.244)
`t 112 = 2.8 ± 3.1 h (0.75-7.4)
`C1 = 1.00 ± 0.61 ml/kg/min
`(0.33-1.56)
`
`Par Pharmaceutical, Inc. Ex. 1010
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 2 of 7
`
`

`

`R.B. MacArthur eta/. I Molecular Genetics and Metabolism 81 (2004) S67-S73
`
`S69
`
`clinical research center, and was approved by the local
`institutional review board.
`
`Results
`
`Subjects
`
`Two volunteer cohorts were studied, each undergoing
`two separate treatment periods. The first cohort (n = 3),
`during period 1, received 5.5g/m2 ofNAPNNABZ as a
`90-min infusion, with blood sampling at 0, 0.25, 0.5, 0.75,
`1.5, 2, 2.5, 3.5, 4.5, 5.5, 6.5, 7.5, 8.5, 9.5, 10.5, 11.5, and
`12.5 h. After a one-week washout, during the second
`treatment period, a 90-min bolus dose of 5.5 g!m2 was
`administered, followed by a 24-h continuous infusion of
`the same dose. Blood was sampled at 0, 0.25, 0.75, 1.5,
`2.5, 4.5, 6.5, 8.5, 10.5, 12.5, 16.5, 25.5, 26, 26.5, 27, 27.5, 28,
`28.5, 29.5, 30.5, and 32.5 h. The second cohort of subjects
`(n = 17) underwent an identical sequence of drug admin(cid:173)
`istration and sampling; only, in this group, a dosage of
`3.75 g/m2 was used for each of the combination drugs.
`Infusions were administered via a peripheral venous
`catheter.
`
`Sample handling and analytics
`
`Ten milliliters of venous blood was drawn into
`sodium heparin vacutainers. They were centrifuged at
`l600g at a temperature of 4°C for 10-15min. The
`plasma was separated into two storage tubes and frozen
`at -20 °C. Following precipitation with methanol and
`addition of internal standard the plasma samples were
`analyzed by reverse phase high-pressure liquid chroma(cid:173)
`tography using ultraviolet absorption detection. For the
`analytes, phenylacetic acid, and benzoic acid, the calibra(cid:173)
`tion curve ranged from 5.000 to 500.0 J..Lg/mL, and for the
`metabolites phenylacetylglutamine (PAG), and hippuric
`acid (HIP) the calibration curve ranged from 5.000 to
`500.0 J..Lg/mL.
`
`Pharmacokinetic analysis
`
`Ucyclyd Pharma (Scottsdale, AZ) supplied analyte
`concentration data as Microsoft Excel® 4.0 spreadsheets.
`Data were analyzed using WinN onlin Professional®
`(version 4.1). The concentration:time data for phenylace(cid:173)
`tate and benzoate for each dose group (periods 1 and 2
`for each cohort) were analyzed using non-compartmen(cid:173)
`tal methods, and the results are presented as treatment
`group means. For each active drug and its metabolite,
`the following model-independent parameters were calcu(cid:173)
`lated: AUC, Cmax• and T max· Data from each cohort and
`treatment period were then modeled individually, using
`WinNonlin Michadis--Menlen model 302 (one-comparl(cid:173)
`ment with constant IV input and Michaelis-Menten out(cid:173)
`put), with fitting via the Gauss-Newton (Levenberg and
`Hartley) algorithm.
`
`The first cohort of three volunteers received 5.5 g!m2
`,
`initially as a single bolus dose (period 1 ), and then as a
`bolus dose, followed by a continuous infusion (period 2).
`During the second treatment period, significant nausea,
`vomiting, and somnolence occurred, and as a result no
`additional volunteers were treated at this dose level. The
`dose was then lowered to 3.75 g/m2
`; 17 volunteers
`received this bolus dose (period 1), and 14 of them
`received the combined bolus followed by the continuous
`infusion (period 2).
`Time-concentration curve graphs of the bolus dose
`followed by continuous infusion are provided for phen(cid:173)
`ylacetate and PAG (Figs. 1 and 2), as are graphs for ben(cid:173)
`zoate and HIP (Figs. 3 and 4). Non-compartmental
`analysis results for each compound and its correspond(cid:173)
`ing metabolite are provided (Tables 3 and 4). Because
`the bolus dose followed by continuous infusion mimics
`the manner in which the drug is infused according to
`the investigational drug protocol, only these modeled
`data are presented here. Graphs of modeled data are
`provided (Figs. 5 and 6) for phenylacetate and benzoate
`respectively.
`
`Sodium phenylacetate and phenylacetylglutamine
`
`Following the bolus dose alone (period 1), plasma
`levels of phenylacetate peaked at the end of the 1.5-h
`infusion, with a maximum plasma concentration of
`2225 (± 309) and 3022 (± 220) J..Lmol/L, in the low- and
`high-dose groups, respectively (Table 3).
`When the bolus dose was followed by a 24-h contin(cid:173)
`uous infusion, a plateau phase was observed following
`
`T----------------
`
`Mean (SD) Phenylacetate and PAG Plasma
`Levels Following 3 75 g/m2 Bolus and 24 hr
`lnfus1on, 1n Normal Healthy Volunteers
`
`- • Phenylacetate
`-io- Phenylacetylglutamine
`"' Bolus Dose
`--Constant Infusion
`
`4000
`
`3500
`
`3000
`
`2500
`
`~ 2000
`0
`E 1500
`
`::l.
`
`1000
`
`500
`
`0
`
`0
`
`5
`
`10
`
`15
`20
`Hours
`
`25
`
`30
`
`35
`
`Fig. 1. Mean phenylacetate and PAG plasma levels at the 3.75 g!m2
`dose level.
`
`Par Pharmaceutical, Inc. Ex. 1010
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 3 of 7
`
`

`

`S70
`
`R.B. MacArthur eta/. I Molecular Genetics and Metabolism 81 (2004) S67-S73
`
`6000
`
`5000
`
`4000
`
`<::! 3000
`0
`E
`='- 2000
`
`1000
`
`0
`
`T------------------
`
`Mean (SO) Phenylacetate and PAG Plasma
`Levels Following 5.5 g/m 2 Bolus and 24 hr
`Infusion, in Normal Healthy Volunteers
`
`ll 1 I I -
`-- - ~
`j
`
`I
`
`-•- Phenylacetate
`r
`Phenylacetylglutamine
`.,. Bolus Dose
`'~ I~----,,- Constant Infusion
`
`~.,
`
`~"-."ITk
`---~~Iii~
`
`T------------------
`
`Mean (SD) Benzoic and Hippuric Acid Plasma
`Levels Following 5.5 g/m 2 Bolus and 24 hr
`Infusion, in Normal Healthy Volunteers
`
`-•- Benzoic Acid
`Hippuric Acid
`.,. Bolus Dose
`--Constant Infusion
`
`6000
`
`5000
`
`4000
`
`3000
`
`--'
`'5
`E 2000
`=<.
`
`1000
`
`0
`
`0
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`0
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`Hours
`
`Hours
`
`Fig. 2. Mean phenylacetate and PAG plasma levels at the 5.5 g/m2 dose
`level.
`
`Fig. 4. Mean benzoic and hippuric acid plasma levels at the 5.5 g/m2
`dose level.
`
`T-----------------
`
`Mean (SO) Benzoic and Hippuric Acid Plasma
`Levels Following 3. 75 gfm2 Bolus and 24 hr
`Infusion, in Normal Healthy Volunteers
`
`-•- Benzoic Acid
`-6 , - Hippuric Acid
`.,. Bolus Dose
`---Constant Infusion
`
`4000
`
`3500
`
`3000
`
`2500
`
`--' 2000
`'5
`E 1500
`
`::t
`
`1000
`
`500
`
`0
`
`5500
`
`5000
`
`4500
`
`4000
`
`3500
`
`3000
`
`=:::
`0 2500
`E
`::l.. 2000
`
`1500
`
`1000
`
`500
`
`0
`
`Mean Phenylacetate Plasma Levels versus Modeled
`Concentrations Following 3. 75 g/m2 and 5.5 g/m2 Bolus
`and 24 hr Infusions, in Normal Healthy Volunteers
`• Measured Phenylacetate ~ 5.5 g/m 2 Bol + CIVI
`-Michaelis-Menten Model
`• Measured Phenylacetate- 3.75 g!m 2 Bol + CIVI
`-Michaelis-Menten Model
`
`0
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`0
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`Hours
`
`Hours
`
`Fig. 3. Mean benzoic and hippuric acid plasma levels at the 3.75 g/m 2
`dose level.
`
`Fig. 5. Modeled phenylacetate concentrations at the 3.75 g/m2 and
`5.5 g/m2 dose levels.
`
`the end of the bolus dose, during which phenylacetate
`concentrations remained essentially unchanged for 5
`and 9 h, respectively. In the high-dose group, the mean
`T max occurred at 7.2 (± 5.0) h, well after the initial bolus
`dose and Cmax was 2261 (± 272) and 3422 (± 365) Jl.mol/
`L, in the low- and high-dose groups, respectively
`(Table 3). Thereafter, plasma level concentrations
`declined over the remaining 24-h period, during the
`ongoing continuous infusion. For phenylacetate, AUC
`increased disproportionately with dose, while
`this
`parameter for P AG was approximately dose-propor(cid:173)
`tional (Fig. 7). PAG was not detected until 0.75 h after
`the start of the infusion, and plasma levels increased
`slowly, with T max occurring at 8.8 (± 1.7) and 19.5
`(± 5.2) h, in the low- and high-dose continuous infusion
`groups, respectively.
`
`Sodium benzoate and hippurate
`
`In all dose groups, peak plasma levels of be=oate
`occurred at the end of the 1.5-h infusion (Table 4). During
`period 1, Cmax was 2136 (±319) and 3444 (±286) and
`during period 2,
`it was 2182 (± 298) and 3746
`(± 593) Jl.mol/L, in the low- and high-dose groups, respec(cid:173)
`tively. Also, in all dose groups NABZ concentrations
`declined rapidly following T max, and approached the limit
`of detection in these treatment groups at 6.5 and 12.5 h,
`respectively (Figs. 3 and 4). In contrast to P AG, plasma
`levels of HIP were detectable at the first sampling point,
`15 min following the start of the bolus dose, in all groups.
`As with PAG, the AUCs of HIP increased in proportion
`to the dose (Fig. 7), while the AUC for the parent drug,
`be=oate, increased disproportionately with dose. This
`
`Par Pharmaceutical, Inc. Ex. 1010
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 4 of 7
`
`

`

`R.B. MacArthur eta/. I Molecular Genetics and Metabolism 81 (2004) S67-S73
`
`S71
`
`Differences in AUC/D by dose group, in the continuous
`infusion groups
`
`5000.-----------------------------------------,
`Mean Benzoic Acid Plasma Levels versus Modeled
`4500 Concentrations Following 3.75 g/m 2 and 5.5 g/m2 Bolus
`and 24 hr Infusions, in Normal Healthy Volunteers
`
`4000
`
`• Measured Benzoic Acid - 5.5 g/m 1 Bol + CIVI
`-Michaelis-Menten Model
`• Measured Benzoic Acid 3.75 g/m" Bal CIVI
`-Michaelis-Menten Model
`
`3500
`
`3000
`
`2500
`
`~
`E
`='- 2000
`1500
`
`1000
`
`500
`
`0
`
`13000
`
`12000
`
`11000
`
`10000
`
`~
`(.)
`::::J
`
`<{ .
`
`9000
`
`8000
`
`7000
`
`6000
`
`5000
`
`4000
`
`3000
`
`2000
`
`1000
`
`0
`
`0
`
`5
`
`10
`
`15
`Hours
`
`20
`
`25
`
`30
`
`35
`
`"AU C/O units are in 11.mol x hrx m='/g/1
`
`3.75 5.5
`NAPA
`
`3.75 5.5
`PAG
`
`3.75 5.5
`NABZ
`Dose (g/m2
`
`)
`
`3 75 55
`HIP
`
`Fig. 6. Modeled benzoic acid concentrations at the 3.75 g/m2 and 5.5
`g/m2 dose levels.
`
`Fig. 7. Comparison of parent drug and metabolite AUCs at each dose
`level.
`
`Table 3
`Pharmacokinetic parameters for PA and PAG determined using
`non-compartmental methods
`NAPA dose (g/m 2)
`
`Table 4
`Pharmacokinetic parameters for BZ and HIP determined using
`non-compartmental methods
`NABZ dose (g/m 2
`
`)
`
`3.75 (n = 20)
`
`5.5 (n = 3)
`
`3.75 (n
`
`20)
`
`5.5 (n
`
`3)
`
`PK parametera
`Bolus Period
`PA
`Cmax (11mol/L)
`Tmax (h)
`AUC (11mol/L x h)
`PAG
`Cmax (11mol/L)
`Tmax (h)
`AUC (11mol/L x h)
`
`2225 (309)
`1.7 (0.3)
`15004 (2445)
`
`281 (42)
`7.9 (0.9)
`2047 (300)
`
`Continuous IV infusion period
`PA
`Cmax (11mol/L)
`Tmax (h)
`AUC (11mol/L x h)
`PAG
`Cmax (11mol/L)
`Tmax (h)
`AUC (11mol/L x h)
`
`2261 (272)
`2.0 (1.1)
`25970 (5041)
`
`333 (50)
`8.8 (1.7)
`5099 (858)
`
`3022 (220)
`1.7 (0.3)
`28156 (3791)
`
`NR
`NR
`NR
`
`3422 (365)
`7.2 (5.0)
`61560 (7834)
`
`462 (58)
`19.5 (5.2)
`9213 (2120)
`
`PK parametera
`Bolus period
`BZ
`Cmax (11mol/L)
`Tmax (h)
`AUC (11mol/L x h)
`HIP
`Cmax (11mol/L)
`Tmax (h)
`AUC (11mol/L x h)
`
`2136 (319)
`1.5 (0.0)
`4666 (859)
`
`351 (65)
`3.0 (0.5)
`1330 (278)
`
`Continuous IV infusion period
`BZ
`cmaxCilmol/L)
`Tmax (hr)
`AUC (11mol/L x h)
`HIP
`CmaxCilmol/L)
`Tmax (h)
`AUC (11mol/L x h)
`
`2182 (298)
`1.5 (0.0)
`5949 (1150)
`
`323 (69)
`3.4 (1.0)
`2666 (640)
`
`3444 (286)
`1.7 (0.3)
`13216 (3827)
`
`435 (113)
`5.2 (0.6)
`2912 (257)
`
`3746 (593)
`1.5 (0.0)
`20430 (7255)
`
`419 (120)
`5.8 (1.2)
`4725 (1284)
`
`NR, not reported due to limited 12 h sampling period.
`a Values are mean (SD).
`
`a values are mean (SD).
`
`relationship between sodium benzoate dose, plasma ben(cid:173)
`zoate and hippurate, and urinary hippurate has been pre(cid:173)
`viously described [5]. HIP concentrations reached Cmax
`more quickly than those of P AG, and appeared to plateau
`at 10.5 h during period 2, remaining at approximately
`0.2 Jlmol/L until the end of the continuous infusion.
`
`Mode led parameters
`
`Michaelis Menten parameters for pheny1acetate and
`benzoate derived from the continuous infusion regimens
`are provided in Tables 5 and 6. In the 5.5 g/m2 dose
`group, the following parameters (mean (SEM)) were
`
`derived for phenylacetate, vd 10.27 (0.3439)Lim2
`, vmax
`0.2780 (0.03046) Jlmol!m2/h, and km 67.55 (135.6) Jlmol/L,
`and for benzoate, Vct 9.722 (0.4179) Llm2
`, Vmax 0.7417
`(0.1 064) Jlmol/m2/h, and km 304.1 (250.0) Jlmol/L. Similar
`values were computed for the respective lower-dose
`groups. The data fit the model well. The largest variabil(cid:173)
`ity in individual parameters was seen in Km.
`
`Discussion
`
`When used for the treatment of hyperammonemia,
`the objective of NAPNNABZ therapy is to bind and
`
`Par Pharmaceutical, Inc. Ex. 1010
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 5 of 7
`
`

`

`S72
`
`R.B. MacArthur eta/. I Molecular Genetics and Metabolism 81 (2004) S67-S73
`
`Table 5
`Model-dependant PA pharmacokinetic parameters from continuous infusion regimens
`
`3.75
`
`5.5
`
`Parameter
`vd
`vmax
`Km
`
`vd
`vmax
`Km
`
`Units
`Llm2
`mmol/m2/h
`Jlmol/L
`
`Llm2
`mmollm2/h
`Jlmol/L
`
`Estimate
`
`10.03
`0.2961
`254.9
`
`10.27
`0.2780
`67.55
`
`SEM
`
`0.1355
`0.01385
`61.73
`
`0.3439
`0.03046
`135.6
`
`SEM
`
`0.1427
`0.09821
`163.4
`
`0.4179
`0.1064
`250.0
`
`cv (%)
`1.35
`4.68
`24.22
`
`3.35
`10.95
`200.73
`
`cv (%)
`1.43
`10.3
`31.49
`
`4.3
`14.34
`82.8
`
`Table 6
`Model-dependant BZ pharrnacokinetic parameters from continuous infusion regimens
`Dose (g/m2
`)
`3.75
`
`Parameter
`vd
`vmax
`Km
`
`Units
`Llm2
`mmollm2/h
`Jlmol/L
`
`Estimate
`
`9.949
`0.9539
`518.8
`
`5.5
`
`vd
`vmax
`
`Llm2
`mmol/m2/h
`Jlmol/L
`
`9.722
`0.7417
`304.1
`
`excrete as much nitrogen as possible, and as quickly as
`possible, in order to offset excessive ammonia produc(cid:173)
`tion, and to return ammonia levels to as near normal as
`possible. Therefore, the rate and extent of formation of
`the respective nitrogen-containing byproducts, PAG and
`HIP, becomes quite important. Also important is main(cid:173)
`taining the plasma levels of phenylacetate and benzoate
`below the levels associated with toxicity, while providing
`enough of these scavenging agents to maximize waste
`nitrogen removal.
`
`Phenylacetate and benzoate clearance and toxicity
`
`These compounds have similar chemical structures,
`molecular weights (Fig. 8 displays the sodium salts), and
`solubility profiles. Initial peak levels achieved following
`
`Sodium Benzoate
`C6HsC02Na
`MW 144.1
`
`Sodium Phenylacetate
`CsH7Na02
`MW 158.1
`
`Fig. 8. Chemical structures of sodium phenylacetate and sodium
`benzoate.
`
`a bolus infusion are nearly identical, indicating that the
`volumes of distribution are comparable. The clearance
`of phenylacetate appears to be much slower and, unlike
`benzoate, clearance can become saturated at the plasma
`levels attained with doses used to treat hyperammone(cid:173)
`mia. When used in oncology protocols, phenylacetate is
`administered as twice-daily bolus doses of 150 mg/kg,
`and one objective of therapy is to maintain serum
`concentrations above 1850 J-Lmol/L (250 J-Lg/mL). In this
`setting, grade 1 and grade 3 neurologic toxicities (som(cid:173)
`nolence, fatigue, headache, lightheadedness, dysgeusia)
`have been associated with phenylacetate concentrations
`of 3694-7520 J-Lmol/L [2]. Similarly, in our study, the
`high-dose regimen was poorly tolerated by the normal
`volunteers due to nausea, vomiting, and somnolence,
`and peak plasma phenylacetate concentrations ap(cid:173)
`proached the lower end of the above range. Although
`patients may become more tolerant to repeat infusions,
`it is possible that keeping the plasma phenylacetate lev(cid:173)
`els below those observed in the high-dose group would
`lessen the risk of these adverse reactions. Thus, adminis(cid:173)
`tration of phenylacetate needs to be optimized to lessen
`the risk of attaining inappropriately high plasma phen(cid:173)
`ylacetate levels, while maximizing its conversion to P AG.
`In contrast, benzoate was rapidly cleared, and plasma
`levels did not display a plateau phase. Information con(cid:173)
`cerning the relationship between dose, plasma concen(cid:173)
`trations and toxicity for benzoate is less available than
`for phenylacetate. Doses of sodium benzoate alone,
`above 500 mg/kg/day have been administered in a differ(cid:173)
`ent clinical setting for the management of seizures in
`non-ketotic hyperglycinemia [7]. Anorexia and vomiting
`were reported at "higher doses," and renal tubular
`
`Par Pharmaceutical, Inc. Ex. 1010
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 6 of 7
`
`

`

`R.B. MacArthur eta/. I Molecular Genetics and Metabolism 81 (2004) S67-S73
`
`S73
`
`dysfunction, electrolyte abnormalities, and repeated
`nausea and vomiting were reported at doses of 900-
`1000 mg/kg/day. Whether, doses higher than 500 mg/kg/
`day (investigational day 1 dose for infants with hyper(cid:173)
`ammonemia) may be safely given in light of benzoate
`pharmacokinetics remains to be determined. Such treat(cid:173)
`ment would have to take place within the context of an
`investigational protocol with drug and metabolite
`monitoring. There may be other toxicities related to
`cumulative exposure, nutritional status, or concurrent
`phenylacetate administration that cannot be predicted.
`It seems that more benzoate could be administered,
`thereby, increasing the formation and clearance of HIP,
`and thus maximizing the therapeutic effect. Dose optimi(cid:173)
`zation is required however, and would depend on
`prompt availability of drug level monitoring.
`
`Metabolite formation
`
`HIP is the first of the two ammonia-scavenger metabo(cid:173)
`lites to be detected, and it appears to be both produced
`and excreted more rapidly than P AG. This observation
`still needs to be confirmed, as urine collections were not
`performed in this study, and therefore, quantification of
`metabolite excretion was not possible. A reasonable goal
`of therapy would be to alter the infusion rate ofNABZ, so
`that more HIP is produced. This objective should be
`achievable, as elevated benzoate levels were not sustained
`for the full course of the continuous infusion, and HIP
`concentrations actually decreased during the continuous
`infusion. As such, optimization of the benzoate infusion is
`required, and the required studies need to include plasma
`and urine measurements of benzoate and HIP, to relate
`urine levels of HIP to the corresponding plasma levels and
`determine total body and renal clearance rates. P AG
`appeared to be formed at a slower rate, although this also
`remains to be confirmed. Optimization ofPAG formation
`is also required, especially as it binds two moles of nitro(cid:173)
`gen for every mole of NAPA, making phenylacetate the
`more potent component of this two-drug regimen.
`
`Conclusions
`
`Both NAP A and NABZ were infused into healthy
`volunteers in a manner similar to that followed when
`treating hyperammonemia in urea cycle enzyme-defi(cid:173)
`cient patients. The pharmacokinetics of these infusions
`were characterized, and it appears that present dosing
`regimens need to be optimized to maximize the elimina(cid:173)
`tion of waste nitrogen via PAG and HIP, and to keep
`phenylacetate and benzoate doses and plasma levels
`below those associated with toxicity. Dose optimization
`will require both plasma and urine measurement of the
`metabolites, and phenylacetate and benzoate. Additional
`work in treated patients will also be required to ensure
`that the results obtained in healthy volunteers are appli(cid:173)
`cable to patients with urea cycle disorders.
`
`References
`
`[l] M. Summar, Current strategies for the management of neonatal
`urea cycle disorders, J. Pediatr. 138 (Suppl. 1) (2001) S30-S39.
`[2] A. Thibault, D. Samid, M.R. Cooper, W.D. Figg, A.C. Tompkins,
`N. Patronas, D.J. Headlee, D.R. Kohler, D.J. Venzon, C.E. Myers,
`Phase I study of phenylacetate administered twice daily to patients
`with cancer, Cancer 75 (1995) 2932-2938.
`[3] A. Thibault, M.R. Cooper, W.D. Figg, D.J. Venzon, A.O. Sartor,
`A.C. Tompkins, M.S. Weinberger, D.J. Headlee, N.A. McCall, D.
`Samid, A phase I and pharmacokinetic study of intravenous phen(cid:173)
`ylacetate in patients with cancer, Cancer Res. 54 (1994) 1690-
`1694.
`[4] K. Oyanagi, Y. Kuniya, A. Tsuchiyama, T. Nakao, E. Owada, .T.
`Sato, K. Ito, Nonlinear elimination of benzoate in patients with
`congenital hyperammonemia, J. Pediatr. 110 (1987) 634-636.
`[5] K. Kubota, T. Ishizaki, Dose-dependent pharmacokinetics of ben(cid:173)
`zoic acid following oral administration of sodium benzoate to
`humans, Eur. J. Clin. Pharmacal. 41 (1991) 363-368.
`[6] T.P. Green, R.P. Marchessault, D.K. Freese, Disposition of
`sodium benzoate in newborn infants with hyperammonemia, J.
`Pediatr. 102 (1983) 785-790.
`J. Wolff, S. Kulovich, A. Yu, C. Qiao, W. Nyhan, The effectiveness
`of benzoate in the management of seizures in nonketotic hypergly(cid:173)
`cinemia, Am. J. Dis. Child. 140 (1986) 596-602.
`
`[7]
`
`Par Pharmaceutical, Inc. Ex. 1010
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 7 of 7
`
`