ONCOLOGY
`
`Disposition of Phenylbutyrate and its
`Metabolites, Phenylacetate and
`Phenylacetylglutamine
`
`Stephen C. Piscitelli, PharmD, Alain Thibault, MD, William D. Figg, PharmD,
`Anne Tompkins, RN, Donna Hcadlec, RN, Ronald Lieberman, MD,
`Dvorit Samid, PhD, and Charles E. Myers, MD
`
`Plienylocetnte. an inducer of tumor cylostasis and dijfereiitiotiori, sliows promise as a
`relatively nontoxic antincaplastie agent. Phenylacetate, however. has an unpleasant odor
`that might limit patient acceptability. Phenylbutyrate, an odorless compound that also
`has activity in tumor models. is known to undergo rapid Conversion to phenylacetate by
`beta-oxidation in viva. This phase I study examined the phormacokinelics of phcnylbu-
`tyrale and clmracterized the disposition of the two metnliolite.-.'. phenylnrretale and pine-
`nylacetylglutnmine. Fourteen patients with cancer [aged 51.8 4: 13.8 years} received n 30~
`minute infusion of phenyibutyrate at 3 dose levels (600, 1200. and 2000 mg/m‘’]. Serial
`blood samples and 24-hour urine collections were obtained. Samples were assayed by
`high-performance liquid chromatography. A model to simultaneously describe the phar-
`macokinetics ofall three compounds was developed using ADAPT ll. Data were modeled
`as molar equivalents. The model fit the data well as shown by mean [iSD] coefficients of
`determination tr’) for phenylliulymle, phenylacetate, and phenylacetylglutamine. which
`were 0.96 :r 0.07, 0.83 i 0.10, and 0.92 1- 0.06. respectively. The intrapatient coefficient of
`variation percentage [CV%} around the parameter estimates were small [range 7.2-
`33.5%l. Phenylbutyratc achieved peak conccntrotiorls in the range ofin vitro tumor ac-
`tivity |'5UU—-2000 nmol/Ll and exhibited saturablc elimination {Km = 34.1 i: 18.1 pg/ml.
`and Vm, = 18.1 I 18 iiigfli/kg]. Melribolisiri was rapid: the times to maximum concentra-
`tion for phenylacetatc and phenylocelylglulamine were I and 2 hours, respectively. The
`conversion of phenylbutyrale to phenylacetate was extensive {B9 1 12.6%}, but serum
`concentrations ofphenylacetate were low owing to rapid. subsequent conversion to phe-
`nylacetylglutamine. The ratio of phenylbutyrate AUG to phenylacetate AUC was 2.66.
`‘Thus, phenylhutyrote may not be a prodrug for plienyloceiole and should be pursued as
`an independent antilumor agent.
`
`The amino acid phenylalanine is degraded by a
`combination of hydroxylatiun and deamination,
`leading to a range of metabolic products including
`
`From the Clinical Pharmacokinetics Research Laboratory, Pharmacy
`Department, Clinical Center. National Institutes of Health. Bethesda,
`Maryland (Dr. Piscitelli); the Clinical Pharmacology Branch, National
`Cancer Institute, National Institutes of Health. Bethesda, Maryland
`(M55. Tompkins and Headlee. and Drs. Thibault. Fi. Samid, and My-
`ers); and the Center for Drug Evaluation and Research. Fnod and Drug
`Administration, Rockville, Maryland (Dr. Lieberman). Address for corre-
`spondence: William D. Fig, PharmD. Clinical Pharmacekinctics, Sec-
`tion. Clinical Pharmacology Branch, National Cancer Institute, Building
`10. Room 5A0], Bethesda. MD 20892.
`
`368 0 J Clin Pharmacul 1995;315:368-373
`
`phenylacetate. a compound used to treat children
`with hyperammonernic urea cycle disorders.‘ Man
`and higher primates conjugate phenylacetate with
`glutamine to form phenylacetylglutarnine, whereas
`in rodents this compound is conjugated with gly-
`cine.‘ The fact that phenylacetate is conjugated with
`and depletes circulating glulamine is ofspecial inter-
`est, because tumor cells are highly dependent on this
`amino acid. rendering glutamine a target for thera-
`peutic intervention. ln addition to potential gluta-
`mine starvation, phenylacatate can arrest
`tumor
`growth by modulating the expression of genes criti-
`cal to growth control and differentiation.“
`Recently. phenylacetate has been shown to possess
`cytostatic and differentiating properties against a va-
`
`PAR PHARMACEUTICAL, INC.
`PAR PHARMACEUTICAL, INC. EX. 1023
`EX. 1023
`
`

`
`
`
`PHENYIBUTYHATE AND IT'S METABOLI'TE3, PHENYLACETATE AND PHE.'NYLACETYLGLU'l'AMlNl:.'
`
`riety of hematologic and solid tumors in laboratory
`models.” When given to healthy subjects, phenyla-
`cetate undergoes hepatic conjugation with gluta-
`mine by phenylacetyl coenzyme A: glutamine any]-
`transfcrase, which yields phenylacetylglutamine,
`the major urinary metabolite.’ Although previously
`shown to follow lirst-order pharmacoltineticsf the
`drug exhibits nonlinear, saturable pharmacokinetics
`at doses currently being evaluated in patients with
`cancer.“ Phenylacetate. however. has an unpleasant
`odor that might limit its acceptance and develop-
`ment as an oral drug.
`In contrast. phenylbutyrate is an odorless com-
`pound and has also been safely given to children for
`hyperauunoneulic urea cycle disorders.°'"’ Recent
`laboratory studies have documented that phenylbu-
`tyrate. like phenylacetate. can {1} induce selective
`cytostasis and maturation of cultu red tumor cells de-
`rived from various erythropoietic and solid neo-
`plasms {incluciing adenocarcinomas of the prostate,
`breast. ovary. colon, and long, as well as central ner-
`vous system tumors and malignant melanoma]: [2]
`modulate the expression of genes implicated in tu-
`morgrowth, metastasis, and immunogenicity; and [3]
`enhance the efficacy of other agents of clinical inter-
`est including retinoids. interferon alfa, surarnin. 5-
`aza-2'-deoxycytidine. and hydroxyurea {Samid et
`al.“; Liu et al., lludgins et al.. Fig et al., submitted:
`Sand et al., unpublished data]. Phenylbutyrate is
`converted in vivo to phenylacetate by mitochondrial
`beta—oxidation.“ Therefore. phenylbutyrate is cur-
`rently being investigated as a new autineoplastic
`agent, and as a prodrug for phenylacetate in the treat-
`ment of cancer.
`To better understand the disposition ofthese com-
`pounds after intravenous administration of phenyl-
`butyrate. a pharmacokinetic model
`that simulta-
`neously characterizes the disposition of phenylbu-
`tyrate, phenylacetate, and phenylacetylglutamine
`was developed from plasma and urine data collected
`during a phase I clinical trial.
`
`METHODS AND DEVELOPMENT OF MODEL
`
`Adults with adva nced solid tumors refractory to con-
`ventional therapy. a performance status greater than
`60% on I(arnofsky's scale.” normal hepatic transam-
`inases and bilirubin. a serum creatinine less than 1.5
`mg/dL. and normal leukocyte and platelet counts
`were eligible for this study. The clinical protocol was
`approved by the National Cancer Institute’s Institu-
`tional Review Board, and all patients gave written in-
`formed consent before participating in the study.
`Patients were enrolled into the study in cohorts of
`at least 3 per dosage level (600. 1200, and 2000 mg/
`mg]. Each patient received a single 30-minute infu-
`
`sion of phenylbutyrate, and serial blood samples
`were collected before, immediately post-dose. and at
`0.15. 0.3. 0.5, 0.75. 1, 1.5, 2.5, 3.5, and 5 l1Dt1I‘S after
`the infusion. Blood samples (5 ml.) were collected in
`5-mt. glass tubes {Vacutainer®; Becton Dickinson,
`Rutherford. NJ] either via an intravenous catheter
`[separate from the drug administration catheter} or
`venipuncture. Blood was centrifuged. and the serum
`was transferred to 5-mL polypropylene tubes and fro-
`zen at —85°C until the time of analysis. A 24-hour
`urine collection for cumulative phenylacetylglutam-
`ine excretion was done in a subset ofpatients.
`The reversed phase high—performance liquid chro-
`matography method for measuring serum concentra-
`tions of phenylacetate, phenylbutyrate. and phe-
`nylacetylglutamine has been previously described.”
`Briefly. 100 at. of 10% perchloric acid was used to
`precipitate the proteins ofa 200-p.l_. aliquot of serum,
`which was then centrifuged. The supernatant was
`neu tralized with 25 pl. ofa 20% solution ofpotassium
`bicarbonate. After centrifugation. 20 pL of superna-
`tant was injected onto a C-18 column heated at 60°C.
`Urine samples were processed similarly. after a 1:20
`dilution with water. Elution was done with an in-
`
`creasing gradient of acetonitrile in water from 5 to
`30% over 45 minutes. its progress was followed by
`monitoring ultraviolet absorbance at 203 nm. Char-
`acteristic elution times for phenylacetylglutaminc.
`phenylacetate. and phenylbutyrate were 10.1, 17.4,
`and 27.3 minutes. respectively. The assay yielded a
`lower limit of detection of 2 pg/ml. and was linear
`for concentrations as high as 2,000 pg/mL. Between
`20 and 1.000 pg/mL. the interassay CV% was less
`than 10%.
`A model to simultaneously describe the pharma-
`cokinetics ofall three compounds was developed us-
`ing ADAPT Il.” Several models were constructed to
`compare one and two compartments for each drug.
`as well as the possibility of nonlinear pharmacoki-
`netics. Model selection was determined by Akailce’s
`Information Criterion (AlC}.‘5 and by visual inspec-
`tion of the difference between measured and com-
`puter-fittecl concentrations [residl1als). Data were
`modeled as molar equivalents. The pharmacokinetic
`parameters were estimated using weighted nonlin-
`ear least squares by an adaptive process that used se-
`quential updating of priors for parameter values.
`Weighting was by the inverse of the observation vari-
`ance for all compounds.
`-
`Drug input was by intermittent intravenous infu-
`sion. To make the model identifiable. the volume of
`distribution of phenylacetate was fixed at 0.3 L/kg
`based on previous phase I data in which phenylacet-
`ate was given intravenously? Complete conversion
`of phenylacetate to phenylacetylglutamine and
`elimination of all phenylacetylglutamine in the
`
`ONCOLOGY
`
`369
`
`

`
`PISCITELLI ET AL
`
`ination rate constant for the metabolites owing to
`sparse data describing the terminal slope.
`
`RESULTS
`
`Patients
`
`Fourteen male patients were included in the study.
`Three patients received 600 mg/In” of phenylbutyn
`ate. 8 received 1200 mg/tn”, and 3 received 2000 mg]
`mg. Patient demographics are shown in Table l.
`
`Pharmacokinelics
`
`The model fit the data well as shown by mean [iSD)
`coefficients of determination {r’] for phenylbutyrate,
`phenylacelate, and plienylacetylglutamine. which
`were 0.96 : 0.07, 0.88 i 0.10, and 0.92 i 0.05, respec-
`tively. Pharmacokinetic parameters are shown in
`Table I]. The intrapatient CV93 around the parame-
`ter estimates were small, ranging from 7.2 to 33.5%
`of the fitted values. The mean inlerpatient CV% for
`parameter values ranged from 11.85 to 34.6%.
`Serum concentration-lime plots for a representa-
`tive patient in each dosage group are shown in Figure
`2. Similar fits were seen for the other patients. Peak
`concentrations of phenylbutyrate after 600 mg/m’
`ranged from 31 to 57 pg/mL. After 1200 mg/m2 and
`2000 mg/mz, peak concentrations in serum ranged
`from 57 to 115 pg/mL and 114 to 184 pg/mL, respec-
`tively. Concentrations at 5 hours after dosing were 2
`
`
`
`TABLE I
`
`
`
`Figure 1. Model to describe the disposition of phenylbutyrote [PB],
`ol1enylocetotetPAl. and phcnvlocctylglutominc {PAC} illustrating
`the phormncoldnetic parameters.
`Abbreviations: V”, volume of distribution for PR: Um. volume
`of distribution for PAL Vmg. volume of distribution for FAG: MM.
`Michoelis-Menten elimination: Km. Michoelis-Menten constont:
`V,,,,,,,, maximum elimination rate: CL,.j'ormotior1cIl clcuronce ofPA
`to PAC: CL}. clearance of PAC into the urine: Cl..." clearance of PB
`out of the control compartment.
`
`urine was also assumed based on our previous phase
`I experience.“ Thus, the fraction of phenylbutyrate
`converted to phenylacetate was determined using
`the following equation:
`
`urinary phenylacetylglutainine [,umol)
`dose of phenylbutyrate [,umol}
`
`'
`
`The pharmacokinetic parameters for phenylacet-
`ylglutamine are dependent on this fraction, which is
`analogous to oral drugs where clearance and volume
`are dependent on the value of bioavailability [i.e.,
`CL/F or Vs‘./F].
`The model eventually used was a one-comparb
`ment nonlinear model for phenylbutyrate with con-
`version to a one-contpartnlent linear model for phe-
`nylacetate, and further conversion of phenylacetate
`to phenylacetylglutamine {one-compartment]. Phe-
`nylbutyrate was parameterized by a central volume
`[VH3]. a minor elimination pathway [CI.m]. and a non-
`linear function consisting of intrinsic clearance
`[CLi,..] and the Michaelis—Menten constant {Km}. The
`Vma, is equal to CL“-Km. The CL, and CL3 describe
`the clearances of phenylacetate to phenylacetylglu-
`tamine and phenylacetylglutamine into the urine.
`respectively. The Va“; describes the volume of dis-
`tribution [V2] for phenylacetylglutamine. The V”
`represents the volume of distribution of phenylacet-
`ate. The model displaying the pharmacokinetic pa-
`rameters is shown in Figure 1.
`The area under the serum concentration versus
`
`time curve [AUG] was calculated by the trapezoidal
`rule according to Gibaldi and Perrier.” The AUC was
`determined from time zero until the last time point
`{5 hours). because concentrations of each compound
`were usually below detectable limits at this point
`and because of the difficulty in determining an elim-
`
`370 I Jclin Pharmanol !995:35:368—373
`
`Individual and Mean Patient Demographics
`Patient
`Age
`N°-
`0'')
`
`Height
`(Chit
`
`Weight
`(kg)
`
`Dose
`
`Dose/m’
`(ms)
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`ll
`12
`13
`14
`
`75
`E0
`55
`66
`55
`61
`39
`29
`48
`35
`42
`46
`7 1
`43
`
`174
`183
`177
`164
`180
`180
`167
`179
`170
`169
`184
`152
`180
`158
`
`63.9
`87.0
`79.9
`70.6
`101.4
`101.4
`80.2
`70.6
`71.4
`72.5
`103.0
`54.0
`82.0
`48.7
`
`600
`600
`600
`1200
`1200
`1200
`1200
`1200
`1200
`1200
`1200
`2000
`2000
`2000
`
`1080
`1278
`1188
`2112
`2640
`2352
`2292
`2340
`2143
`2195
`2700
`3000
`4020
`2940
`
`77.6
`173.0
`51.8
`Mean
`16.7
`10.2
`13.8
`SD
`3-
`
`

`
`
`
`PHENYLBUTYHATE AND ITS METABOLITES. PHENYLACETATE AND PHENYLACE'I'YLGLU'I'AMINb‘
`
`
`
`TABLE II
`
`Pharrnacoklnetic Parameters Derived
`From the Model
`
`Mean
`
`SD
`
`.
`
`-
`
`0.08
`0.21
`V... (L/ks)
`0.04
`0.1 0
`CL... (L/hr/kg)
`18.]
`34.1
`K... (us/mt)
`0.30
`0.50
`CL.... (L/hr/kg)
`18.0
`18.1
`V....,..(mg/hr/kg)
`Fixed
`0.30
`v... (L/kg)
`0.13
`0.37
`CLt1(L/hrjkg)
`0.11
`0.19
`11...... (L/kg)
`0.11
`0.17
`CLt2 (L/hr/kg)
`139.6
`265.4
`AUC PB 1500 mg/m’
`167.9
`557.8
`AUC PB 1200 mg/m2
`689.2
`1214.5
`AUC PB 2000 mg/I21’
`16.6
`120.0
`AUC PA 600 mg/m
`90.6
`220.2
`AUC PA 1200 mg/m?
`160.0
`608.3
`AUC PA 2000 mg/m2
`119.2
`401.3
`AUC PAG 600 mg/m’
`269.7
`438.0
`AUC PAG 1200 mg/m’
`389.6
`1055.4
`Auc FAG 2000 mg/m2
`Visa — volume of distribution for PE; \I'p.. — volume of distribution for PA". V9.5 =
`volume oidistlibution for PEG: K... = Michae|is—Mentericonstant. CL... — intrinsic
`clearance: \I..._. = rnaicimum elimination rate: CLt1 = lormationalclearance of PA
`to FAG: (LL12 = clearance of FAG into the urine: CLm = clearance oi PB out oi
`central compartment: AUC = area under the curve irorn time 0 to 5 hours post-
`dose: PB = phenylbutyratej PA = phenyiacetate; P.FiG = phenylacetylglutamine.
`AUC vaiues are represented as great -hr/L.
`
`.ug/mL or lower in all patients. Phenylbutyrate ex-
`hibited saturable elimination pharmacokinetics as
`evidenced by concave log-linear plots on visual in-
`spection, an AUCN, disproportionate to dose {Figure
`3), and improved tits at high doses using a nonlinear
`function as assessed by AIC. Final estimates for Mi-
`chaelis-Menten parameters were a K... of 34.1 i 18.1
`pg/mL and a V...“ of18.1 i 18.0 mg/h/kg.
`Six patients had complete 24-hour urine collec-
`tions for determination of phenytbutyrate conver-
`sion to phenylacetatc. 1 at 600 rng/mz. and 5 at 1200
`mg/m2. The percentage of conversion was high with
`a mean (:SD] of80.U i 12.6%. The conversion ranged
`from 68 to 100% of phenylbutyrate accounted for in
`the urine by phenylacetylglutarnine.
`Phenylacelate was detectable in plasma immedi-
`alely after phenylbutyrate infusion with mean [iSD]
`peak concentrations of 20.7 .t 13.6 pg/mL. The time
`to maximum concentration most commonly oc-
`curred 30 to 60 minutes after the infusion. The serum
`concentrations of phenylacetate that were seen in
`this study were much lower than those after intrave-
`nous administration ofphenylacetate.“ After phenyl-
`butyrate administration. phenylacetatc followed
`
`first-order elimination. The Michaelis-Menten con-
`stant of phenylacetate from our previous trial" was
`105.1 i 44.5 pg/n1L. The highest concentration of
`phenylacetaie achieved in this study was 57 pg/mL
`with 11 of the 14 patients exhibiting peak concentra-
`tions less than 30 pg/mL. Because the peak phenyla-
`cetate concentrations were less than or equal to one-
`half the K..., the nonlinear function of phenylacetate
`collapses to a first-order rate constant.” As a compar-
`ison ofthe total exposure between the 2 compounds,
`the mean [:80] ratio of phenylbutyrate AUC to phe-
`nylacetate AUC was 2.66 i 1.57.
`Phenylacetylglutamine
`serum concentrations
`were also observed immediately after phenyIbutyr-
`ate dosing. However, peak concentrations appeared
`1 to 3.5 hours after the infusion. which was later
`than those of phenylacetate. Phenylacetylglutamine
`achieved maximum serum concentrations of 59.5 4;
`
`34.2 .ug/mL. which ranged from 27 to 129% of those
`of plwnylbutyrate (mean 1 SD. 61.2 i 29.0%]. Com-
`paratively. phenylacetate achieved peak concentra-
`tions that were only 38.8 i 19.2% of those of pheny-
`lacetylglutarnine.
`
`DISCUSSION
`
`Pharmacolcinetic models of anticancer agents can be
`used for a variety of purposes. In addition to describ-
`ing a drug’s disposition. these models can be used to
`[1] determine sampling schemes based on a small
`nurnber of blood samples [using optimal sampling
`theory"]; [2] predict plasma concentrations of new
`regimens: or [3] optimize dosing for maximal efficacy
`and minimal toxicity in patients receiving multiple
`courses oftherapy.
`The simultaneous modeling approach used in this
`analysis accurately characterized the conversion
`and disposition ofphenylbutyrate and its two metab-
`olites, phenylacetate and phonylacetylglutamine.
`There is increasing interest in phenylacetate as a rel-
`atively nontoxic antitumor agent.3‘“'“ The unpleas-
`ant odor of phenylacelate. however, may limit its ac-
`ceptance by patients. Phenylbutyrate is the odorless
`precursor of phenylacetate. with demonstrable anti-
`tumor activity in laboratory models. Phenylbutyrate
`was converted to phenylacetate with subsequent
`conversion to phenylacetylglutamine. These conver-
`sions were rapid with detectable amounts of both
`metabolites occurring less than 10 minutes after ini-
`tiation of the phenylbutyrate infusion. Phenylbutyr-
`ate was characterized by nonlinear elimination
`pharmacokinctics with a K... of 34.1 pg/mL and a
`V,,._.,,. ol"’l8.11ng/h/kg.
`Our group has previously reported the results of a
`phase I study of intravenous phenylacetate that
`showed nonlinear pharmacokinetics for pheny1acet-
`
`ONCO L0 GY
`
`371
`
`

`
`PISCITELLI ET AL
`
`Figure 2. Actual {squares} and computer-fitted (line) concentration-
`time profiles of phenylbutyrnte, phenylncetule, and phenylncetyl-
`glutamine in [A] patient 1 receiving 600 mg/m’ of phcnylbulyrate;
`{B} patient 5 receiving 1200 mg/m' of phenylbutyrute; [Cl putient
`12 receiving 2000 mg/m’ of phenylbutymte.
` FF—~—
`
`ate characterized by saturable metabolism to phe-
`nylacctylglulamine.” In this study. where concentra-
`tions of phenylacetate were smaller than the re-
`ported Km. the Michaelis-Menten function reduces
`to a lirst-order rate constant. Thus. no saturability of
`phenylacetate was observed. The low concentrations
`of phenylacetate seen in this study may also he re-
`lated to the small doses of phenylbutyrate used here
`compared with the initial phase I trial. which used
`a 150-mg/kg [approximately 6000 mg/m2] bolus of
`phenylacetate."
`Preclinical antitumor activity has been observed
`for phenylhutyrate at concentrations of 500 to 2000
`pmol/L [94-376 pg/ml.]. This concentration range
`was shown here to be clinically achievable after a
`30-minute infusion. It will be important to further
`evaluate the pharrnacokinctics of phenylbutyrate
`using alternative dosing strategies [e.g., continuous
`infusion] or higher doses to determine whether these
`concentrations can be maintained for longer periods
`of time. In addition, continuous infusion may yield
`higher pherlylacetate concentrations, especially if
`saturation of phenylacetate is achieved.
`Phenylbutyrate is known to undergo rapid conver-
`sion to phenylacetate in vivo by beta—oxidation.
`
`Dose [mg]
`
`C‘!
`Ed
`
`E5
`
`1,
`
`034
`
`1:
`
`1cm
`
`200 D
`
`3000
`
`4 D D 0
`
`
`
`concentration(nu)
`
`.-..
`E:..
`I:o
`.1on&¢-l
`n0I:
`0u
`
`Et
`
`
`
`concentration(UM)
`
`372 I J Clin Pharmacot l.995:35:358—373
`
`Figure 3. Plot of phenyltlutyrote (lose [mg] Illlti area under the
`curve. Line of best fit is shown: 3! = 105.95 - I0""°°°’°”", [H = .78].
`
`

`
`PHENYLBUTYRATE AND 113 ME'l'AB()LiTES PHENYLACETATE AND PHENYLACE'tYLGLUTAltttNtI
`
`However, this conversion has not been extensively
`studied to date. Therefore, one purpose of this study
`was to examine the percentage of phenylbutyrate
`that is metabolized to phenylacetate. Comparison of
`phenylacetate and phenylbutyrate AUC would be
`inappropriate because metabolites often have con-
`siderable differences in clearance and half-life, as
`well as different distributional properties. More im-
`portantly. phenylacetate is only an intermediate me-
`tabolite and undergoes further metabolism to phe-
`nylacetylglutamine; rapid conjugation with gluta-
`mine would clearly result in a low AUC for phenyla-
`cetate. We therefore determined the conversion of
`
`phenylbutyrate based on the amount of phenylacet-
`ylglutamine recovered in the urine. This method is
`valid based on the more than 99% conversion of phe-
`nyl-acetate to phenylacetylglutamine that our group
`has previously shown.“ Although one patient did ex-
`hibit a 100% recovery, several patients showed in-
`complete conversion. The latter could he owing to
`assay variance of 10% for each compound and any
`error incurred in collection of 24-hour urine sam-
`
`ples. Another possible factor may have been a urine
`collection of only 24 hours for determination of uri-
`nary phenylacetylglutamine. An additional period of
`collection may have yielded a greater recovery. Also,
`small traces of phenylbutytate and phenylacetatc.
`below the limit of quantification of the assay. were
`observed in the urine. The combination of all these
`
`factors is likely to play a role in explaining the in-
`complete conversion.
`In summary, phenylhutyrate exhibits saturablc.
`nonlinear pharmacokinetics after intravenous ad-
`ministration and achieves peak concentrations in the
`range of in vitro antitumor activity. Concentrations
`of the active, intermediate metabolite [phenylacet-
`ate) were low in this study and did not achieve levels
`at which saturation occurs. The conversion of phe-
`Itylbutyrate to phenylacetate was high (80%), but the
`rapid. subsequent conversion to phonylacotylglu—
`tamine resulted in serum levels of phenylacetate
`that were much lower than those seen when the drug
`is given intravenously. ‘We conclude that phenylbu-
`tyrate should not be considered a clinically useful
`prodrug ofphenylacetate and that both phenylbutyr-
`ate and phenylacetate should be pursued as indepen-
`dent antineoplastic agents.
`
`The authors thank Natalie McCall. Bernadette Altman. and ]en—
`nifer Stevens for laboratory assistance.
`
`REFERENCES
`
`‘l. Brusilow SW,Danney M. Weber L], Batshaw M. Burton B. Lev-
`itsky i.. Roth K. McKeethren C. Ward I: Treatment of episodic hy.
`perarnrnonemia in children with inborn errors of urea synthesis.
`N Englj Med 1984;310:1639-1634.
`
`2. ]_ames MO, Smith RL, Williams RT. Reidenberg M: The conju-
`gation of phenylacetic acid in man. sub-human primates. and
`some non-primate species. Pros it Soc Lond B t9?2;‘182:25—35.
`3. Samid D. Shack S, Sherman LT: Phenylacetate: A novel non-
`toxic induccr of tumor cell differentiation. Cancer Hes 1992;52:
`1933-51 992.
`
`:1. Samid D. Yell 1'5. Prasana P: ittduction of ervthroid differentia-
`tion and fetal hemoglobin production in human leukemic cells
`treated with phenylacetate.Bloodt992;8t):i5?6-1581.
`5. Sarnicl Ll. Shack S. Myers CE: Selective growth arrest and phe-
`notypic reversion of prostate cancer cells in vitro by nontoxic
`pharmacological concentrations of phenylacetate. J Clin invest
`1993;91:2288-2295.
`6. Samid D. Ram Z. Hudgins WR. Shack S. Liu L. Walbridge S.
`Oldfield El-[. Myers CE: Selective activity ofphenylacetate against
`malignant gliurltas: Rescnitilaoce to fetal brain damage in pl1enyl-
`ketonuria. Cancer Fles 1994; 54:891-895.
`T. Sirneil CI. Sipila I. Rajantie ], Valle DL. Brusilow SW: Waste ni-
`trogen excretion via amino acid acylation: Eenzoate and phenyl-
`acetate in lysinuric protein intolerance. Pedialr iles1986:2{):‘l1‘t?'-
`1121.
`
`It. Thihault A. Cooper MR. Figg WD. Venzon DJ‘. Sartur AD. Tom-
`pkins AC. Weinberger MS. Headlee DJ. McCall NA. Samid D. My-
`ers CE: A phase I and pharmacokinetic study of intravenous phe-
`nylacelate in patients with cancer. Culrccr Hes1994:54:1690-1694.
`El. Brusilow SW. Horowich AL; Urea cycle enzymes. in Scriber C.
`Beaudet A. Sly W. Valle DR {eds}: Metabolic Basics of inherited
`Diseases. New York: Mctlraw Hill. I9B9;fi29—fiti-t.
`1|]. Brusilow SW: tlltenylacetylglutamino may replace urea as a
`vehicle for waste nitrogen excretion. Pediotrflcs 1991; 29:14?-150.
`11. Knoop F‘: Der Abbau aromatischer feltsaure Tierkorper. Bietr
`Chem Physio!Potlrol19tJ5:Ei:150—162.
`12. Karnofsky DA. Abelmzut Wtl. Craver LF. Burcltenzrl Ill: The
`use of the nitrogen mustard-s in the palliative treatment of carci-
`noma. Concer 19=‘t8;1:63-1-656.
`13. Thibault A, Fig WU. McCall N. Myers CE, Cooper MR: it si-
`multaneous assav of the differentiating agents phenylacelate and
`pheoylbutyrale. and one of their metabolites. phenylacety|giu-
`tamine. by reversed-phase, high performance liquid chromatogra-
`phy. lLiq Chromotogr 1994;1l?’:2t]95-290i].
`14. I)'Argenio DZ. Sehumitzky A: ADAPT User's Guide. Biomedi-
`cul Simulations Resource. Los Angclcs: University of Southern
`California, 1990.
`
`15. Yamaoka K. Nakagawa T, Uno T: Application of Akaike’s to-
`formation Criterion in the evaluation of linear pharmacokinctic
`equations. J‘ Phormucokinct Hiophorm ‘t978;6:1ti5-]?5.
`16- Gibtlldi M. Perrier D: Pharmocokinetics, 2nd ed. New York:
`Marcel Deltker. Iuc., 1932.
`'17. D'Arget1iu DZ: Optimal sampling times for pharmacokinetic
`experiments. IPl1orn1ocokinet Biophurm I9ttI;9:?39—?56.
`
`ONCOLOGY
`
`373