Vol. 7, 3047–3055, October 2001
`
`Clinical Cancer Research
`
`3047
`
`A Phase I Clinical and Pharmacological Evaluation of Sodium
`Phenylbutyrate on an 120-h Infusion Schedule1
`
`Michael A. Carducci,2 Jill Gilbert,
`M. Katherine Bowling, Dennis Noe,
`Mario A. Eisenberger, Victoria Sinibaldi,
`Yelena Zabelina, Tian-ling Chen,
`Louise B. Grochow, and Ross C. Donehower
`Divisions of Medical Oncology [M. A. C., J. G., M. K. B., M. A. E.,
`V. S., L. B. G., R. C. D.] and Experimental Therapeutics and
`Pharmacology [M. A. C., D. N., Y. Z., T-l. C., L. B. G.], The Johns
`Hopkins Oncology Center, The Johns Hopkins University School of
`Medicine, Baltimore, Maryland 21231
`
`ABSTRACT
`Purpose: Sodium phenylbutyrate (PB) demonstrates
`potent differentiating capacity in multiple hematopoietic
`and solid tumor cell lines. We conducted a Phase I and
`pharmacokinetic study of PB by continuous infusion to
`characterize the maximum tolerated dose, toxicities, phar-
`macokinetics, and antitumor effects in patients with refrac-
`tory solid tumors.
`Patients and Methods: Patients were treated with a
`120-h PB infusion every 21 days. The dose was escalated
`from 150 to 515 mg/kg/day. Pharmacokinetics were per-
`formed during and after the first infusion period using a
`validated high-performance liquid chromatographic assay
`and single compartmental pharmacokinetic model for PB
`and its principal metabolite, phenylacetate.
`Results: A total of 24 patients were enrolled on study,
`with hormone refractory prostate cancer being the predom-
`inant tumor type. All patients were evaluable for toxicity
`and response. A total of 89 cycles were administered. The
`dose-limiting toxicity (DLT) was neuro-cortical, exemplified
`by excessive somnolence and confusion and accompanied by
`clinically significant hypokalemia, hyponatremia, and hy-
`peruricemia. One patient at 515 mg/kg/day and another at
`345 mg/kg/day experienced this DLT. Toxicity resolved <12
`h of discontinuing the infusion. Other toxicities were mild,
`including fatigue and nausea. The maximum tolerated dose
`
`Received 3/14/01; revised 6/11/01; accepted 6/13/01.
`The costs of publication of this article were defrayed in part by the
`payment of page charges. This article must therefore be hereby marked
`advertisement in accordance with 18 U.S.C. Section 1734 solely to
`indicate this fact.
`1 Supported by UO-1 CA 70095, American Society of Clinical Oncol-
`ogy Young Investigator Award; NIH-K08-CA-69164; NIH-R01-
`CA75525 (to M. A. C.); Aegon Scholarship in Oncology (to J. G.),
`General Clinical Research Center N01-CM07302 from the National
`Cancer Institute.
`2 To whom requests for reprints should be addressed, at The Johns
`Hopkins Oncology Center, 1 M88 Bunting-Blaustein Cancer Research
`Building, 1650 Orleans Street, Baltimore, MD 21231-1000. E-mail:
`carducci@jhmi.edu.
`
`was 410 mg/kg/day for 5 days. Pharmacokinetics demon-
`strated that plasma clearance of PB increased in a continu-
`ous fashion beginning 24 h into the infusion. In individuals
`whose Vmax for drug elimination was less than their drug-
`dosing rate, the active metabolite phenylacetate accumu-
`lated progressively. Plasma PB concentrations (at 410 mg/
`kg/day) remained above the targeted therapeutic threshold
`of 500 ␮mol/liter required for in vitro activity.
`Conclusion: The DLT in this Phase I study for infu-
`sional PB given for 5 days every 21 days is neuro-cortical in
`nature. The recommended Phase II dose is 410 mg/kg/day
`for 120 h.
`
`INTRODUCTION
`Differentiation therapy for epithelial malignancies may po-
`tentially alter tumor growth and progression, slow or inhibit
`metastases, inhibit angiogenesis, and/or effect response to other
`forms of therapy (1–5). Nonretinoid agents such as hexameth-
`ylenebisacetamide and sodium butyrate have been evaluated
`clinically, but sustained systemic levels required for activity
`have not been achieved because of toxicity and/or lack of a
`suitable formulation (6 –9). Sodium PB,3 an aromatic fatty acid,
`is a lead candidate as a cancer differentiating agent and a histone
`deacetylase inhibitor (10 –13). Sodium PB is the precursor to
`PA, and both compounds are potent differentiating agents in
`vitro (10 –18). PA is formed by ␤-oxidation of PB (19). PB is
`Food and Drug Administration approved for children and adults
`with hyperammonemia associated with urea cycle disorders
`(recommended dose of 13 g/m2/day) and has been used for adult
`patients with hyperammonemia secondary to high-dose chemo-
`therapy for leukemia and transplant therapies (19 –22). PB is
`also under investigation for cystic fibrosis and adrenal leukodys-
`trophy.
`Preclinical studies have looked at the ability of these drugs
`to alter gene expression and promote differentiation (23–25). In
`various tumor model systems, PA/PB can arrest cells in G1-G0
`with induction of p21WAF1 and other cdk-2-associated cell cycle
`checkpoint proteins, alter expression of invasion products such
`as urokinase-plasminogen activator (UPA), induce apoptosis,
`inhibit telomerase, and increase MHC class I expression at
`concentrations of 500-2500 ␮mol/liter PB (23–31). Delay in
`tumor progression has been noticed in prostate and malignant
`glioma models (12, 16, 17). It should be noted that tumor
`markers, such as PSA, might not be the most accurate measure-
`ments of progressive disease in patients treated with PB, as a
`rise in tumor markers may signal cell differentiation rather than
`
`3 The abbreviations used are: PB, phenylbutyrate; PA, phenylacetate;
`PG, phenylacetylglutamine; PSA, prostate-specific antigen; MTD, max-
`imum tolerated dose; NCI, National Cancer Institute; DLT, dose-limit-
`ing toxicity; PCA, prostate cancer.
`
`
`
`clincancerres.aacrjournals.org Downloaded from
`
`on August 3, 2018. © 2001 American Association for Cancer
`Research.
`
`Par Pharmaceutical, Inc. Ex. 1007
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 1 of 10
`
`

`

`3048 Phase I Study of Phenylbutyrate
`
`disease progression. Proposed mechanisms of action to elicit
`these effects include modification of lipid metabolism, activa-
`tion of the peroxisome proliferator activator receptor, mainte-
`nance of histone acetylation through inhibition of histone
`deacetylase, inhibition of protein isoprenylation, and/or gluta-
`mine starvation (10, 27–31). There may be other cellular and
`molecular effects of PB that contribute to its antitumor activity.
`In addition to all of the potential cellular and molecular
`effects, PB is minimally toxic in children with urea cycle dis-
`orders and hemoglobinopathies (31). Serum concentrations of
`PB have been reported in the range (500 –2500 ␮mol/liter)
`required for in vitro effects of growth arrest and differentiation.
`Preclinical data from our laboratory and others suggest PB may
`be more potent and produce more molecular effects than PA
`(12). PB has activity similar to butyrate (induction of apoptosis
`and histone acetylation) that PA may not possess. For these
`reasons, Phase I studies of PB were initiated.
`The purpose of this Phase I trial was to determine the MTD
`and toxicity profile of PB administered i.v. as an 120-h infusion
`every 21 days and to describe the pharmacokinetic behavior of
`this novel agent in patients with refractory solid tumors.
`
`PATIENTS AND METHODS
`Patient Eligibility
`The protocol was approved by the Joint Committee on
`Clinical Investigation of the Johns Hopkins University School
`of Medicine. Male and female patients ⱖ18 years of age with a
`histologically or cytologically documented diagnosis of cancer
`(solid tumor or lymphoma) refractory to conventional therapeu-
`tic modalities were eligible. All patients enrolled onto the study
`had progressing, evaluable, or measurable disease. Patients were
`required to have an Eastern Cooperative Oncology Group per-
`formance status of 0, 1, or 2 and a life expectancy of ⱖ3 months.
`Patients were not to have had major surgery, radiotherapy, or
`chemotherapy in ⱕ28 days and were to be fully recovered from
`the toxicity of prior therapy. Patients with prostate cancer were
`allowed to continue on a leutenizing hormone releasing hor-
`mone (LHRH) agonist. A 4 – 6-week period of antiandrogen
`(flutamide and bicalutamide) withdrawal and evidence of PSA
`progression was required of all men on an antiandrogen before
`entry. Patients that previously received suramin were eligible if
`their suramin level was ⬍50 ␮g/ml. Adequate organ function
`was required at study entry: WBC ⬎ 2,000 or absolute neutro-
`phil count (ANC) ⬎ 1,500/mm3, platelets ⬎ 100,000/mm3, and
`hemoglobin ⬎ 9 g/dl, a serum creatinine ⬍ 2 mg/dl, total
`bilirubin ⬍ 1.5, and aspartate aminotransferase/alanine amino-
`transferase ⬍ 1.5 ⫻ the upper limit of normal, left ventricular
`ejection fraction (LVEF) by MUGA or echocardiogram ⬎ 40%
`and no history of congestive heart failure (CHF) requiring
`hospitalization, or uncontrolled hypertension (diastolic blood
`pressure (BP) ⬎110 mmHg) and a forced expiratory volume
`⬎ 1.5 liter/min. Patients with obstructive uropathy were
`(FEV)1
`eligible if obstruction was relieved by nephrostomy or other
`appropriate intervention. Patients were required to give in-
`formed consent with understanding of the investigational nature
`of the treatment and its potential risks. Patients were ineligible
`if they had an active infectious process, including HIV or viral
`hepatitis; a malignant brain tumor, or known central nervous
`
`system metastases even if treated previously and not clinically
`active; an active seizure disorder; or a baseline dementia with
`mini-mental score ⬍ 23 (32). Pregnant or nursing females were
`ineligible. Patients with requirement for steroids for any reason,
`including previous suramin therapy, were excluded because of
`concerns about potential interference with the mechanism of
`action of PB.
`
`Pretreatment Evaluation
`Patients were assessed before receiving the first dose of PB
`with a complete history, including prescription and nonprescrip-
`tion drug history, physical exam with a thorough neurological
`examination, including a mini-mental status examination (32),
`and performance status assessment. Documentation of evaluable
`or measurable disease was performed ⱕ4 weeks of initiation of
`therapy. Pretreatment laboratory evaluations included a com-
`plete blood count with platelet and differential, chemistry panel
`(including electrolytes, creatinine, bilirubin, aspartate amino-
`transferase, alanine aminotransferase, alkaline phosphatase, uric
`acid, total protein, albumin, calcium, phosphorous, and magne-
`sium), partial
`thromboplastin time and prothrombin time,
`urinalysis, human chorionic gonadotropin (if woman of child-
`bearing age), EKG, MUGA or echocardiogram for LVEF,
`spirometry with lung volumes, chest roentgenogram (or chest
`computed tomography if part of the tumor assessment), appro-
`priate tumor marker (i.e., PSA, CA-125, CA15-3, and CA19-9),
`and a suramin level, if the patient had been treated previously
`with suramin.
`
`Evaluation During Therapy
`Patients were monitored with laboratory and clinical eval-
`uations weekly by an investigator or research nurse while on
`study. The history and physical was performed at the beginning
`of each 5-day infusion. Patients were also seen as needed for
`additional toxicity assessment. Toxicity was graded using the
`NCI common toxicity criteria, version 2.0. Tumor markers were
`assessed monthly. Evaluation of measurable and evaluable dis-
`ease was performed after each two cycles of therapy. Documen-
`tation of performance status was required with all visits. Com-
`plete response was defined as the disappearance of all clinical
`evidence of active tumor and symptoms for ⱖ1 month. Partial
`response required a ⬎50% decrease in the sum of the products
`of the perpendicular tumor diameters of all measurable lesions
`for ⱖ4 weeks without simultaneous increase in the size of any
`lesion or appearance of any new lesion. Progressive disease was
`defined as a ⬎25% increase in the size of any measurable lesion
`or the appearance of any new lesion. Stable disease represented
`a response less than a partial response or growth less than
`progressive disease. Changes in tumor markers did not factor in
`to the definition of response, e.g., a rising PSA did not define
`progression. Changes in PSA are merely descriptive.
`
`Treatment Plan
`The Investigational Drug Branch of the NCI received phe-
`nylbutyric acid, sodium salt (National Service Center number
`657802; EL532) from Elan Pharmaceutical Research Corp.,
`Gainesville, GA. PB was formulated in 50-ml glass vials con-
`taining 40% viscous solution of sodium PB in sterile water (400
`
`
`
`Downloaded from on August 3, 2018. © 2001 American Association for Cancerclincancerres.aacrjournals.org
`
`
`Research.
`
`Par Pharmaceutical, Inc. Ex. 1007
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 2 of 10
`
`

`

`Clinical Cancer Research
`
`3049
`
`mg/ml). The sodium content of one vial was 4.9 grams of
`sodium. PB was provided for the patients on this study by the
`Division of Cancer Treatment, NCI.
`Patients were admitted to the General Clinical Research
`Center of the Johns Hopkins Hospital for each course of therapy.
`PB was administered as a continuous i.v. infusion (centrally or
`peripherally) for 120 h every 21 days. A course was defined as
`the 21-day cycle, consisting of the 5-day infusion and 16 days of
`rest. The infusional starting dose was 150 mg/kg/day for 5 days,
`a dose below that shown previously to be safe in children with
`urea cycle disorders. Subsequent dose levels were 225, 285,
`345, 410, and 515 mg/kg/day for 5 days. Doses could be
`escalated within individual patients if 3 patients at the next
`higher dose level had received therapy for ⬎1 month without
`DLT. Patients remained on study until evidence of progressive
`disease, severe DLT, or patient request to discontinue therapy.
`DLT was defined using the NCI common toxicity criteria, as
`neurotoxicity ⱖ grade 2 or any other toxicity ⱖ grade 3. A
`registered nutritionist met with every patient to discuss salt
`intake and to provide recommendations of lowering sodium
`intake while receiving PB.
`Cohorts of 4 patients were initially enrolled at each dose
`level to fully evaluate the central nervous system toxicity and to
`gather more pharmacokinetic data. If no patients in a cohort
`experienced a DLT during the initial 3-week observation period,
`the dose was escalated, and the next cohort was commenced. If
`1 of 4 patients experienced a DLT, 2 additional patients were
`added at that dose level. If 2 patients experienced a DLT at any
`dose level, no additional patients were enrolled at that dose
`level, and an additional 2 patients were entered at the next lower
`dose level. Dose escalation could occur once all 4 patients in a
`given cohort were treated, and at least 2 patients had completed
`one course (21 days) with the 3rd and 4th patients ⱖ1 week into
`therapy without meeting criteria for the MTD. Although the
`study was not designed to escalate doses within patients, in-
`trapatient dose escalation was allowed. A patient who received
`at least two courses of treatment without toxicity (except alo-
`pecia) greater than grade II at a given dose level could have their
`subsequent treatments at the next higher dose level. Before such
`an escalation could occur, 2 PB-naı¨ve patients must have com-
`pleted at least one course (22 days) at the next higher dose level
`without any DLT. The MTD was defined as that dose level at
`which consistent, definable toxicity occurs that is reversible and
`does not subject patients to excessive risk or discomfort.
`Patients who received ⱖ1 day of PB were assessable for
`toxicity, but only those patients who received one full 21-day
`course of PB were assessable for response. Patients for whom a
`drug was held because of DLT still had to complete the full
`21-day observation period. Any patient who experienced DLT
`and recovered fully from that toxicity could resume PB at the
`next lower dose level; however, for MTD determination, pa-
`tients were included only at their initial, higher dose level.
`Patients who did not recover fully from DLT ⱕ14 days or had
`a DLT at a subsequent lower dose were taken off study. The
`study was defined as complete for a given patient if the patient
`did not recover fully from a DLT ⱕ14 days or had a DLT at a
`subsequent lower dose. Patients were also removed from the
`study if they had evidence of progressive disease or if the patient
`requested to be taken off the study.
`
`Table 1 Gradient profile for HPLC
`
`Time (min)
`0.00
`20.00
`23.00
`35.00
`35.01
`38.00
`38.01
`45.00
`
`A (%)
`47.5
`38.5
`38.5
`27.5
`5.0
`5.0
`47.5
`47.5
`
`B (%)
`5.0
`23.0
`23.0
`45.0
`90.0
`90.0
`5.0
`5.0
`
`C (%)
`47.5
`38.5
`38.5
`27.5
`5.0
`5.0
`47.5
`47.5
`
`Recognizing that the recommended Phase II dose is often
`one dose level below the MTD, the recommended Phase II dose
`was to be dictated by toxicities, patient tolerability, and plasma
`levels achieved. An objective of this study was to achieve a
`range of plasma levels that may be clinically relevant as dictated
`by preclinical models.
`
`Pharmacokinetic Study
`Pharmacokinetic Sampling. Drug disposition studies
`were performed during the first course of treatment. Blood
`specimens were collected before the start of the infusion, at 0.5,
`1, 2, 4, 6, 12, 24, 36, 48, 60, and 96 h during the infusion and
`just before completion of the infusion (120 h). Postinfusion
`specimens were obtained at 0.5, 1, 2, 4, 8, and 24 h as measured
`from the end of the infusion. Urine specimens (every 24 h) were
`collected between 0 and 24 h and between 96 and 120 h as
`measured from the start of the infusion.
`Analytic Assay. Plasma PB, PA, and PG concentrations
`were determined in all blood specimens by reverse-phase high-
`performance liquid chromatography assay. These methods have
`not been published previously. Plasma (200 ␮l) containing
`compound was transferred to a microcentrifuge tube, followed
`by the addition of 50 ␮l of 10% perchloric acid (Sigma Chem-
`ical Co., St. Louis, MO) for protein extraction. The samples
`were vortexed and centrifuged (International Equipment Co.,
`Needham Heights, MA) at 4°C at 8500 rpm for 10 min. After
`centrifugation, 150 ␮l of supernatant was added to 5 ␮l of
`super-saturated potassium bicarbonate solution for neutraliza-
`tion and centrifuged again at 4°C at 8500 rpm for 10 min. All of
`the supernatant was transferred to an autosampler vial for anal-
`ysis. The chromatographic apparatus consisted of a Hewlett
`Packard Series II 1090 Liquid Chromatographic (Hewlett Pack-
`ard Corp., Palo Alto, CA) with an autosampler compartment,
`solvent delivery system, and a diode-array UV absorbance de-
`tector with a resolution of 2 nm. The absorbance wavelength
`was 208 nm (bandwidth, 10 nm) with a reference wavelength of
`400 nm (bandwidth, 80 nm). Mobile phase A consisted of 100%
`deionized water (Milli-QUV Plus; Millipore Corp., Bedford,
`MA) with 0.005 M phosphoric acid (Sigma Chemical Co.)
`buffer. Mobile phase B consisted of 100% high-performance
`liquid chromatography grade acetonitrile (J. T. Baker, Phillips-
`burg, NJ) with 0.005 M phosphoric acid buffer, and mobile
`phase C was identical to mobile phase A. All mobile phases
`were run at a flow rate of 1 ml/min for a run time of 45 min at
`the gradient profile listed on Table 1.
`A Waters Nova-Pak C18 guard column was placed in line
`before the analytical column. The samples were injected onto a
`
`
`
`clincancerres.aacrjournals.org Downloaded from
`
`on August 3, 2018. © 2001 American Association for Cancer
`Research.
`
`Par Pharmaceutical, Inc. Ex. 1007
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 3 of 10
`
`

`

`3050 Phase I Study of Phenylbutyrate
`
`Table 2 Patient characteristics
`
`Total patientsa
`Male
`Female
`Diagnosis
`Prostate, hormone-refractory
`Kidney
`Melanoma
`Endometrium
`Performance status
`0
`1
`2
`Prior therapy
`Chemotherapy, XRT, hormone
`XRT,b hormone
`Hormone
`Chemotherapy, hormone
`Chemotherapy, radiation
`Immunotherapy only
`Chemotherapy, immunotherapy
`No therapy
`a Age: range, 48 – 80, median, 66.
`b XRT, X-ray therapy.
`
`24
`23
`1
`
`19
`3
`1
`1
`
`5
`18
`1
`
`7
`6
`4
`3
`1
`1
`1
`1
`
`lyzed by nonlinear mixed effect modeling. In the pharmacoki-
`netic models used, PA was modeled as arising from PB via the
`Michaelis-Menten-type saturable elimination process for PB
`(Fig. 1). PG was modeled as arising from PA via a Michaelis-
`Menten-type saturable process. For both species, it was assumed
`that biotransformation was complete. PA was modeled as dis-
`tributing in a single compartment with elimination solely via the
`process that gives rise to PG. The Vmax of this elimination
`process was treated as increasing in an exponential fashion
`beginning 24 h after the institution of drug therapy. PG was
`modeled as distributing in a single compartment with first-order
`elimination. In one model, the elimination was treated as time
`invariant, and in another, it was modeled as increasing in an
`exponential fashion beginning 24 h after the institution of drug
`therapy.
`Urine clearance rates of PG were calculated as the amount
`of PG excreted in the urine during the period of urine collection
`divided by the area under the PG plasma disposition curve
`during the same period of time. The areas were determined
`using a linear trapezoidal rule.
`
`RESULTS
`Patients (24) were enrolled at six dose levels. Patient char-
`acteristics are listed in Table 2. A total of 89 full cycles of
`treatment were administered. All patients were evaluable for
`toxicity and response assessment. The majority of patients were
`men with advanced hormone refractory prostate cancer. Over
`half of the patients had received previous chemotherapy. One
`patient with renal cell cancer had had no previous therapy, and
`4 men with prostate cancer received PB as their first therapy in
`the hormone refractory state. The median number of courses per
`patient was four (range, one to eight) for an average time on trial
`of 84 days.
`Patients (21) were withdrawn from therapy secondary to
`progressive disease documented by radiograph or symptoms.
`
`Fig. 1 Single compartment model of PB disposition. PB is ␤-oxidized
`to PA, and glutamine binds PA to form PG, which is cleared in the urine.
`The disposition model for these three compounds is defined by a
`specific Vmax and km for the disposition of one compound to another.
`
`reverse-phase (Waters Bond-Pak C18, 3.9 ⫻ 300 mm; Millipore
`Corp) column, which was maintained at 60°C. PA, PB, and PG
`had retention times of ⬃18.2, 31.4, and 10.2 min, respectively.
`Chromatographic peak area was used for quantitation by linear
`regression analysis. The lower limit of detection for the assay
`was 5 ␮g/ml. Quality control samples were assayed at concen-
`trations of 15, 35, and 85 ␮g/ml, and the inter-day coefficients
`of variation were 11% for PA, ⬍7% for PB, and ⬍11% for PG.
`One set of quality control samples were placed before the
`calibration standards, before the patient samples, and then im-
`mediately after the patient samples were assayed. Samples from
`1 patient were assayed with each analytical run. Each analytical
`run was equivalent to ⬃26 h; autosampler stability for 35 h
`demonstrated that metabolites are stable for this period of time
`with a ⬍10% change (loss or gain) in concentration. The lower
`limit of quantitation was 5 ␮g/ml for PA and PG and 10 ␮g/ml
`for PB. PG concentrations were determined in the urine speci-
`mens, and the amount of PG excreted during the urine collection
`period was calculated as concentration ⫻ urine volume. The
`accuracy of the assay was 93–98% with precision of the assay to
`ⱕ1– 4%.
`Pharmacokinetic Methods. The plasma PB disposition
`curves were analyzed in two ways: (a) individual subject dis-
`position curves were fit by nonlinear regression (PCNONLIN;
`Scientific Consulting, Apex, NC) using a pharmacokinetic
`model consisting of a single distribution compartment and a
`first-order elimination process that increases in magnitude in an
`exponential fashion beginning 24 h after the institution of drug
`therapy, i.e., clearance rate(t ⬎ 24 h) ⫽ clearance rate(t ⫽
`0)⫻exp[kinc⫻(t ⫺ 24 h)], where kinc is the rate constant of the
`exponential process; and (b) the data were analyzed using a
`pharmacokinetic model consisting of a single distribution com-
`partment and a Michaelis-Menten-type saturable elimination
`process in which Vmax increases in magnitude in an exponential
`fashion beginning 24 h after the institution of drug therapy. For
`this model, the disposition curves were analyzed by nonlinear
`mixed effect modeling (P-Pharm; MicroPharm International,
`Champs-sur-Marne, France). With this software, population pa-
`rameter values were estimated using an EM-type iterative algo-
`rithm, and individual subject parameter values were estimated
`by MAP Bayesian fitting.
`The plasma PA and PG disposition curves were also ana-
`
`Downloaded from
`
`on August 3, 2018. © 2001 American Association for Cancerclincancerres.aacrjournals.org
`
`
`Research.
`
`Par Pharmaceutical, Inc. Ex. 1007
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 4 of 10
`
`

`

`Table 3 No. of patients with grade 3 or 4 toxicity by dose level
`
`150
`mg/kg/d
`
`225
`mg/kg/d
`
`285
`mg/kg/d
`
`410
`mg/kg/d
`
`345
`mg/kg/d
`1
`
`1b
`
`2c
`
`Neurocortical
`Hypokalemia
`Hyperuricemia
`Nausea
`Neutropenia
`Fatigue
`a Toxicities occurred in the same patient.
`b Lasted 2 days only.
`c Felt to be related to disease progression rather than drug.
`
`515
`mg/kg/d
`1a
`1a
`1a
`1a
`
`Clinical Cancer Research
`
`3051
`
`emetics, such as the phenothiazines (prochloperazine). This
`observation led to the use of nonsedating antiemetics such as the
`serotonin 5-HT3 receptor antagonists, ondansetron and granis-
`etron. Patients had less sedation and minimal emesis with this
`change in antiemetic.
`Hematological Toxicity. No dose-limiting hematologi-
`cal toxicity was noted. Minor decreases in hematocrit were
`noted over the course of therapy. No platelet toxicities were
`noted in any patient. One patient had a transient grade 3 neu-
`tropenia (absolute neutrophil count of 600) noted at dose level 1
`that resolved in 2 days. No other declines in total white cell
`count or in subpopulations were noted.
`Nonhematological Toxicity. Patients experienced mild
`grade 2 nausea at the higher dose levels, in particular, the 410
`mg/kg/day dose. Fatigue was dose related as well. Grade 2
`fatigue was noted in 1 patient at 345 mg/kg/day and 2 patients
`at 410 mg/kg/day. Grade 3 fatigue was noted in 2 of 7 patients
`receiving 410 mg/kg/day infusions. This degree of fatigue was
`not felt to be drug related but rather attributable to tumor
`progression. In fact, the 2 patients with grade 3 fatigue did not
`receive a rechallenge with PB as they were taken off study for
`progressive disease. Of note, 7 (38%) of the patients with bony
`metastatic PCA experienced a flare in bone pain that started on
`discontinuation of the infusion and persisted for 48 h. Only the
`2 patients with grade 3 neuro-cortical toxicity developed the
`characteristic odor of PA. The ability to detect the odor also
`resolved promptly with discontinuation of the study drug.
`Pharmacokinetics and Pharmacodynamics. Of 24 pa-
`tients enrolled in the study, 1 patient had incomplete blood
`sampling for pharmacokinetic data analysis, and 2 patients did
`not have adequate plasma sample volumes to allow determina-
`tion of PA and PG concentrations. These 3 patients were ex-
`cluded from the pharmacokinetic data analysis.
`In the majority of the patients, the plasma PB disposition
`curves were characterized by a rise to an apparent plateau
`achieved by 4 – 6 h into the infusion. However, a number of the
`subjects had curves that showed a substantial decline in PB
`concentrations starting around 24 h into the infusion and con-
`tinuing throughout the remainder of the infusion. This pattern
`was seen in 0 of 4 patients 150 mg/kg/day, 0 of 4 patients at 225
`mg/kg/day, 1 of 4 patients at 285 mg/kg/day, 3 of 4 patients at
`345 mg/kg/day, 5 of 7 patients at 410 mg/kg/day, and 0 of 1
`patient at 515 mg/kg/day. This pattern suggested that the plasma
`clearance of PB increased in a continuous fashion beginning
`24 h into the infusion. Fig. 2 demonstrates the PB disposition
`pattern for 2 patients, 1 at 225 mg/kg/day and the other at 515
`mg/kg/day. This feature was incorporated in the pharmacoki-
`netic models used to describe the disposition curves.
`Individual subject PB disposition curve analysis using a
`pharmacokinetic model with first-order elimination showed an
`inverse relationship between clearance rate and dose, suggesting
`that a model with saturable elimination would be more appro-
`priate. In the model that was used in the population pharmaco-
`kinetic analysis of PB disposition, elimination was modeled as
`being Michaelis-Menten in character. The population pharma-
`cokinetic estimates for PB disposition are listed in Table 4. The
`mean kinc value of 0.0029/h equates to a 1.32-fold increase in
`Vmax by 120 h into the infusion.
`The population pharmacokinetic estimates for PA and PG
`
`One patient refused additional therapy after six cycles for per-
`sonal reasons despite continued stable disease. Another patient
`discontinued therapy secondary to toxicity during his first cycle
`of therapy at 410 mg/kg/day. Lastly, 1 patient had toxicity at
`two different dose levels despite dose reduction and met criteria
`for withdrawal secondary to toxicity. This patient had a 10%
`decline in his PSA and subjective improvement in bone pain
`over the two cycles that he received therapy.
`Three patients had their dose escalated during the course of
`therapy. Two of these patients remained on the study for a total
`of eight cycles. In fact, 1 of these patients had his dose increased
`twice from 150 mg/kg/day (dose level 1) to 285 mg/kg/day
`(dose level 3). Two patients had their dose reduced secondary to
`toxicity at their starting dose during cycle 1. The starting doses
`for these patients were 345 and 515 mg/kg/day. No patient at
`410 mg/kg/day had their dose reduced or escalated and was
`determined to be the recommended Phase II dose. Table 3
`describes all grades 3 and 4 toxicities during the study period.
`Neuro-cortical Toxicity. DLT was neuro-cortical in na-
`ture. One patient on 515 mg/kg/day experienced grade 3 neuro-
`cortical toxicity ⱕ48 h of the start of drug infusion. The toxicity
`was exemplified by excessive somnolence and confusion and
`was accompanied by metabolic changes consisting of grade 4
`hypokalemia (2.1 mmol/liter), grade 3 hyponatremia (128
`mmol/liter), grade 2 hypocalcemia (7.9 mg/dl), and grade 4
`hyperuricemia (15.2 mg/dl). The metabolic nadirs occurred 4
`days after initiation of infusion. The patient also had grade 3
`nausea before the decline in mental status. Full recovery and
`return to baseline clinical state occurred ⱕ10 –12 h after dis-
`continuation of the drug at approximately the same time as the
`electrolyte abnormalities returned to normal values after aggres-
`sive repletion. The patient received a second cycle at the next
`lower dose and experienced a similar toxicity syndrome on
`rechallenge. An additional patient developed grade 3 neuro-
`cortical toxicity at the 345 mg/kg/day dose. This was not ac-
`companied by electrolyte abnormalities and occurred at 96 h
`into the infusion. The symptoms resolved 10 –12 h after discon-
`tinuing the drug. A 3rd patient developed grade 2 neuro-cortical
`toxicity with only grade 1 hyponatremia and hypocalcemia at
`410 mg/kg/day. This patient was not rechallenged at a lower
`dose because he developed obstructive urinary symptoms and
`was taken off study for progression of disease. Of note, plasma
`ammonia levels in the grade 3 toxicity patients remained in the
`normal range.
`Patients with nausea had more sedating effects from anti-
`
`Downloaded from
`
`on August 3, 2018. © 2001 American Association for Cancerclincancerres.aacrjournals.org
`
`
`Research.
`
`Par Pharmaceutical, Inc. Ex. 1007
`Par v. Horizon, IPR of Patent No. 9,561,197
`Page 5 of 10
`
`

`

`3052 Phase I Study of Phenylbutyrate
`
`Table 4 Population pharmacokinetic estimatesa
`
`Mean
`
`SD
`
`0.169
`148
`0.0029
`546
`
`PB (n ⫽ 20)
`volume of distribution, liter/kg
`Vmax, ␮mol/h/kg
`kinc, 1/h
`km, ␮mol/liter
`PA (n ⫽ 19)
`volume of distribution, liter/kg
`Vmax, ␮mol/h/kg
`kinc, 1/h
`km, ␮mol/liter
`Phenylacetylglutamine (n ⫽ 19)
`0.281
`volume of distribution, liter/kg
`1.83
`clearance rate, ml/min/kg
`a Kinc, rate constant of increase in Vmax after 24 h.
`
`0.720
`77
`0.0009
`62
`
`0.043
`23
`0.0015
`159
`
`0.562
`10
`0.0006
`28
`
`0.130
`1.15
`
`collections in which the calculated urine clearance rates were
`unphysiologically large, the mean urine clearance rate for the
`0 –24-h collection period is 116 ml/min (SD, 62 ml/min), and for
`the 96 –120-h collection period, it is 136 ml/min (SD, 97 ml/
`min). The coefficient of correlation for paired values is 0.84. A
`paired t test indicated that the increase in the mean value of the
`urine clearance rate between the first and second urine collec-
`tions is not significant (P ⫽ 0.10, two-tailed test). Previous
`studies have shown that PB is almost completely metabolized to
`PA via ␤-oxidation