(CANCER RESEARCH 49, 241-247, Januar) l, 1989]

Pharmacology Studies of 1-ß-D-ArabinofuranosyIcytosine in Pediatrie Patients with

Leukemia and Lymphoma after a Biochemically Optimal Regimen of LoadingBolus plus Continuous Infusion of the Drug1

Vassilios I. Avramis,2 Kenneth I. Weinberg, Judith K. Sato, Carl Lenarsky, Michael L. Willoughby,3

Thomas D. Coates, M. Fevzi Ozkaynak, and Robertson ParkmanDivisions ofHematology/Oncology and Research Immunology and Bone Marrow Transplantation, Department of Pediatrics, School of Medicine, University of SouthernCalifornia, Childrens Hospital of Los Angeles, Los Angeles, California 90027

ABSTRACT

In an attempt to maximize the therapeutic index and to overcome thelarge variations in l-/3-D-arabinofuranosylcytosine (ara-C) plasma levelsand host toxicities that have been documented with standard HDara-Cregimens (3 g/m2 over 3 h every 12hx8orxl2 doses), pediatrie patients

with acute lymphocytic leukemia or lymphoma in relapse were treatedwith a regimen of loading bolus followed immediately by continuousinfusion of ara-C. In addition, patients received a single dose of etoposide(VP-16, I g/m2) prior to the ara-C administration. In four patients, total

body irradiation was administered as part of a bone marrow transplantation preparative regimen after the ara-C administration. The regimenwas designed to attain and maintain plasma steady-state concentrations(C,,) of ara-C three to four times the Km¡value of ara-C, which wasdetermined with purified deoxycytidine kinase from the patients' tumor

cells prior to treatment. Eight patients age 0.75 to 16 years with relapsedacute lymphocytic leukemia (three patients) or lymphoma (five patients,one with bone marrow involvement), received a test dose of 3 g/m2 ara-C

injected over I h, and the plasma kinetics were determined. The peakplasma ara-C concentration of ara-C ranged from 57 to 199 MMwith anaverage concentration of 103 ±49 ¿i\l;the half-lives of distributionCi .-.„)and elimination (/, ;. ) averaged 17 ±7 min and 4.04 ±3.1 h,respectively. The mean area under the plasma concentration time curvefrom 0 to 12 h (AUCo_,2h) of ara-C averaged 386.8 ±328.0 pMh (mean,±SD, n = 8). The peak concentration of uracil arabinoside averaged 501±123 *iM,and it was eliminated with a t, ?...,of 2.3 ±0.6 h. The patientsthen received an individualized loading bolus (mean = 0.5 g/m2) followedby a continuous infusion regimen of ara-C (mean = 130 mg/m2/h), to

achieve a Ca in the range of 20 to 35 JIM.The obtained plasma C,, weresimilar to the desired ones, averaging in variation 10.7% ±8.2%. Thepercentage of variation of correlation of the Ali following the loadingbolus plus the continuous infusion from 12 to 72 h was only 12.4% (mean= 2158 >tMh,n = 8), whereas the percentage of variation of correlationof the AUC after the test dose of ara-C in the same patients was 84.8%.The total dose of ara-C administered with this regimen averaged 41.4 ±16.0% (mean ±SD, n = 8) of the dose that the patients would havereceived if they had been given the standard HDara-C regimen (3 g/m2)

for the same duration of treatment. Both of the évaluablepatients withlymphoma undergoing bone marrow transplantation achieved a completeremission after ara-C, etoposide and total body irradiation. In a non-bonemarrow transplantation trial one complete remission and one partialremission were observed among three patients with relapsed acute lymphocytic leukemia and one patient with lymphoma that had relapsed inthe bone marrow. This regimen uses a loading bolus followed immediatelyby continuous infusion, which administers an intermediate dose level ofara-C, achieves a uniform drug exposure (AUC) in each patient, withfewer variations in plasma concentrations. This new ara-C regimen shows

Received 2/29/88; revised 8/15/88; accepted 10/5/88.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 inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

' Supported in part by Grant CA 38905 from NIH, National Cancer Institute,

and by grants from The Upjohn Company and from the T. J. Marteli Foundation.2To whom requests for reprints should be addressed, at Division of Hematol-

ogy/Oncology, Childrens Hospital of Los Angeles. 4650 Sunsel Boulevard. LosAngeles, California 90027.

•'Present Address: The Princess Margaret Hospital for Children, Perth, West

ern Australia.

promising antitumor activity with tolerable toxicity and can be administered as part of a combination therapy.

INTRODUCTION

ara-C4 is a pyrimidine nucleoside analogue with proven an-

tileukemic activity against animal and human leukemias (1-6).Earlier studies in pediatrie patients showed that intra- andinterpatient plasma ara-C concentrations vary greatly afterchildren received an HDara-C (6-8). These studies also showedthat the pharmacological parameters of the pro-drug ara-C, didnot correlate with clinical response (6, 7). The pharmacody-namic studies of the active anabolite of ara-C, ara-CTP, havealso been studied in circulating blast cells in pediatrie patientswith ALL and acute nonlymphocytic leukemia, and some ofthese parameters, such as inhibition of DNA synthetic capacityand tumor cell kill, appear to correlate with response (6). Thevariations in the cellular ara-CTP concentrations that have beendocumented in ALL patients have been attributed to differencesin the ara-C levels in biological fluids and to the variability ofthe activating kinase, dCk, that activates ara-C (6-12).

In order to improve the therapeutic index of HDara-C, thisregimen has been combined with L-asparaginase (6, 13), an-thracyclines (14), and amsacrine (15, 16) in patients with relapsed acute leukemia. In a similar manner, combination regimens using ara-C and etoposide or the related compoundteniposide, have been studied as a means of increasing theefficacy of ara-C (17, 18). Additionally, etoposide has shownefficacy in lymphoma and leukemia both as a single agent andin combination with ara-C (18). We have been studying thecombination of ara-C and etoposide as therapy for leukemiaand lymphomas.

ara-C is transported into the cell by carrier-mediated facilitated diffusion (17, 19, 20), where it is phosphorylated by dCkto ara-CMP and then to ara-CTP, the active metabolite (3, 4,6, 21). Studies of human leukemic cells during early relapsehave shown that this activation process is saturable at exogenous ara-C concentrations greater than 10 nM (3, 22, 23). Asimilar mechanism of saturable activation of ara-C occurs inmurine L1210 leukemia (24). The peak plasma concentrationsof ara-C during HDara-C therapy have consistently yielded 5-to 100-fold higher concentrations than the saturation concentration of ara-C (6, 25, 26). Recent studies have shown thatdCk is subject to substrate activation by ara-C and other pyrimidine nucleosides (27). Substrate activation yields two Km values, one for the low range and the other for the higher range of

4The abbreviations used are: ara-C. I-f)-D-arabinofuranosylcy tosine: ara-U.uracil arabinoside: ara-CTP, cytosine arabinoside 5'-triphosphate; dCk, deoxycytidine kinase: HDara-C, high dose ara-C; ALL, acute lymphocytic leukemia;AUC, area under the plasma concentration-time curve; HPLC, high-performanceliquid chromatography; A",,,amount of drug to achieve a steady-state concentration in plasma; Ca, steady-state concentration in plasma: I,,.. volume of distribution of the central compartment; A„,constant rate of infusion: C„,concentrationof drug in plasma at time zero, TBI, total body irradiation, BMT, bone marrowtransplantation.

241

on April 9, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

PHARMACOLOGY OF ara-C IN PEDIATRIC PATIENTS

substrate concentrations (9, 27). The Km values for ara-C ondCk from patients' leukemic cells ranges from 1 to 16 MM(9,

11,23).A pharmacologically directed study with intermittent short-

term infusions of HDara-C showed that even though there isconsiderable variability in both plasma ara-C and cellular ara-CTP concentrations, it is possible to attain and maintain theminimum therapeutic concentration of ara-CTP in the blastsof 75 UM(4). Earlier studies with continuous infusion of ara-C,where the administered dose was based on body surface area,reported wide variations in the plasma ara-C concentrationsattained (28, 29). The plasma ara-C concentrations without aloading bolus ranged from 3.6 to 22.6 ßMafter 1.33 g/m2/dayto 6 g/m2/day for 3 days, and further escalation was prohibited

by combined organ system side effects, which included hepatic,pulmonary, renal, and gastrointestinal tissues (29).

In an effort to avoid the toxicities associated with the highpeak levels (8) and to counteract the fluctuations of plasma ara-C concentrations seen during HDara-C administration, a regimen consisting of a loading bolus followed immediately by acontinuous infusion of ara-C was designed to attain and maintain a desired plasma steady-state concentration (Css) of ara-C.To saturate the activation of ara-C to ara-CTP, the plasma Cssconcentrations were chosen to be three to four times the apparent Km2 value of ara-C, which was determined with purifieddCk from the individual patients' malignant cells prior to

treatment (9). We report here pharmacokinetic studies of ara-C and its deamination catabolite, ara-U in pediatrie patientsafter such a biochemically directed regimen.

MATERIALS AND METHODS

Patients. During a period of 18 months, eight children age 0.75 to16 years, three with ALL and five with lymphoma in relapse, weretreated with a loading bolus followed by a continuous infusion of ara-C. Each course of ara-C was preceded by a single dose of VP-16, l g/m2. Four patients with lymphoma also received TBI as part of the

preparative regimes for bone marrow transplantation. Two of thepatients with ALL did not have any appreciable circulating blast count,thus no cellular anabolism studies of ara-C were performed. The thirdchild with ALL was treated at another hospital and the cellular studieswere not performed. Five patients, two with ALL and three withlymphoma, had previously received ara-C during induction chemotherapy.

Treatment Protocols. Four children with advanced lymphoma received the loading bolus followed by continuous infusion of ara-C aspart of a conditioning regimen for bone marrow transplantation. Thefirst patient treated (Patient 1. Table 2) received a loading bolus and a6-day infusion of ara-C and developed fatal noncardiogenic pulmonaryedema thought to be due to ara-C toxicity. The dose and duration ofinfusion of ara-C was reduced for subsequent patients. The treatmentregimen consisted of etoposide (1 g/m2) on Day 0, followed by a testdose followed by a loading dose and continuous infusion of ara-Clasting 60 h on Days 1 to 4. TBI, at 200 rads x6 (1200-rad total) wasgiven on Days 5 through 7 with the marrow infusion given on Day 8.One patient (Patient 5, Table 2) who had previously received high dosesof chest and abdominal irradiation, was given cyclophosphamide (60mg/kg) and reduced TBI, 200 rads x3 (600 rads instead of 1200 rads).Patient 2 (Table 2), who relapsed soon after his first BMT, received asecond transplant, using a modified preparative regimen with a reducedTBI dose (800 rads) and lung shielding. All BMT patients were hospitalized in a protective sterile environment until hematopoietic andimmunological reconstitution was achieved.

Two children with acute leukemia in relapse and one child withrelapsed lymphoma who had bone marrow involvement were hospitalized in the oncology unit and received a loading bolus and continuousinfusion of ara-C 24 h after VP-16 dosing as remission induction

therapy. An additional patient with ALL (Patient 8, Table 2) was nottreated at Childrens Hospital of Los Angeles and thus individualizedadjustment of drug dosing from pharmacokinetic studies could not beperformed. This patient was treated with loading bolus and continuousinfusion doses derived from other patients of comparable size and age,after adjusting for surface area. Blood specimens were obtained fromthis patient and analyzed post hoc for plasma ara-C levels.

Criteria for Response and Toxicity. Complete remission was definedas the absence of all symptoms and physical signs due to leukemia orlymphoma cells in the bone marrow (<5% lymphoblasts) and radio-graphic studies showing absence of lymphoma nodules. Partial remission was defined as the complete disappearance of symptoms andphysical signs of leukemia or lymphoma, with <25% leukemia orlymphoma cells in the bone marrow, and a >50% shrinkage of definablelymphomatous masses. No evaluation of efficacy or toxicity was possible if the patient (Patient 1, Table 2) expired within 7 days ofcompletion of chemotherapy. Toxicity was graded by standard WorldHealth Organization criteria with Grade O = none, Grade I = mild,Grade II = moderate, Grade III = severe, and Grade IV = life-threatening. AH patients were treated in concordance with clinicalprotocols approved by the Childrens Hospital of Los Angeles Committee on Clinical Investigations (Institutional Review Board) and informed consent from the patient and/or the patients' parents were

obtained.Materials and Drugs. Cytosine arabinoside (ara-C, Cytosar-U) was

purchased from The Upjohn Co., Kalamazoo, MI. Ficoll-Hypaque,specific density 1.083 g/ml (Histopaque 1083) was purchased fromSigma Chemical Co., St. Louis, MO. Radioactive [3H]ara-C and [3H]-

dCyt were purchased from Moravek Biochemicals Inc., Brea, CA. Allother chemicals for the kinetic studies, extractions and HPLC assayswere of analytical or HPLC grade.

Pharmacokinetic Considerations. The plasma concentration data ofara-C after a test dose of 3 g/m2 ara-C were fitted to a two-compartmentopen model and the data of ara-U were fitted to a one-compartmentopen model. The exponential equations, including the constant ratesof distribution and elimination, which describe the kinetics of the twodrugs were derived by a pharmacokinetic analysis least-squares program. The volume of distribution of ara-C in the central compartmentwas calculated from the equation:

V* = Dose/Co, where C, = A + B (A)

and A and B are the abscissa intercepts of the best fit computer estimatedtheoretical lines describing the distribution and elimination phases ofara-C in the body. The AUC was calculated utilizing the Simpsons'

approximation rule (logarithmic trapezoidal method) for the first 12 h.The AUCi2-72h was calculated utilizing the trapezoidal method.

The i.v. loading bolus to achieve a selected steady-state concentrationof ara-C was calculated from the equation:

Xa = Ca.Y« (B)

where X„is the loading bolus in mg and C„is the selected steady-stateconcentration of ara-C.

The constant rate of infusion of ara-C to maintain the selectedconcentration of ara-C at steady-state was derived from the equation:

k = C . K, ñ IOKo t- u ' de P \*~J

where ßis the constant rate of elimination of ara-C and k0 is the rateof infusion and is expressed in mg/h of ara-C. There is an assumptionbeing made in this equation, in that I',/, approximately equals the

volume of distribution of the total body ( F»)which for a highly hydro-philic drug, such as ara-C, is correct within an acceptable margin oferror.

Isolation of Human Lymphoblasts and Lymphoma Tumor Cells. Theperipheral blood leukocytes were isolated using Ficoll-Hypaque gradients as described previously (6). Lymphoma cells were obtained bysurgical biopsy of nodules under general anesthesia. Bone marrow blastcells from three patients, two patients with ALL and one with lymphoma who had bone marrow involvement, were collected in heparin.

142

on April 9, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

PHARMACOLOGY OF ara-C IN PEDIATRIC PATIENTS

The bone marrow aspirate was immediately placed in an ice bath (0°C)

and transported to the laboratory- Heparinized peripheral blood specimens for ara-C and ara-U levels were collected in test-tubes containing

10 MMtetrahydrouridine, a cytidine deaminase inhibitor (24), to inhibitfurther deamination of ara-C in vitro.

HPLC Determination of the Nucleoside Analogues ara-C and ara-Uin Plasma and the Anabolite ara-CTP in Leukemic Cells. The nucleosideanalogues ara-C and ara-U were assayed by HPLC on a reverse-phaseMBondapak CIS column (Waters Associates, Milford, MA) as describedpreviously (6, 24). ara-CTP, the active anabolite of ara-C, was assayedin perchloric acid extracts of leukemic cells on a strong anion-exchangeresin column as described previously (6, 24).

Purification of dCk by HPLC Protein Columns. The tumor andhealthy WBC cell dCk was purifiai as described previously (9, 10).Briefly, the cells were lysed by freezing at -70°C and thawing to 0°C

in an ice bath. 1 ml of Tris-HCl 50 mM + 10 mM MgCl2 + 1 mMdithiothreitol (Buffer A) was added per 2 x 10" cells and the cellsuspension was sonicated for 2 min at 0°Cwith a Sonifier, Danbury,

CT. The nucleic acids were precipitated with streptomycin sulfate, 0.3/l V/V of 5% solution and centrifugation for 20 min at 20,000 x g in aSorvall RC5-B high speed refrigerated centrifuge at 4°C.1 ml of the

crude cell homogenate was injected onto an analytical cation-exchangeprotein column (Pro Pak SR, Waters Associates, Milford, MA) andthe proteins were eluted with an isocratic elution over 30 min. Theelution conditions were KH2PO4. 35 mM (pH 6.80). containing 1 mMMg2+ (MgClz), and 0.1 mM /3-mercaptoethanol at 1 ml/min flow rate.

The UV absorbance of the column eluates were followed at 280 nm.The HPLC column eluates were collected in test tubes placed in an icebath by a programmable fraction collector (Gilson, model 201 ). Proteinwas assayed with the Bio-Rad micro-method (30). The dCk activity wasassayed as described previously (6, 10). The test tube aliquot sampleswith the most dCk activity were combined and injected onto a secondHPLC size exclusion protein column with weak hydrophobic interactions, TSR G3000 SW (Varian Micropak, Varian, Sunnyvale, CA) (10).The elution buffer remained the same but the flow rate was reduced to0.5 ml/min for better resolution of dCk and to avoid high back pressuresin the column. The eluates were collected again with the fractioncollector and the procedure of protein and dCk determination in eachsample was repeated. A 1000-fold of protein loss was observed with thepassage through both protein HPLC columns, whereas the dCk activitywas maintained to 65-70% and 90-95% in the absence and presenceof 30% saturation with (NH4)2SO4, respectively. The time required forthis procedure was 2 to 3 days, thus reducing drastically the deterioration of enzymatic activity. Overall a >2000-fold purification of dCkwas achieved with this method (9).

Kinetic Characterization of dCk with ara-C and dCyt as Substrates.Various concentrations of [3H]ara-C and [3H]dCyt were incubated with

the purified dCk in the presence of ATP and the phosphorylationproducts were absorbed onto aniónexchange resin filter disks DE 81(Whatman, Maidstone, England) as described previously (6, 9, 10). Thedouble reciprocal plots of velocity versus concentration of substratewere bimodal in nature yielding two K„values (9, 27).

RESULTS

Deoxycytidine Kinase Characterization with dCyt and ara-Cfrom Normal Peripheral Blood Leukocytes (WBC), LeukemicCells (ALL), and Lymphoma Cells. The lymphoma was surgically removed from the patient (Patient 2, Table 2) and theleukemic cells were isolated by Ficoll-Hypaque gradients. Table1 depicts the kinetic data of dCk from normal WBC, leukemiccells (ALL), and lymphoma cells. Purified dCk was subject tosubstrate activation by both dCyt and ara-C, thus yielding twoKm values (Fig. 1). The second Km (Kmi) value of ara-C rangedfrom 2.7 to 9.7 /¿Mand for dCyt, it ranged from 0.7 to 1.25 I¿M(Table 1). The Fmax2of ara-C was higher in the lymphoma andhealthy WBCs when compared with dCyt (Fig. 2 and Table 1).The Kmax;of ara-C in the two patients with ALL was substantially lower than the Vma%:iof dCyt (Table 1). The Kmaxwas

attained at 20 /iM or higher ara-C concentrations from the

lymphoma and WBC origin dCk (Fig. 2). A whole cell assay ofara-C activation in malignant cells from a patient (Patient 6,Table 2) showed that ara-CTP cellular concentrations reached

a peak at 20 nM and declined at higher concentrations ofexogenous ara-C, probably due to drug-induced cell death (Fig.

3).Pharmacokinetic Studies of ara-C in Pediatrie Patients after

High Dose ara-C (3 g/m2) Administration as a Test Dose. Theconcentration of ara-C measured in plasma, after pediatriepatients received a test dose of 3 g/m2 of ara-C when injected

over l h i.v., declined rapidly following a biexponential decaycurve over time (Fig. 4). There is considerable variability in thepeak plasma concentration of ara-C in these patients after thetest dose of HDara-C (Table 2). The area under the plasmaconcentration-time curve (AUC) has an even larger variability,due to variations in the terminal half-lives (f1/2.0)of eliminationof ara-C in pediatrie patients. The half-lives of distribution andelimination of ara-C averaged 17.4 ±7 min and 3.82 ±2.97 h,respectively (Table 2). Plasma ara-U concentrations averaged501.4 ±123 /iM, and it was eliminated from plasma monoex-ponentially with an average fi/2,ei of 2.3 h. The volume ofdistribution ( yd) of ara-C in the patients was estimated as 88.87±46.63 liters/m2 (mean ±SD, n = 8).

Pharmacokinetic Studies of ara-C in Pediatrie Patients afterthe Loading Bolus followed by a Continuous Infusion of ara-CRegimen. 12 h after a test dose of ara-C, the patients were givena loading bolus followed immediately by a continuous infusiondetermined by the patient's individual volume of distribution

and the slope of the elimination phase, ß,of ara-C. The amountsof loading bolus, the continuous infusion rate and the Cssplasmalevels attained in each patient are depicted in Table 3. Theplasma Cs, of ara-C was arbitrarily chosen for the first patient(70 Õ/M)and the duration of infusion was 6 days. Because oftoxicity seen in this patient (Patient 1, Table 2), subsequentpatients received a regimen designed to achieve a C„of 20 UMara-C, and the duration of infusion was reduced to 60 h. TheCss of 20 fiM was chosen because it was approximately theconcentration at which Fmaxwas obtained in patient's malignant

cells (Fig. 2). A representative pharmacokinetic profile ofplasma ara-C data is depicted in Figure 5.

The plasma AUC of ara-C, after the loading bolus and duringthe infusion of ara-C, from 12 to 72 h was relatively uniformin the patients with an average % CV of 12%, when normalizedfor the differences in the CM(Table 3). The % CV in the AUCafter the test dose of ara-C in the same patients was approximately 85% (Table 2).

One patient (Patient 8, Table 2) was treated with 185-mg

loading bolus plus 50 mg/h infusion rate based on the datacollected from other patients with similar body surface area, toempirically achieve a plasma C„of20 UM.This patient had anaverage Cssof ara-C of 14.6 = 7.1 ^M without any dose adjustments and had on average 27% variation from the theoreticalCMof 20 (¿M.This patient achieved a short term completeremission. In the other patients' adjustments of either the

infusion rate and/or a second mini-bolus was necessary to loweror elevate the plasma Ca of ara-C so that it would approximatethe selected plasma ara-C level. This effort yielded a lowervariation on the achieved Ca in comparison to the chosen ones,with a range from 2.5 to 20.0% and an overall average of 10.7±8.2% (mean ±SD, n = 8). This regimen achieved andmaintained the selected Ca with a reduced total dose of ara-C,averaging 41.4 ±16.1 % in comparison with the dose that would

243

on April 9, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

No.

PHARMACOLOGYOF ara-C IN PED1ATRICPATIENTS

Table 1 Kinetic parameters of ara-C and dCyt substrates on deoxycytidine kinasefrom lymphoma, leukemic cells, and healthy WBCRelative efficiency*

Cells Substrate No. 1 No. 2

1.2.34.5.6.7.8.WBCcellsWBCcellsLymphoma

(P.B.)Lymphoma(P.B.)ALL

cells (S.G.)ALLcells (S.G.)ALLcells (S.H.)"ALL

cells (S. H.)dCytara-CdCytara-CdCytara-CdCytara-C0.540.190.161.100.211.350.461.252.700.769.702.256.401.864.4056.598.745.6111.70.970.484.1829.9137.645.8293.61.941.0033.312.7100.0496.5272.497.1100.049.6C100.0213.1251.9126.5100.051.5C100.015.2'°nmol/h-mg protein.* Relative efficiency is the ratio of Km„//f„;the values of dCyt are set as 100.0.' % of dCyt value from the same patient.'' The patient was originally diagnosed with lymphoma. At the time of treatment she had bone marrow involvement; the BM cells appeared as ALL cells on

microscopic examination.' % of dCyt value from the same patient.

1750

| 1000

-£-750

i 500

250,

Km,=0.46 UM

Km,=4.40uM

-2.2 124 10l /[S], [3H]ar<j-C, UM'1

Fig. 1. Lineweaver-Burk plots of dCk activity with ara-C as the variablesubstrate. The initial velocities were determined by varying the ara-C concentration and by using HPLC-purified tumor (leukemic) cell dCk at fixed concentrationof ATP and Mg2* in a 100-fold excess of ATP. The reaction volume was 15Xand

the incubation time was 1 h.

Lymphoma^C~

0 2 4 ó8 !O 20[SIMM

Fig. 2. Activation of ara-C and dCyt by a purified preparation of dCk from alymphoma tumor and healthy white blood cells (WBC). The assays were run asdescribed in "Materials and Methods" for 1 h in a total reaction volume of 15X

and in excess of ATP.

have been administered after the standard HDara-C for thesame duration of time.

A correlation exists between surface area of patients and the

loading bolus in these patients with a correlation coefficient R2

of 0.76. The equation that describes the function of the loadingbolus estimates 500 mg/m2 as the average dose (Table 3). Thecontinuous infusion rate in a similar manner had an A2 value

of 0.84 and the equation estimates an average k0 of 130 mg/m2/h.

Clinical Response and Host Toxicities. The clinical data onall eight patients are summarized in Table 4. Two of the fourpatients who received a biochemically optimal loading bolusplus continuous infusion of ara-C as part of a BMT preparativeregimen for lymphoma, were évaluablefor clinical response.Both évaluablepatients attained a complete remission. One ofthese children (Patient 2, Table 4) relapsed 2 months afterBMT and a second transplant was performed after which heagain quickly relapsed. Patient 1, Table 4, expired of toxicityand was not évaluable;however, at autopsy no evidence oflymphoma was present. Patient 4, Table 4, received his BMTwhile in second remission but relapsed 2 months post-BMT.

One complete remission and one partial remission wereobserved among the three ALL patients and the one lymphomapatient with bone marrow involvement, who received VP-16and loading bolus plus continuous infusion of ara-C as remission induction therapy. The complete remission was observedin a 3-year-old boy (Patient 8, Table 4) in second bone marrowrelapse with ALL, which had not previously responded to highdose ara-C (3 g/m2 every 12 h X8 doses) plus i.-asparaginase.The partial remission was seen in a 16-year-old girl with non-Hodgkin lymphoma with marrow involvement. No responsewas observed in a child with acute lymphocytic leukemia and a7-month infant with congenital leukemia of mixed lineage.

Toxicity in these patients was predominantly myelosuppres-sion, Grades III and IV. Myelosuppression in the three patientswho did not receive a marrow transplantation was profound

Table 2 Pharmacokinelic parameters of ara-C and ara-V in pediatrìepatients after a single 3 g/nf dose of ara-C

No.1.2.3.4.5.6.7.8.9.Mean

±SDPatientB.

P.°P.

B.P.B.cS.

G.R.P.S.H.°C.

L.A.V.S.

G."DiseaseLymphomaLymphomaLymphomaLymphomaLymphomaLymphoma'*ALLALLALLSurface

area(m2)0.61.01.01.21.32.00.30.50.87Peak[ara-C]

OIM)78.5136.884.260.873.7199.057.0132.4134.0106.3

±47.2r»,„

(min)22241128917IS1317.4

±7.0fvw

(h)11.10*2.571.802.822.501.765.232.184.463.82

±2.97AUC„-,,2h

(¡M,h)362.8360.8182.7451.7188.4379.0160.11208.4168.9384.8

±328.0Peak

[ara-U](CM)514.9473.8484.4442.5482.3319.0758.4535.6501.

4 ±122.8ÕM

(h)2.781.802.842.071.811.502.802.762.30

±0.56" Patients had not received prior treatment with ara-C.* Statistical outlier, not considered for the estimation of the mean.c Second treatment.

The patient had bone marrow involvement at the time of treatment.

244

on April 9, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

PHARMACOLOGY OF ara-C IN PEDIATR1C PATIENTS

1.300

gC 200o

¿ 100

_o=i ou O 20 50 100

[ora-C], MM

Fig. 3. Activation of ara-C by dCk from a leukemic cell population as determined by a whole cell assay. The cells were incubated with various concentrationsof ara-C for 1 h. The cells were washed and extracted with perchloric acid. Theacid-soluble neutralized extract was assayed by a strong aniónexchange HPLCcolumn for NTPs and ara-CTP as described in "Materials and Methods."

1000

i.8 100

10

01357Time, hours

Fig. 4. Kinetics of ara-C and its catabolite ara-U in patient plasma after a 3g/m2 test-dose of ara-C. The nucleoside analogues were assayed by reverse-phase/jC18 HPLC column as described in "Materials and Methods."

with absolute granulocyte counts of 0 cells/mm3 lasting a mean

of 18 days before recovery to absolute granulocyte counts >300.Thrombocytopenia with platelet count less than Ix IO4 platelets/mm3 was also observed and showed a similar rate of recov

ery.The BMT patients and one child with ALL were hospitalized

in sterile environments and had no infectious complications.The three other patients, who were hospitalized on generalpediatrie oncology units, developed either documented bacte-remia or fever and neutropenia requiring empiric treatmentwith antibiotics. As previously noted, the first patient treatedwith a BMT regimen developed fatal pulmonary toxicity aftera 6-day ara-C infusion at an approximate steady-state concentration of 70 fiM.

No other patient had pulmonary or skin toxicity. In contrastto our experience with conventional HDara-C regiments forBMT, in which Grade II to IV mucosal toxicity is usuallyobserved, no BMT patient had greater than Grade II mucosaltoxicity with this regimen. One patient (Patient 6, Table 4)developed severe cholestatic liver disease and had a transient

cerebrovascular accident a week after ara-C chemotherapy,which may have resulted from either the chemotherapy or thehigh dose intravenous steroids that she was receiving duringthis time for massive uterine bleeding. Except for this possiblecase, no central nervous system toxicity, typically seen withHDara-C regimens, was seen with this loading bolus pluscontinuous infusion ara-C regimen.

DISCUSSION

ara-C administered as a high dose regimen (HDara-C) is oneof several treatments available for patients with acute leukemiain relapse (6, 7, 12). Studies examining the rate of eliminationof ara-C from patient plasma and ara-CTP from leukemicblasts, as well as the ara-CTP AUC in circulating blasts, haveshown that these pharmacological parameters differ significantly in individual patients, and none were predictive forclinical response to the drug regimen (6, 7). The variation maybe due to the net result of the following variable processes: Thesaturation of ara-C phosphorylation by dCk in the blast cells,the different levels of cytidine deaminase activity in the host,and the different rates of ara-C elimination from patients'

plasma.The primary objective of this study was to demonstrate that

it is plausible to attain and maintain a desired plasma steady-state concentration of ara-C by utilizing pharmacokinetic principles. The present study documents that the AUC of ara-C canbe maintained relatively constant in individual patients in spiteof the differences in deamination and elimination rates of thedrug in pediatrie patients.

ara-C is phosphorylated in the cells by dCk to ara-CMP,which is believed to be the rate-limiting step in the cascade ofactivation of ara-C to ara-CTP (20, 31). ara-CMP is phosphorylated by pyrimidine nucleoside monophosphate kinase to thediphosphate and this anabolite in turn is phosphorylated bypyrimidine nucleoside diphosphate kinase to ara-CTP (31, 32).The affinity of dCk for ara-C is nearly 10-fold higher than thatof the pyrimidine nucleoside monophosphate kinase for ara-CMP (Km = 6.8 x IO2 MM),but the later enzyme is found in100-fold higher levels than those of dCk in leukemic cells (31).The difference in the enzymatic content suggests that dCk isthe rate-limiting step in the activation of ara-C to ara-CTP andis the appropriate target for efforts to optimize ara-C administration.

Our regimen was designed to give plasma steady-state concentrations of ara-C sufficient for its optimal activation to ara-CTP, its active anabolite. The information for the optimal

Table 3 Pharmacokinetic parameters of loading bolus plus continuous infusion of ara-C regimen in pédiatriepatients with cancer

PatientB.

P.P.B.P.B.'S.G.R.

P.S.H.C.L.A.V.S.G.Mean

±SDSurfacearea(mj)0.61.01.01.21.32.00.30.50.87Loading

bolusofara-C(mg)5033651247510850961228185293*„rate*(mg/h)329628080225378415095SelectedCu

(MM)702035202025202026ObtainedC„GIM)65.6

±40.418.2±4.331.0±11.419.1

±3.619.3±3.228.0

±8.319.5±5.814.6±7.120.8±4.9Variation

%ofselectedCa6.3%9.0%11.4%4.5%3.5%12.0%2.5%27.0%20.0%10.7

±8.2%AUC|2—

72h(cMh)6793.02239.13237.22278.92290.93307.52348.41900.62511.12989.6

±1500.4AUC12-.,2hnormalized'(/iMh)1940.02239.11849.92278.92290.92646.02348.41900.61931.62158.5 ±267.6%

ofdoseequiv.ofHDara-C*22.434.0100.324.561.365.749.835.438.341.4

±16.1°The relationship between body surface area and loading bolus is: R2 = 0.76, ¥= 40.07 + 460.03.V.* The relationship between body surface area and CI rate is: R2 = 0.84, ¥= (-71.2) + 201.6A'.' Area under the plasma concentration-time curve normalized for 20 «Mplasma Ca.''Equivalent of 3 g/m: every 12 h for a total time-period of 72 h.' Second treatment.

245

on April 9, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

PHARMACOLOGY OF ara-C IN PEDIATRIC PATIENTS

bolus 365 mg oro-C

0 12 24 36 48 60 72Time, hours

Fig. 5. Kinetic profile of the test-dose of ara-C followed by a loading bolusplus a continuous infusion rate of ara-C at 12 h after the treatment began. Arrows,adjustment of the constant rate of infusion as it was suggested by the phartnaco-kinetic parameters and the plasma Ca of ara-C for this individual patient.

activation of ara-C was derived from the tumor dCk kineticcharacterization from these patients. The plots of inverse concentration versus inverse velocity for ara-C or dCyt on dCkfrom patient's malignant cells were bimodal in nature, thus

yielding two Km values. This is supported by other studiespublished previously on this enzyme (27). The Michaelis-Men-ten rate constants for ara-C were higher from tumor (lymphomaand leukemic cells) than from healthy white blood cell dCk,ranging from 0.46 to 9.70 ¿IMand from 0.19 to 2.70 /UMin thetumor and the WBC cells, respectively. The data are in agreement with Kmvalues for ara-C reported elsewhere, which rangedfrom 1 to 16 ¡IM(11, 23, 27). The plasma concentrations ofara-C were chosen to be in the range of 20 to 35 /¿M,in orderto saturate the activation of the drug to ara-CTP. The data inFig. 3 further supported this idea. It was demonstrated that inblast cells isolated from a bone marrow specimen from a patientthe accumulation of ara-CTP was saturated at 20 pM extracellular concentration of ara-C, after the cells were incubated withthe drug for 1 h. The K2mvalue of ara-C was 4.4 ¿iMin thistumor cell population. The apparent decline in cellular ara-CTP concentrations after the higher concentrations of ara-Cmay be due to drug-induced decline in the nucleoside triphos-

phate pools and subsequent cell death.Significant variations were observed in peak plasma ara-C

concentrations after the test dose of HDara-C. The half-life ofelimination (fi/2.t<),and the AUC of ara-C in plasma had %CVranging from 78 to 85% in these patients (Table 2). Despite allthese variables, the regimen of loading bolus followed by continuous infusion of ara-C, maintained steady-state plasma con

centrations of ara-C with an average variation of 10% from theselected Css. This was responsible for the rather uniform AUCfrom 12 to 72 h with a %CV of 12%, after accounting fordifferences in the CM in these patients (Table 3). The resultsalso showed that even though there was a wide variation in thepharmacokinetic parameters after the test dose of ara-C in thesepatients, the regimen of loading bolus plus continuous infusionof ara-C achieved the desired steady-state plasma concentrationof the drug at a fraction of the total ara-C dose that the HDara-C (3 g/m2) regimen would have administered for the same time

period, averaging 41.4 ±16% (Table 3). This regimen, whichadministered an intermediate dose of ara-C, showed weak correlations between the surface area of the patients and either theloading bolus or the constant rate of infusion of ara-C (Table

3).Previous investigators have administered continuous infu

sions of ara-C to patients with refractory leukemia (28, 29).Ochs et al. reported the results of a 4-day continuous infusionfollowing a loading dose of 0.5 g/m2 ara-C (28). In contrast toour results, the doses were calculated on the basis of patients'

body surface area rather than the actual plasma concentrationof ara-C to be achieved. As expected, wide interpatient variations in the actual steady-state plasma concentrations of ara-Cwere noted. Attempts to escalate the dosage resulted in unacceptable nonhematopoietic toxicity. Donehower et al. reportedon a trial of continuous infusion of 4 to 18 g/m2/72 h of ara-C

using a dose determined by body surface area (29). In theirstudy no loading bolus was given and the mean ara-C levels

achieved were low (3.6 to 22.6 UM). They also reported thatnonhematopoietic toxicity, such as hepatic, pulmonary, renaland gastrointestinal, limited further escalation of doses thatcould be administered.

Our efforts differ in several significant ways from the previousstudies. Because of the wide variations in pharmacokinetichandling with ara-C that we and others have observed, theindividualized doses we administered were calculated to providethe desired plasma steady-state concentration of ara-C. The useof the loading bolus plus the continuous infusion resulted inessentially constant plasma steady-state concentration of ara-Cwith minor pharmacokinetic adjustments. Secondly, the desiredplasma concentration of ara-C was based on actual biochemicaland pharmacodynamic determinations from the patients' ma

lignant cells and from the test dose of ara-C administered inthe beginning of therapy. The individualized administration ofara-C could thus be based on kinetic characteristics of the

Table 4 Clinicalresponseand host toxicitiesof treatmentregimenof etoposidefollowedby ara-C in pediatrìepatients withleukemiaor lymphoma

No.1.2.

3.

4.

5.

6.PatientB.

P.P.

B.

P. B.*

S.G.

R. P.

S. H.Age2yr

8yr

8yr

12 yr

14 yr

16 yrDisease,

statusBurkitt

lymphoma 2ndrelapseLymphoblastic

lymphoma, 2ndrelapse

Lymphoblastic lymphoma, 3rdrelapse

Large cell lymphoma 2ndrelapseHodgkin

disease, 2nd relapse

Non-Hodgkin lymphoma + BMTreatment

regimenTBI1200R + VP-16 + LBC+ CI

ara-C x 144 hVP-16 + LB CI ara-C" + 1200

rads TBI, BMTVP-16 + LB -1-CI ara-C" + 800

rads TBI, BMT 2VP-16 + LB + CI ara-C1 + 1200

rads TBI, BMTLB + CI ara-C*, cycloph. + 600

rads TBI, BMTVP-16 + LB + CI ara-CEfficacyNE,

no tumorat autopsy

CR, 2moCR,

2 mo

NE

CR, 5 mo

PRHost

toxicityDied

of Grade IV pulmonarytox.Grade

III myelosuppression +Grade I mucosal

Grade III myelosuppression +Grade II mucosal

Grade III myelosuppression tGrade I mucosal

Grade IIImyelosuppressionGrade

III myelosuppression.involvement, 2nd relapse

C. L. 7 mo Congen. ALL mixed lineage 2ndrelapse

A. V. 3 yr ALL, 2nd relapseS. G. 5 yr ALL, 2nd relapse

VP-16 + LB + CI ara-C NR

VP-16 + LB + CI ara-C" CRVp-16 + LB + CIara-C NR

Grade III CNS, Grade IIICholestasis

Grade IV myelosuppression

Grade HI myelosuppressionGrade IV myelosuppression

°The duration of ara-C continuous infusion was 60 h. The loading bolus and continuous infusion were begun 12 h after the test-dose of ara-C.* Second treatment.' LB, loading bolus; NE, not évaluable,CR, complete remission; CI, continuous infusion; NR, no response.

246

on April 9, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

PHARMACOLOGY OF ara-C IN PEDIATRIC PATIENTS

patients' tumor for optimum ara-C activation and for pharma-

cokinetic variations between patients. As a result, more uniformpharmacological parameters were obtained (Table 3).

A third difference is that our studies incorporated ara-C into

a successful combination chemotherapy and radiation therapyregimen. Because of this and the small numbers of patientstreated thus far, it is not possible to generalize about the efficacyof the mode of administration of ara-C. Previous studies of ara-C as a single agent have shown only short duration of completeremissions in pediatrie patients (6). Successful use of ara-C willbe based on combination either with other chemotherapeuticagents or other modalities of cancer treatment to not onlyachieve a maximum therapeutic index, but to attempt to obtainsuccessful long term results in the treatment of leukemias andlymphomas.

Overall, our preliminary data are comparable to other trialsconducted in heavily pretreated patients with relapsed leukemia.The toxicities we have observed have been primarily hemato-poietic and the significant, dose-limiting nonhematopoietic toxicities usually seen in patients receiving HDara-C in combination therapy, were observed less frequently in our patients(Table 4). This regimen of loading bolus plus continuous infusion of ara-C, which shows weak correlations with body surfacearea, may allow optimal ara-C doses to be administered inconjunction with other modalities, including other chemotherapeutic agents, radiation therapy and bone marrow transplantation.

REFERENCES

1. Skipper, H. E., Schabel, F. M., Jr., and Willcox. W. S. Experimentalevaluation of potential anticancer agents. XXI. Scheduling of arabinosyl-cytosine to take advantage of its S-phase specificity against leukemia cells.Cancer Chemother. Rep., 51:125-165, 1967.

2. Bodey, G. P., Rodriguez, V., McCredie, K. B., and Freireich, E. J. Earlyconsolidation therapy for adults with acute leukemia in remission. Med.Pediatr. Oncol., 2: 299-307, 1976.

3. Plunkett, W., I .ilk-mark, J. O., Adams, T. M., Nowak, B., Estey, E., Kumarjian, H., and Keating, M. J. Saturation of 1-/3-D-arabinofuranosylcytosine 5'-triphosphate accumulation in leukemia cells during high-dose l-.i-n-arabi-nofuranosylcytosine therapy. Cancer Res., 47: 3005-3011, 1987.

4. Plunkett, W., lacoboni, S., Estey, E., Danhouser, L., Liliemark, J. O., andKeating. M. J. Pharmacologically directed ara-C therapy for refractoryleukemia. Semin. Oncol., /2(Suppl. 3): 20-30, 1985.

5. Gahrton, G. Treatment of acute leukemia-advances in chemotherapy, immunotherapy and bone marrow transplantation. Adv. Cancer Res., 40: 255-329, 1983.

6. Avramis, V. I., Biener, R., Krailo, M.. Finklestein, J., Ettinger, L., Wil-loughby. M., Siegel S. E., and Holcenberg, J. S. Biochemical pharmacologyof high dose 1-/3-D-arabinofuranosylcytosine in childhood acute leukemia.Cancer Res., 47: 6786-6792, 1987.

7. Avramis, V. I., Mecum, R., Nyce, J., Steele, D. A., and Holcenberg, J. S.DNA methylation and pharmacodynamic studies of HDara-C before andafter 5-Azacytidine treatment in-children with refractory acute lymphocyticleukemia (ALL) to ara-C. Proc. Am. Assoc. Cancer Res., 29: 190, 1988.

8. Weinberg, K. L, Avramis, V. I., Lenarsky, C., Falk, P.. Kogut, N., Feig, S.,and Parkman, R. Variable high dose cytosine arabinoside (HDara-C) pharmacology in bone marrow transplant (BMT) patients. Proc. Am. Soc. Clin.Oncol., 7: 65, 1988.

9. Avramis, V. I., Alajajyan, M., Huang, S-H., and Holcenberg, J. S. Deoxy-cytidine kinase (dCk) purification and characterization from CEM/O, CEM/dCk(-) and LI210/0 cells. Prog. Am. Assoc. Cancer Res., 28:8, 1987.

10. Kreis, W., Graham, J., and Damin, L. A. Phosphorylation of selectedsubstrates by semipurified cytoplasmic kinases of three P815 murine leukemias exhibiting collateral sensitivity (P815/TG), sensitivity (P815) or resist

ance (P815/ara-C) to l-/3-D-arabinofuranosylcytosine. Biochem. Pharmacol.,31: 3831 -3837, 1982.

11. Muus, P., Drenthe-Schonk, A.. Haanen. C., Wessels, H., and Linssen, P. In-Vitro studies on phosphorylation and dephosphorylation of cytosine arabinoside in human leukemic cells. Leukemia Res., / /.- 319-325, 1987.

12. Avramis, V. I., Antonsson, B. E., and Holcenberg, J. S. Biochemical pharmacology of HDara-C before and after 5-Aza-C treatment in pediatrieleukemic (ALL) patients with clinical resistance to HDara-C. Proc. Am.Assoc. Oncol., 6: 34, 1987.

13. Capizzi, R. L., Poole, B. L.. Cooper, M. R., Richards, F., II, Smuri. J. J.,Jackson, D. V., Jr., White, D. R., Spurr, C. L. Hopkins, J. O., Muss, H. B.,Rudnick, S. A., Wells, R., Gabriel, D., and Ross, D. Treatment of poor riskacute leukemia with sequential high dose ara-C and asparaginase. Blood, 63:694-700, 1984.

14. Herzig, R. H., Lazarus, H. M., Wolff, S. N., Phillips,^. L., and Herzig, G.P. High-dose cytosine arabinoside therapy with and without anthracycolineantibiotics for remission reinduction of acute nonlymphoblastic leukemia. J.Clin. Oncol., 3: 992-997, 1985.

15. Hiñes,J. D., Oken, M. M., Mazza, J. J., Keller. A. M., Streeter, R. R., andGlick, J. H. High-dose cytosine arabinoside and m AMSA is effective therapyin relapsed acute nonlymphocytic leukemia. J. Clin. Oncol., 2: 545-549,1984.

16. Plunkett, W., Nowak, B., and Keating, M. J. Effect of amsacrine on ara-CTP cellular pharmacology in human leukemia cells during high dose cytar-abine therapy. Cancer Treat. Rep., 71:479-483, 1987.

17. White, J. C., Hiñes,L. H., and Rathmell, J. P. Inhibition of 1-0-D-arabino-furanosylcytosine transport and net accumulation by teniposide and etoposidein Ehrlich ascites cells and human leukemic blasts. Cancer Res., 45: 3070-3075, 1985.

18. Rivera, G., Aur, R. J., Dahl, G. V., Pratt, C. B., Wood, A., and Avery, T. L.Combined VM-26 and cytosine arabinoside in treatment of refractory childhood lymphocytic leukemia. Cancer (Phila.), 45: 1284-1288, 1980.

19. Mulder, J. H., and Harrap, K. R. Cytosine arabinoside uptake by tumor cellsin vitro. Em. J. Cancer, //: 373-379, 1975.

20. Plagemann, P. G. W., Martz, R., and Wohlhueter, R. M. Transport andmetabolism of deoxycytidine, and 1-d-D-arabinofuranosy Icytosine into cultured Novicoff rat hapatoma cells, relationship to phosphorylation, andregulation of triphosphate synthesis. Cancer Res., 38: 978-989, 1978.

21. Chabner, B. A. Cytosine arabinoside. In: B. Chabner (éd.),PharmacologiePrinciples of Cancer Treatment, pp. 387-401. Philadelphia: W.B. SaundersCo., 1982.

22. Plunkett, W., Liliemark, J. O., Estey, E., and Keating, M. J. Saturation ofara-CTP accumulation during high-dose ara-C therapy: pharmacologie rationale for intermediate-dose ara-C. Semin. Oncol., 14(Supp\. 1): 159-166,1987.

23. Chou, T-C, Arlin, Z., Clarkson, D. B., and Phillips, F. S. Metabolism of 1-0-D-arabinofuranosylcytosine in human leukemic cells. Cancer Res., 37:3561-3570,1977.

24. Avramis, V. L, and Powell, W. C. Pharmacology of combination chemotherapy of cytosine arabinoside (ara-C) and uracil arabinoside (ara-U) or tetra-hydrouridine (THU) against murine leukemia L1210/0 in tumor bearingmice. Cancer Invest., 5: 293-299, 1987.

25. Capizzi, R. L., Yang, .1-1...Cheng, E., Bjornsson, T., Sahasrabudhe, D., Tan,R. S., and Cheng, Y-C. Alteration of the pharmacokinetics of high-dose urnC by its metabolite, high ara-U in patients with acute leukemia. J. Clin.Oncol.,/: 763-771, 1983.

26. Liliemark, J. O., Plunkett, W., and Dixon, D. O. Relationship of 1-/3-D-arabinofuranosylcytosine in plasma to I-/3-D-arabinofuranosylcytosine 5'-triphosphate levels in leukemic cells during treatment with high-dose 1-ß-D-arabinofuranosylcytosine. Cancer Res., 45: 5952-5957, 1985.

27. Sapur, J. C., and I ridiami. A. Identification of purine deoxyribonucleosidekinases from human leukemia cells: Substrate activation by purine andpyrimidine deoxyribonucleosides. Biochemistry, 26: 590-597, 1987.

28. Ochs, J., Sinkule, J. A., Danks, M. K., Look, T., Bowman, W. P., and Rivera,G. Continuous infusion high-dose cytosine arabinoside in refractory childhood leukemia. J. Clin. Oncol., 2: 1092-1097, 1984.

29. Donehower, R. C., Karp, J. E., and Burke, P. J. Pharmacology and toxicityof high-dose cytarabine by 72-hour continuous infusion. Cancer Treat. Rep.,70:1059-1065, 1986.

30. Bradford, M. M. A rapid and sensitive method for the quantitation ofmicrogram quantities of protein utilizing the principle of protein-dye binding.Anal. Biochem., 72: 248-254, 1976.

31. Hande, K. R., and Chabner, B. A. Pyrimidine nucleoside monophosphatekinase from human leukemic blast cells. Cancer Res., 38: 579-585, 1978.

32. Cheng, Y-C., Agarwal, P., and Parks, R. E. Erythrocytic nucleoside diphos-phokinase IV. Evidence for electrophoretic heterogeneity. Biochemistry, 10:2139-2142,1971.

247

on April 9, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1989;49:241-247. Cancer Res   Vassilios I. Avramis, Kenneth I. Weinberg, Judith K. Sato, et al.   Continuous Infusion of the DrugBiochemically Optimal Regimen of Loading Bolus plusPediatric Patients with Leukemia and Lymphoma after a

-D-Arabinofuranosylcytosine inβPharmacology Studies of 1-

  Updated version

  http://cancerres.aacrjournals.org/content/49/1/241

Access the most recent version of this article at:

   

   

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/49/1/241To request permission to re-use all or part of this article, use this link

on April 9, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from