(CANCERRESEARCH48, 1337-1342,March 1, 19881

The results indicate that when 100 mg/rn2 of ara-C is coinfused with THU at 250 or 350 mg/rn2 over 3 h, the plasma araC levels achieved are comparable to levels obtained when 1 g/rn2 of ama-C was infused over the same time. A preliminaryreport of this work has been presented elsewhere (7).

MATERIALS AND METHODS

Patient Selection.Heavily pretreated patients with solid tumors wereselected for this study, 16 males and 14 females, with mean age of 58.3yr (range, 36 to 75) were enlisted in the protocol. Written informedconsent was obtained from all patients. Admission criteria includedhemoglobin of > 10.0 and creatinine < 1.5. With few exceptions, thepatients were outpatients.

Study Design. The study was performed in two parts. The first partestablished the effect of various doses of THU (10, 40, 60, 80, 100,150, 200, 250, 300, and 350 mg/rn2) on a fixed dose of ara-C (25 rng/m2).Once the maximum effective dose of THU was determined, thesecond part was designed to evaluate the effect of two high doses ofTHU (250 and 350 mg/rn2)on various doses of ama-C(50, 75, and 100mg/rn2).In all cases, the patients receivedam-C first (Day 1;“control―)and later (Day 3 or within 1 wk) ma-C with THU (“treated―).Underthese conditions all patients served as their own controls and the effectof THU was evaluated by the differencesobserved between parametersdetermined following control and treatment. The total of 30 patientsparticipating in this study received49 courses (ma-C and ara-C + THUpairs). At least 3 patients were scheduled for each treatment regimen.In some instances, the same patients participated in more than onetreatment regimen (same ma-C but with different doses of THU ordifferent doses ofara-C and THU). In such cases, each treatment coursewas given after a rest period of 2 wk. The purity of THU, Lot 0584,was assessed to be 88.2% pure and was prepared fresh for each treatment.

Drug Administration. ama-Calone was prepared fresh in 250 ml D5Wshortly before administration and infused over 3 h with a Travenolpump. For the combined drug studies, 20% of the dose of THU,dissolvedin sterilewater,wasinjectedin the form of an i.v. bolus 10mm before the start of the infusion. The rest ofthe THU was coinfusedwith nra-C in a total volume of 250 ml.

Sample Collection. Two ml of blood were collected in heparinizedtubes containing 0.5 mg THU at time 0 (before the start of infusion) at10 and 30 mm and 1, 2, 2 h and 45 mm, 3 h and 15 mm, 3 h and 30mm, 4, 5, 6, and 24 h after initiation of the infusion. The tubes wererapidly cooled in ice and plasma was removed after centrifugation at3000 x g for 5 mm. Urine of all patients was collected over 24 h. NoTHU was added to the urine samples. Plasma and lO-ml aliquots ofurine sampleswere stored at —20'Cpending analysis.

Analysis of am-C. The method of Piall et a!. (8) was used for ama-Canalysis. Antibody for radioimrnunoassay was purchased from theUniversity of Surrey, Guildford, England. Cross-reaction of the assayfor nra-U and THU was found to be minimal (< 0.008 and 0.03%,respectively, at 50% inhibition). Separate standard curves were established for the evaluation ofama-C levels in plasma and urine. In addition,for plasma samples containing low levelsof ama-C,the ama-Cstandardswere prepared by spiking appropriate amounts of ama-Cto 2 ml ofplasma (freeze-dried human plasma Sigma Chemical Co., St. Louis,MO) containing 0.5 mg of THU. This step is essential to avoidinaccuracies due to interference of THU with ama-C in the radio

ABSTRACF

When l-@1-D-arabinofuranosylcytosine(ara-C), 25 mg/rn2,is infusedover 3 h together with tetrahydrouridine (THU) at 10 to 350 mg/rn2toheavily pretreated patients with solid tumors, Michaeis-Menten typekinetic values are observed with leveling off of i@area under the curve, i@ara-C levels at 3 h, and@ total body clearance after 150 mgJm2 of THU.When the ara-C dose was increased to 50, 75, and 100 mgJm2coinfusionof 250 or 350 mg/rn2 of THU significantly increased plasma ara-C atpeak and area under the curve. In contrast, total body clearance andvolume of distribution decreased significantly. At 100 mg/rn2 of ara-Ccoinfusedwith high doses of THU, i.e., at 350 mgJm2,the pharmacokinetics of plasma ara-C was changed from a biphasic decay of Plasma maC at peaks (control) to a curve similar or identical to a monophasic curve,indicating that THU not only inhibits deamination but also changes thedistribution of ma-C. This combination provides plasma am-C levels(1O giM)comparable to high dose ara-C at 1 g/m2. Such plasma am-Clevels are considered to be sufficient for saturation of the kinases catalyzingtheproductionof 1-fi-o-amabinofumanosylcytosine5'-tmiphosphate.This reduced ama-C dose necessary to achieve saturation of kinases alsoreduces plasma 1-@8-r-ambinofumanosylumacillevels substantially. Toxicity of this combination was predominantly confined to bone marrow andgastrointestinal toxicity.

INTRODUCHON

ama-C3is an agent commonly used in the treatment of acutemyelo- and myelo-monocytic leukemias. A limit to cytotoxicblood levels of ara-C is its rapid deamination to uracil arabinoside. Measures to circumvent this enzymatically catalyzedreaction include the use of high dosage ara-C (1), continuousinfusion of ara-C (2), and the combination of ama-C with deaminase inhibitors. The most prominent deaminase inhibitor isTHU (3). In rhesus monkeys, which exhibit a high plasma levelof CR/deoxycytidine deaminase (E.C.3.5.4.5), this enzyme wascompletely inhibited for up to 2 h after 12.5 to 100 mg/kg ofTHU (4). Although to date little evidence has accumulated inhumans for an increased chemotherapeutic activity or substantial changes of the pharmacokinetic activity or substantialchanges of the pharmacokinetic parameters of ama-C due to thebolus administration of ara-C plus THU over ara-C alone (5,6), the area has received inadequate evaluation.

The present study was undertaken to evaluate the pharmacokinetic parameters of ara-C in combination with THU bydetermining plasma ara-C levels achieved under conditionscomparable to those reported for high-dose ara-C regimens.

Received7/23/87; revised11/30/87; accepted12/3/87.The costsof publicationof this articlewere defrayedin part by the payment

of pagecharges.This article must thereforebe herebymarkedadvertisementinaccordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

I Supportedin part by the Don Monti MemorialResearchFoundation,andNational Cancer Institute Grant 1 ROl CA38980.

2To whomrequestsfor reprintsshouldbeaddressed.3 The abbreviations used are: ara-C, 1-@9-r-arabinofuranosylcytosine; ara-U, I-

fi-o-arabinofuranosyluracil; THU. tetrahydrouridine; AUC, area under the curve;V.., volumeofdistribution;TBC, total bodyclearance;CR, cytidine;HDA, highdoseama-C;MRT, meanresidencetime;C3h, plasmaara-C at peak;CNS, centralnervous system;CSF, cerebrospinal fluid.

1337

Effect of Tetrahydrouridine on the Clinical Pharmacology of 1-j9-D-Arabinofuranosylcytosine When Both Drugs Are Coinfused over Three Hours'

Willi Kreis,2 Keith Chan, Daniel R. Budman, Phillip Schulman, Steven Allen, Lora Weiselberg, Stuart Lichtman,Vicki Henderson, Jean Freeman, Margaret Deere, Michael Andreeff, and Vincent Vinciguerra

Division ofHematology/Oncology, Department ofMedicine, North Shore University Hospital and Cornell University Medical CollegefW. K., D. R. B., P. 5., 5. A., L.w., V.H.,J. F.,M. D., V. V.],Manhasset,NewYorlçCiba-GeigyCorporation[K.C.],Ardsley,NewYork,10502;andMemorialSloan-KetteringCancerCenter[M.A.J, New York, New York 11021

Research. on August 22, 2021. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

Table I Pharmacokineticparametersfor3-h infusionofara-C aloneorincombinationwithTHUara-C

ara-C alone am-C + ara-C+dose(control) 250 mg/m2THU 350 mg/m2THUC3h

(ng/ml)2577.1 ±17.1 (31) 353.9±101.2(5) 347.9±113.8(4)50

131.1 ±68.3 (6) 977.5 ±326.7 (3) 633.6 ±212.4(3)75331.5 ±161.7 (6) 1,173.8±593.2 (3) 1,900.8 ±182.1(3)100411.3±127.3(6) 2,702.1±802.0 (3) 2,283.3±214.4(3)AUC

(ng )(h/mI)25286.3 ±75.7 (31) 1,926.0±1,184.9 (5) 1,814.9 ±546.8(4)50486.3 ±149.3 (6) 4,600.7 ±966.7 (3) 3,395.2 ±1,200.4(3)75

1,012.6±394.4 (6) 7,745.5±4,250.2 (3) 15,497.4±5,312.3(3)1001,271.8±267.5 (6) 19,224.0±2,321.4(3) 17,610.6±1,601.0(3)TBC

(liters/h/m2)2593.16 ±23.60 (31) 16.87 ±8.38 (5) 14.77 ±4.45(4)50

110.68±31.10(6) 11.17±2.18(3)16.27±6.53(3)7582.98 ±29.25 (6) 11.44 ±4.77 (3) 5.22 ±1.52(3)10081.16 ±14.62(6) 6.05 ±1.61(3) 5.71 ±0.49(3)Yes

(liters/m2)25

251.0 ±163.3 (31) 82.2 ±39.9 (5) 90.9 ±24.7(4)50261.3 ±148.7 (6) 63.6 ±12.3 (3) 72.1 ±37.4(3)75183.8±197.3(6) 78.3 ±30.2 (3) 43.0 ±5.4(3)100109.2 ±35.1 (6) 39.6 ±6.9 (3) 44.6 ±3.7(3)MR'r

(h)252.82 ±1.75 (31) 5.68 ±2.51 (5) 6.21 ±0.47(4)502.30 ±1.14 (6) 5.73 ±0.78 (3) 5.20 ±3.27(3)752.25 ±1.33 (6) 6.98 ±0.69 (3) 8.62 ±1.73(3)1001.33 ±0.28 (6) 7.51 ±0.48 (3) 7.84 ±0.79(3)%

excreted inurine252.22 ±1.42 (31) 13.93 ±6.55 (4) 9.90 ±3.66(3)501.80 ±1.00 (4) 14.85 ±15.06 (2)9.40(1)753.70±1.2(3) 23.60(1) 30.07±14.18(3)1002.30 ±1.4(6) 27.45 ±6.01 (2) 29.74 ±5.64(3)Renal

clearance(liters/h/m2)252.01 ±1.25 (31) 1.76 ±1.05 (4) 1.21 ±0.24(3)501.93±1.07(4) 1.75±1.70(2)2.22(1)752.28 ±0.43 (3) 1.40(1) 1.43±0.22(3)1001.98±1.31(6) 1.63±0.25 (2) 1.68±0.16(3)a

Valuesin parentheses,no. of valuesusedforcalculation.

PHARMACOIUNETICS OF ara-C + THU IN HUMANS

immunoassay. The coefficient of variation for these plasma sampleswas 3.0% for samples containing 1 ng/ml. For samples containing morethan 25 ng/ml of ama-C,the aqueous ama-Cstandards containing ama-Conly were used.

Pharmacokinetic Data Analysis. Pharmacokinetic parameters forboth the control and treatment period were calculated as reported byGibaldi and Perrier (9) and outlined elsewhere (10). In all cases, theresidual area was small and accounted for less than 2% of the totalarea.Thepeakplasmalevelwasobtainedfromtheobservedplasmaconcentration at the end of the infusion (C3h). For technical reasons,the plasma sample collected at 2 h 45 mm was considered to be the 3-h sample. The MRT and V, for ama-Cwere calculated according to themethodof Chanand Gibaldi (11) as reportedfor i.v. infusion.Afterthe terminal (8) half-life was determined, the a-half-life was calculatedusing the method of residuals (9).

THU Effect. The effect of THU on ama-Ckinetics was expressed, forexample, as increase of AUC, increase of am-C concentration at theend of infusion (C3h), or decreasein am-C TBC (expressedas %decreased) during the first phase of the study. Since THU is a potentcytidine deaminase inhibitor, it is reasonable to assume that the effectof THU on ama-C follows a Michaelis-Menten-type function as described by Equation 1 (12)

Emax X THU doseEffect =

E50 + THU dose

where Emax is the maximum effect and E50 is the THU dose for 50%of the maximum effect. Emax and E50 were estimated by fitting theindividualeffect versusTHU dose curve, using a nonlinear least-squaresregression program PC NONLIN based on a Gauss-Newton algorithm(13). The program was run on an IBM PC/AT computer.

RESULTS

Pharmacokinetics of ama-C Plus THU Combinations. Theeffect of THU on ara-C dearnination was evaluated by measuring the differences observed between parameter values (@-AUC, @-C3h,and @-TBCin percentage) of controls and treatment. Fig. 1 depicts such effects for the @-AUCvalues versusTHU doses. The theoretically fitted curve, as described byMichaelis-Menten-type equations, was also included in thesefigures. The theoretically fitted curve reflexes the expectedsaturation pattern despite the small correlation coefficient (ap

proximately 0.5). Since each subject contributed one data pointonly for each treatment, the deviations probably reflect theintersubject variation. Some patients received two or threetreatments with various doses of THU; the results showed nodeviation indicating that the fitted theoretical curves representthe general trend. Due to the large intersubject variability, thecurve may not be predictive from one subject to another.

Since from the above studies it was concluded that approxi

mately 250 to 350 mg/rn2 of THU are required to obtainoptimal inhibition of am-C deamination, the effects of THUon increasing ara-C doses up to 100 mg/rn2 were evaluated. Ingeneral, using 250 or 350 mg/rn2 did not differ significantly inany of the parameters studied (Table 1). Indeed, these valuesfurther support the observation delineated in Fig. 1, namely,that maximal effect was already reached using 250 mg/rn2 THUdose without further benefits with doses of THU of 350 rng/rn2. From Table 1 it is clear that under the influence of thesemaximal THU doses, ara-C, C3h, AUC, and MRT were drastically increased and TBC and V, significantly decreased. Thepercentage of unchanged ara-C in urine was also significantlyincreased (from approximately 2% increased to more than 10%and as high as 30%) but the renal clearance values remainedunchanged, indicating that THU has no effect on the renalhandling of ara-C.

The semilog plots of ara-C following control (ara-C alone)and treatment (coadministration of ara-C and THU) in tworepresentative patients are presented in Fig. 2. The bottomdepicts a patient who received 25 mg/rn2 ara-C as control andthe same dose coadrninistered with 350 mg/rn2 of THU. Thetop depicts a patient who received 100 mg/rn2 ara-C and thesame dose coadministered with 350 mg/rn2 of THU. It isevident that THU significantly increased the plasma levels ofama-C in both cases. When ara-C was administered alone, theama-Clevels increased rapidly and reached plateau levels withinabout 1 h. Thereafter, a rapid fall of the plasma levels with aninitial t4 of 0. 19 and 0.20 h, respectively, and a long terminal

half-life of 11.23 and 21 .54 h, respectively, were observed.

03200 -

2 2800 -

j24oo@ 0@ 0

,@20O0- 0C

U 0DI4 600-

@ 200 -

0 @cow

4002 ? I

Co‘ 40‘ 80•20@ 160200 240 280 320360THU DOSE (mg/rn2)

Fig. 1. Plot of a-Delta AUC versusTHU Dose.Abcissa,THU dose(mg/m2);ordinate,ia-DeltaAUC (ng/ml x h).

1338

Research. on August 22, 2021. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PHARMACOKINETICSOF ara-C+ THU IN HUMANS

was observed in the control runs. The behavior of the ara-Cplasma level profile was drastically changed when THU wascoadministered with ama-C. There is no indication of a plateaulevel even at the end of infusion. ara-C levels increased rapidlyover the first hour. Thereafter, the plasma levels still increased,but at a significantly slower rate until the end of the infusion.Thereafter, the a-half-lives were double the ones ofthe controls(0.54 and 0.40 h for bottom and top, respectively) but the (3-half-life decreased drastically (3.97 and 3.47 h for the bottomand top, respectively) (Fig. 2). Again, these half-lives pertain tothe two cases represented in Fig. 2. The mean of all of theterminal half-lives of patients receiving 25 or 100 mg/m2 ofara-C in combination with 350 mg/rn2 of THU were 5.27 ±1.36 h (range, 3.97—6.50h) and 3.69 ±0.35 h (range, 3.43—4. 15 h), respectively. These drastic changes of the decay patternfollowing the discontinuation of the infusion were observed forall patients participating in this study. In some instances, especially using high doses ofara-C (100 mg/rn2) and THU (250—350 mg/rn2), the two phases took on the shape of monoexponential decay curves. A comparison of the effect of 350 mg/rn2ofTHU on plasma levels ofara-C of patients who received araC at 25, 50, 75, and 100 mg/rn2 is demonstrated for representative patients in Fig. 3. It is obvious that, at lower levels of araC (25 and 50 mg/rn2), the plasma ara-C pattern after discontinuation of the infusion is clearly biphasic. When ara-C is givenat 75 mg/rn2 and especially at 100 mg/rn2, the plasma decaycurve becomes more monophasic. While for technical reasonsmentioned above, no blood was collected between the time of 6and 24 h for patients receiving 25, 50, or 75 mg/rn2 of ara-C,the top curve, representing values of a patient who received araC at 100 mg/rn2 and THU at 350 mg/rn2, also includes plasmaara-C values at times 10 and 12 h. It is obvious that, at leastfor this dosage of ama-C, further points of blood sampling didnot produce significant changes in the plasma ara-C pattern.

Pharmacokinetics ofara-C Alone (Controls). From the pararneters evaluated for the control dose administration (25, 50, 75,and 100 mg/rn2 ara-C), it was possible to determine doseproportionality of ara-C. In general (Table 1), mean AUC andC3h values clearly show ara-C dose-related increases. MeanTBC, percentage of dose excreted in urine, and renal clearancevalues remained relatively constant as ara-C dose increased,further supporting the dose proportionality. However, trendsfor decrease were observed for V, and MRT with increasingamounts of ama-C dose (Table 1). Inclusion of further datapoints (at 10 and 12 h) in 2 patients receiving 100 mg/rn2 ofara-C alone provided no evidence ofsignificant changes in thesepharrnacokinetic parameters.

Some patients participated in more than one treatment regimen which provided excellent opportunity to evaluate intrasubject and intersubject variability. Intrasubject variability of amaC was evaluated in 9 patients after repeated administration ofama-Calone at 25 mg/rn2 ama-C. In Fig. 4, AUC is the parameterdemonstrated to illustrate the intrasubject variability. The results indicated that the majority of patients (6 of 9) showedsmall intrasubject variability. However, in few cases, the intrasubject variability was high approaching intersubject variability(not shown). The same observations were obtained for the otherparameters studied, namely, C3h and TBC. The overall variability (expressed as coefficient of variation in percentage) wassimilar for the 9 subjects as compared to all subjects completingthe study. Thus, the 9 subjects with repeated administrationwere representative for all subjects evaluated.

Toxicity. Gastrointestinal toxicity, such as nausea and vorniting, were observed in 75% of courses administered. However,

1339

-3000@\ -2000-\ $000

@\ ,500'@@ \

\ NNNk @‘\,,Aro-C.THU@ 50\ 0

@—-.- 5@ -@ 4,0-C

@NN@N@ I

@ __ __ @@Aro-C.THU@Ara-C

@ 1@ I@ I@@@ I

5 10 IS 20 25TIME (hI

Fig. 2. Plasma ama-Clevels in 2 patients receivingama-C,25 mg/m2, plusThU, 350 mg/m2(bottom),or am-C, 100 mg/m2,plusTHU 350, mg/m2(top).t@,control, 25 mg/m2 of ama-C;A, treated, 25 mg/m2 of ama-C+ 350 mg/m2THU; 0, control, 100 mg/m2 of am-C; •,treated, 100 mg/m2 of ama-C+ 350mg/m2 THU. Dotted lines, probable but not experimentally proven developmentofplasmalevelsbetween6 and24h.

I

HOURSFig. 3. Plasma ama-Cprofiles in 4 patients receiving ara-C at 25, 50, 75, and

100 mg/m2 in combination with THU at 350 mg/rn2. Dotted lines, probable butnot experimentallyprovendevelopmentof plasmalevelsbetween6 and 24 h.

_ 00

@50

These half-lives pertain to the two cases represented in Fig. 2.The mean of all of the terminal half-lives of patients receiving25 or 100 mg/rn2 of ara-C alone (controls) were 13.3 ±8.7 h(range, 3.1 to 38.8 h) and 8.5 ±3.5 h (range, 5.4 to 11.2 h),respectively. In three courses of treatment with ara-C at 100mg/rn2 and THU at 350 mg/rn2, blood was collected at theroutine schedule and also at 10 and 12 h after initiation of theinfusion (Fig. 3), and the resulting terminal half-lives and AUCsshowed similar values with or without these two new timepoints. Similar patterns were observed for all patients receivingara-C alone at any of the four doses of ara-C administered. Inmany instances, as indicated in the top, (Fig. 2) a steady state

Research. on August 22, 2021. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

I@ °I8IOC

2= , , , , . , , ,0 I 1 1 4 • S 7 S S

Subject Numbar

Fig. 4. Intrasubject variation of patients treated 2 or 3 times with ama-C, 25mg/m2 (controls), as expressedin AUC (ng/ml x h). Abcissa, subject number,ordinate,AUC (ng/ml x h).

toxicity was unrelated to the doses of ama-C and THU. Thesegeneral gastrointestinal toxicities could be prevented with Cornpazine, 10 rng i.m., before and after infusion. In some instances,Reglan, 50 rng, infused shortly before initiation ofthe treatmentwas also used successfully. Myelosuppression was rare andinconsistent below 75 mg/rn2 of ama-C in the presence of THU.Above that level of ama-C, thrombocytopenia (values below150,000) became evident randomly in 8 of 12 patients. Themean time to nadir was 8.5 ±1.4 days and the mean time torecovery was 9 ±6.2 days. Leukopenia (white blood count< 3.5) was also seen randomly in 3 of these patients.

Therapeutic Effect. A therapeutic benefit of the combinationof ama-C + THU was not observed.

DISCUSSION

High-dosage ara-C administered in multiple doses has definite beneficial effect in some patients with leukemia who arerefractory to conventional-dose ama-C. This treatment has alsoproduced significant toxicity, including neurotoxicity in 16—50% ofthe patients studied (14—16).Methods to reduce toxicityare clearly needed.

Plasma levels achieved by HDA were reported up to 222 @M(equivalent to 54 @g/rnl)with large intersubject variation (17).However, there is good evidence indicating that such high amaC plasma levels may not be necessary for effective treatment ofleukemia. In vitro studies by Chou et a!. (18), Harris andGraharne-Srnith (19), and Riva and Rusturn (20) demonstratedthat saturation of the kinases occurred at ama-C levels of about10 @tM,which is equivalent to 2.5 @sg/rnl,and higher levels ofara-C might, in some instances, even prevent the further phosphorylation of ara-CMP to its di- and triphosphates (21).

Studies on saturation of ama-C phosphorylation in vivo werereported by Plunkett et a!. (22). They concluded that infusionrates of 200 to 250 mg/rn2/h, 10 times lower than thosecommonly in use with HDA schedules, produce plasma levelsof ama-C adequate to achieve maximum intracellular 1-(3-r-arabinofuranosylcytosine 5 ‘-triphosphate accumulation. Theara-C plasma levels measured under the influence of THU inour study are comparable with the ones reported by Plunkett eta!. (22) and Slevin et a!. (23). Slevin et a!. reported ama-Cplasmaplateau levels of 2960 ±882 ng/ml during a 3-h infusion of 1g/rn2 of ama-C. The same peak ama-C levels can be achieved in

our study (2702. 1 ±802.0 ng/ml) using 100 mg/rn2 plus 250mg/rn2 THU. These levels are within the range recommendedby Plunkett et a!. (22) and van Prooijen et aL (24).

The use of a 10 times lower ama-C dose to achieve plasmalevels comparable to HDA implies at least 10 times lowerplasma ama-U levels. Spriggs et a!. (25) and Capizzi et a!. (26)noted that ama-U levels exceeded ama-C levels extensively. Lowlevels of ama-C excretion in urine reported by Ochs et a!. (17)after HDA infusion (10.5—31% of unchanged ama-C), by Hoand Frei (27) for conventional ama-C infusion (7.8%), and forlow-dose ama-C infusion, by Kreis et a!. (10), also imply substantial dearnination of ama-C. This study, demonstrating verylow recovery of ama-C in urine, corroborates these reports.

In preliminary analyses of ama-U performed on two patientsreceiving 100 mg/rn2 of ama-C and 250 mg/rn2 of THU, theplasma ratios of AUC ama-U to AUC ama-C showed strikingdecreases of these ratios from the controls to the treatmentcourses (26.0 to 1.2 in one patient and 8.8 to 0.42 in the otherpatient). Likewise, the ratios ofurinary ama-Uto ama-Cexcretionin percentage of administered ama-Cchanged from 118 to 2 and53 to 1.5 for the controls versus treatment courses.

The deamination product of ama-C, ama-U, has been considered to be nontoxic by many investigators. However, Mullerand Zahn (28) reported significant inhibitory activity of thegrowth of L5178Y cells in vitro by ama-U at levels of 0.5 to 10@sg/ml.Yang et a!. (29) reported enhancement ofara-C cytotoxicity in vitro by ama-U. High plasma concentrations of ama-U, asobserved in the treatment with high-dose ama-C (25, 30), mightwell contribute to the toxicity spectrum of ama-C.

ama-U has also been reported to inhibit the uptake of ama-Cin a subline of Ll210 deoxycytidine kinase and CR deaminase(31). High levels of ama-U present during the clinical use ofHDA may diminish the uptake of ama-C into the target cells(31). High CNS levels of ama-U may also contribute to CNStoxicity. Lopez and Agarwal suggested unexpectedly high, prolonged CNS levels of ama-U to be associated with neurotoxicity(32). Following intraventricular injection of ama-U in a primatemodel, high CSF concentrations of ama-U were found to beassociated with diffuse slowing of electroencephalogram wavepatterns, while intraventriculam injection of ama-C did not alterthe electroencephalogram pattern (32). In humans treated withhigh-dose ama-C, substantial ama-U and ama-C CSF levels werereported by Capizzi et a!. (26). Similar findings for ama-C CSFlevels were reported by Slevin et a!. (23) and Breithaupt et a!.(30). As i.v. THU prolongs CSF ama-C levels in monkeys (33),an enhancement of CNS antileukernic effect might occur fromour present drug combinations.

The lack of significant effect of bolus injection of ama-C plusTHU as reported earlier (5, 6, 34) may be due to the short halflife of THU (6). Unlike ama-U (26), THU is bound tightly tocytidine deaminase with a K, in the range of l0@ M (35). Theonset of inhibition by THU of CR-deaminase extracted fromhuman liver is slow, reversible, and possibly time dependent,and preincubation with THU was recommended (36). In ourstudy, 20% of the dose was injected rapidly, 10 mm before thecoinfusion of the rest of the THU with ama-C.

This study demonstrated the maximum effective dose forTHU to be 250—350 mg/rn2. However, even at this dose range,total inhibition of deaminase was not observed. A markedchange in plasma level ama-Cdisposition as modulated by THUis expressed when 100 mg/rn2 of ama-C are coinfused with 350mg/rn2 of THU (Fig. 3). If the effect of THU on ama-C weresolely the inhibition of dearnination, then the following pharmacokinetic changes were to be expected: higher ama-C plasma

‘-4a00

U

PHARMACOKINETICS OF are-C + THU IN HUMANS

1340

Research. on August 22, 2021. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PHARMACOKINETICS OF ara-C + ThU IN HUMANS

5. Wong, P. P., Currie, V. E., Mackey, R. W., Krakoff, I. H., Tau, C. T. C.,Burchenal, J. H., and Young, C. W. PhaseI evaluation of tetrahydrouridinecombined with cytosine arabinoside. Cancer Treat. Rep., 63: 1245—1249,1979.

6. Kreis,W., Woodcock,T. M., Gordon,C. S., andKrakoff, I. H. Tetrahydrouridine: physiologicdispositionand effectupon deaminationof cytosineamabinoside. Cancer Treat. Rep., 61: 1347—1353,1977.

7. Kreis, W., Chan, K., Budman, D. R., Schulman, P., Allen, S., Weiselberg,L, Henderson, V., Freeman, J., Deere, M., and Vinciguerra,V. Effect oftetrahydrouridine (THU) on ama-Cpharmacokinetics and toxicity when bothdrugs are given by continuous infusion over 3 hrs. Proc. Am. Assoc. CancerRes.,27:165,1986.

8. Piall, E. M., Aherne, G. W., and Marks, V. M. A radioimmunoassay forcytosine arabinoside. Br. J. Cancer, 40: 548—556,1979.

9. Gibaldi, M., and Pemer, D. Pharmacokinetics,Ed. 2. New York: MarcelDekker, Inc., 1982.

10. Kreis, W., Chaudhri, F., Chan, K., Allen, S., Budman, D. R., Schulman, P.,Weiselberg, L, Freeman, J., Deere, M., and Vinciguerra, V. Pharmacokinetics of low dose l-(i-i>-arabinofuranosylcytosine given by continuous intravenous infusion over 21 days. Cancer Res., 45: 6498—6501,1985.

11. Chan, K. K., and Gibaldi, M. Estimationof statisticalmomentsand steadystate volume of distribution for a drug given by intravenous infusion. J.Pharm. Pharmaceut., 10: 551—558,1982.

12. Holford, N. H. S., and Sheiner, L. B. Understanding the dose.effectrelationship: clinical applications of pharmacokinetic-pharmacodynamic models.Clin. Pharmacokinet., 6: 429—453,1981.

13. Metzler, C. M., and Weiner, D. L PC NONLIN and NONLIN 84: softwarefor the statistical analysis of nonlinear models. Am. Statistician, 40: 52,1986.

14. Early, A., Preisler, H., Slocum, H., and Rustum, Y. A pilot study of highdoseama-Cfor acute leukemiaand refractorylymphoma.Cancer Res., 42:1587—1594,1982.

15. Lazarus, H., Herzig, G., Herzig, R. et al. Central nervous system toxicity ofhigh doseara-C. Cancer (Phila.), 48: 2577—2582,1981.

16. Nand, S., Messmore, H. L., Jr., Patel, R., Fisher, S., and Fisher, R. I.Neurotoxicity associated with systemic high dose cytosine arabinoside. J.Clin. Oncol., 4: 571—575,1986.

17. Ochs, J., Sinkule, J. A., Danks, M. K., Look, A. T., Bowman, W. P., andRivera, G. Continuous infusion high dose ama-Cin refractorychildhoodleukemia. J. Clin. Oncol., 2: 1092—1097,1984.

18. Chou, T. C., Arlin, Z., Clarkson,B. D., and Philips, F. S. Metabolismof 1-j9-n-arabinofuranosylcytosine in human leukemic cells. Cancer Res., 37:3561—3570,1977.

19. Harris, A. L., and Grahame-Smith, D. G. Variation in sensitivity of DNAsynthesis to ama-Cin acute myeloid leukemia. Br. J. Haematol., 45: 371—379, 1980.

20. Riva, C. M., and Rustum, Y. M. l-@9-n-arabinofuranosylcytosinemetabolismand incorporationinto DNA as determinantsof in vivomurine tumor cellresponse.CancerRes.,45: 6244—6249,1985.

21. Rustum, Y. M., Slocum, H., Wang, G., Bakshi, D., Kelly, E., Buscaglia, D.,Wrzosek, C., Early, A. P., and Preisler, H. Relationship betweenplasma amaC and intracellularara-CTP poolsunderconditionsof continuousinfusionin highdoseama-Ctreatment.Med. Pediatr.Oncol. Suppl., 1: 33—43,1982.

22. Plunkett, W., Liliemark, J. 0., Adams, T. M., Novak, B., Estey, E., Kantarjian, H., and Keating, M. J. Saturation of l-@9-o-arabinofuranosylcytosine5'-triphosphate accumulation in leukemia cells during high dose l-fi-o-arabinofuranosylcytosinetherapy.CancerRes.,47:3005—3011,1987.

23. Slevin, M. L., Piall, E. M., Aherne, G. W., Harvey, V. J., Johnston, A., andLister, T. A. Effect of doseand scheduleon phannacokineticsof high dosecytosine arabinoside in plasma and cerebrospinal fluid. J. Clin. Oncol., 1:546—551,1983.

24. van Prooijen, H. C., Dekker, A. W., and Punt, K. The use of intermediatedosecytosinearabinoside(ID ama-C)in the treatmentof acutenon-lymphocytic leukemia in relapse. Br. J. Haematol., 57: 291—299,1984.

25. Spriggs, D. R., Robbins, G., Takvorian, T., and Kufe, D. W. Continuousinfusionof high dose l-$-i-arabinofuranosylcytosine:a phaseI and pharmacological study. Cancer Res.,45: 3932—3936,1985.

26. Capizzi, R. L., Yang, J. L, Cheng, E., Bjorusson, T., Saharabudhe,D., Tan,R. S., andCheng,Y. C. Alterationofthe pharmacokineticsofhigh doseamaC by its metabolite,high ara-U in patientswith acute leukemia.J. Clin.Oncol.,1:763—771,1983.

27. Ho, D. H. W., and Frei, E. Clinical pharmacology of 1-@9-n-arabinofuranosylcytosine.Clin. Pharmacol.Ther., 12: 944—954,1971.

28. Muller, E. G., and Zahn, R. D. Metabolism of l-@9-n-arabinofuranosyluracilin mouseL5178 Y cells.CancerRes.,39: 1102—1107, 1979.

29. Yang, J. L., Cheng,E. H., Capizzi, R. L., Cheng,Y. C., and Kute, T. Effectof uracil arabinoside on metabolism and cytotoxicity ofcytosine arabinosidein L5178Y murine leukemia J. Clin. Invest., 75: 141—146,1985.

30. Breithaupt,H., Pralle, H., Eckhardt,T., Von Hattingberg,M., Schick,J.,and IAffler, H. Clinical results and pharmacokineticsof high dose cytosinearabinoside (HD ama-C).Cancer (Phila), 50: 1248—1257,1982.

31. Kessel, D., and Shurin, S. Transport of two non-metabolized nucleosides,deoxycytidineand cytosinearabinoside,in a sublineof the L1210 murineleukemia. Biochem. Biophys. Acta, 163: 179—187,1968.

32. Lopez, J. A., and Agarwal, R. P. Acute cerebellar toxicity after high dosecytarabine associatedwith CNS accumulation of its metabolite, uracil arabinoside. Cancer Treat. Rep., 68: 1309—1310,1984.

33. Riccardi, R., Chabner, B., Glaubiger, D. L., Wood, J., and Poplack, D. G.

concentration, higher AUC, decrease of total body clearancedue mostly to inhibition of metabolic clearance, increase ofboth half-life and mean residence time, whereas the volume ofdistribution would be expected to remain unchanged. In addition, total urinary excretion of ama-C as percentage of dosewould increase while the renal clearance would remain unchanged. It is clearly demonstrated in our study that enzymeinhibition took place under the influence of THU. However,instead of an increase, the terminal half-life of ama-C under thetime schedule chosen for this study actually decreased significantly while at the same time a significant decrease of the TBCwas observed. This decrease of terminal half-life was associatedwith a decrease in the volume of distribution, which, indeed,was confirmed by our observations (Table 1). The steep initialdecrease ofara-C plasma levels in the controls after terminationof the infusion slowed down significantly under the influenceof THU and, in many instances, completely disappeared. Sincethese studies extended only over 24 h, a change in the shape ofthe curve after this time has to be considered. The meanresidence time, an overall expression of all time components ofthe disposition processes, on the other hand, was increasedunder the influence of THU. This observation led to the conclusion that THU not only acts as a strong inhibitor of thedeamination of ama-C but also exerts an effect upon the distribution of ama-C.

Large intrapatient variation may possibly be attributed toprogressive disease and varying concomitant medications. In

terpatient variations can be explained by differences in quantityof deaminase and/or kinases among different individuals. Substantial variation of plasma ama-C levels were also reported forhigh-dose ama-C by several investigators (14, 23, 26, 30).

When ama-C is given in combination with THU. the strikingincrease of ama-C levels, substantial increase of AUC for ama-C,and striking decrease of total body clearance and volume ofdistribution are likely to be beneficial in improving the overallexposure of the target cell to ama-C with possible increase ofspecificity as reported by Ho et a!. (37) and Chou et a!. (18) forin vitro studies, with less CNS toxicity. The observed toxicities,such as nausea and vomiting, controllable with antiemetics, andrnyelosuppression were qualitatively but not quantitatively cornparable to the ones observed in HDA regimens (14, 17, 30).Due to the short interval between administration of ama-Caloneand ama-C plus THU. no conclusions could be drawn on theindividual contribution to toxicity of ama-C or THU, respectively. The clinical usefulness ofthis combination is being testedpresently in patients with leukemia.

ACKNOWLEDGMENTS

The authors are indebted to Dr. Youcef Rustum and Dr. CharlesYoung for their suggestions and comments. The nra-C used for thisstudy was kindly supplied through the courtesy of Dr. R. L. Royer ofthe Upjohn Company. The THU was kindly supplied by the NationalCancer Institute through the courtesy ofJ. P. Davignon.

REFERENCES

1. Rudnick, S. A., Cadman, E. C., Capizzi, R. L., Skeel, R. T., Bertino, J., andMcIntosh, S. High dosecytosinearabinosidein refractoryacuteleukemia.Cancer (Phila.), 44: 1189—1193, 1979.

2. Southwest Oncology Group. Cytarabine for acute leukemia in adults. Effectofschedule on therapeutic response.Arch. Intern. Med., 133: 251—259,1974.

3. Hanze, A. R. Nucleic Acids. IV. The catalytic reduction of pyrimidinenucleosides (human liver deaminase inhibitors). J. Am. Chem. Soc., 89:6720-6725, 1967.

4. Mulligan, L. T., and Mellet, L. B. Inhibition of arabinosylcytosinedeamination by tetrahydrouridine in the monkey. Pharmacologist, 12: 221, 1970.

1341

Research. on August 22, 2021. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

PHARMACOKINETICS OF ama-C + ThU IN HUMANS

Influence of tetrahydrouridine on the pharmacokinetics of intrathecally ad- 36. Camiener, G. W. Studies of the enzymatic deamination of ara-cytidine. V.ministered l-@3-r-arabinofuranosylcytosine.Cancer Res., 42: 1736—1739, Inhibition in vitro and in vivoby tetrahydrouridine and other reducedpyrim1982. idine nucleosides.Biochem. Pharmacol., 17: 1981—1991,1968.

34. Kreis,W., Hession,C., Soricelli,A., and Scully,K. Combinationof tetrahy- 37. Ho, D. H. W., Carter, C. J., Brown, N. S., Hester, J., MCCredie, K.,drouridine and cytosine arabinoside in mouse tumors. Cancer Treat. Rep., &njamin R. S., Freireich, E. J., and Bodey, G. P. Effects of tetrahydrouri61: 1355—1364,1977. dine on the uptake and metabolism of l-fi-n-arabinofuranosylc$osine in

35. Wentworth, D. F., and Wolfenden, R. On the interaction of 3,4,5,6-tetrahy- human normal and leukemic cells. Cancer Res.,40: 2441—2446,1980.drouridine with human liver cytidine deaminase. Biochemistry, 14: 5099—5105,1975.

1342

Research. on August 22, 2021. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from

1988;48:1337-1342. Cancer Res   Willi Kreis, Keith Chan, Daniel R. Budman, et al.   over Three Hours-d-Arabinofuranosylcytosine When Both Drugs Are Coinfused

βEffect of Tetrahydrouridine on the Clinical Pharmacology of 1-

  Updated version

  http://cancerres.aacrjournals.org/content/48/5/1337

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/48/5/1337To request permission to re-use all or part of this article, use this link

Research. on August 22, 2021. © 1988 American Association for Cancercancerres.aacrjournals.org Downloaded from