(CANCER RESEARCH 50. 503-508, February 1, 1989]

Disposition of Amsacrine and Its Analogue 9-({2-Methoxy-4-[(methylsulfonyl)-amino]phenyl}amino)-Ar,5-dimethyl-4-acridinecarboxamide (CI-921) inPlasma, Liver, and Lewis Lung Tumors in Mice1

Philip Kestell,2 James W. Paxton, Pandora C. Evans, Deborah Young, Jeffrey L. Juriina,

Iain G. C. Robertson, and Bruce C. BaguleyCancer Research Laboratory ¡P.K., J. L. J., B. C. B.] and Department of Pharmacology and Clinical Pharmacology [J. W. P., P. C. E., D. Y., I. G. C. R.J, Universityof Auckland Medical School, Private Bag, Auckland, New Zealand

ABSTRACT

9-(|2-Methoxy-4-((methylsuIfonyl)amino|phenyljamino)-Ar,5-di-methyl-4-acridinecarboxamide (CI-921), an analogue of the clinical an-tileukemia drug amsacrine with improved solid tumor activity in mice, iscurrently being evaluated in patients. In order to determine whether CI-921 possesses any advantages over amsacrine in terms of tissue delivery,the pharmacokinetics of amsacrine and CI-921 were determined followingi.v. injection in male B6D2F] mice. Plasma kinetics in normal mice weremeasured following administration of 14.4, 28.9, and 57.7 ftmol/kg. Thekinetics in s.c. Lewis lung tumors, and in plasma and livers of normaland tumor-bearing mice were measured following administration of 57.7«imol/kg.CI-921 and amsacrine were quantitated by high-performanceliquid chrontatography after extraction from plasma and from liver andtumor homogenates. In experiments with appropriate 'I I-lahelcd com

pounds, both total and covalenti) bound radioactivity (determined afterprecipitation and washing with acetonitrile) were measured in plasmaand in liver homogenates. Over this dose range, nonlinear kinetics wereobserved in plasma for unchanged CI-921 and amsacrine, and a reasonable fit was obtained with Michaelis-Menten kinetics to a one-compartment model for CI-921 (#„3.7 »imol/liter;Vm, 18 ^mol/h/kg; V,, 3.3liter/kg) and a two-compartment model for amsacrine (K„3.6«imol/liter;' m.»76 Mmol/h/kg; V,, 4.8 liter/kg). The area under the concentration-time curve (AUC) for plasma following a dose of 57.7 «imol/kgwas 31Minol•¿�h/liicr for CI-921 and 6.3 *<mol•¿�h/liter for amsacrine. However,equilibrium dialysis measurements indicated high plasma protein bindingwith free drug fractions for CI-921 and amsacrine of 0.63 and 6.7%,respectively. In the liver, unchanged drug concentrations and total radioactivity for both compounds were approximately 10-fold those in plasma,and the tissue half-life of CI-921 was approximately 4-fold longer for CI-921 than for amsacrine. Plasma and liver kinetics in mice with s.c. Lewislung tumors were similar to those in normal mice. Tumor half-lives ofunchanged CI-921 and amsacrine were 3.9 and 2.7 h, respectively,considerably longer than those for plasma (1.2 and 0.30 h respectively)or liver (1.2 and 0.28 h, respectively). Tumor AUC values for CI-921 andamsacrine were 68 and 37 ¿iniol•¿�h/liter, respectively, as compared to thecalculated AUC values for free drug in plasma of 0.19 and 0.42 «imol-h/liter, respectively. It is concluded that the uptake into tumors from theplasma free drug fraction is more efficient for CI-921 than for amsacrine.

INTRODUCTION

Amsacrine (Fig. 1) is a 9-aminoacridine derivative first synthesized by Cain and Atwell (1) and now used in the treatmentof acute leukaemia, particularly in combination with agentssuch as cytosine arabinoside (2). Amsacrine was developed andselected predominantly on the basis of testing with murineleukemias (1). Its narrow clinical spectrum of action raised thequestion of whether the use of a preclinical screening protocol

Received 5/8/89: revised 9/11/89; accepted 10/10/89.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.

1Supported by the Auckland Division of the Cancer Society of New Zealand,the Medical Research Council of New Zealand and a Warner-Lambert Fellowship.

2To whom requests for reprints should be addressed, at Cancer Research

Laboratory, University of Auckland Medical School. Private Bag, Auckland, NewZealand.

employing a solid tumor would have resulted in the selectionof a different compound in this series. The Lewis lung carcinoma, only slightly sensitive to treatment by amsacrine, wasthus used in this laboratory to search for more active analoguesof amsacrine (3). A disubstituted derivative, CI-9213 (Fig. 1),

found to be curative against i.v. inoculated Lewis lung tumors(4) was selected for further development and has now completedPhase I clinical trials (5, 6).

CI-921 and amsacrine both bind by intercalation to double-stranded DNA, with CI-921 having a 16-fold higher associationconstant (4). Both compounds appear to act by inducing theformation of covalent links between DNA and the enzymetopoisomerase II (7). In continuous drug-exposure growth-inhibitory assays, carried out with a number of human andmouse cell lines, CI-921 requires approximately 4-fold lowerconcentrations than does amsacrine to inhibit growth (4, 8-9).However, these results do not explain why CI-921 has superiorsolid tumor activity to amsacrine in vivo. We have thereforeinvestigated whether pharmacokinetics plays a role in determining the superiority of CI-921 over amsacrine, using theadvanced s.c. Lewis lung tumor in mice which is known to bemore sensitive to CI-921 than to amsacrine (10). We have alsodetermined whether the presence of a s.c. tumor modifies thepharmacokinetics of these drugs.

MATERIALS AND METHODS

Materials. The hydrochloride salts of amsacrine and [acridinyl-G-3H]amsacrine (406 mCi/mmol; 99% radiochemical purity), and the 2-hydroxethanesulfonate salts of CI-921 and [acridinyl-G-3H]C\-92l (4.1mCi/mmol; 97% radiochemical purity) and [carboxamide-ltC]Cl-92l

(specific activity, 59.9 mCi/mmol; >99% radiochemical purity) werekindly provided by Dr. Lloyd Whitfield, Parke-Davis Division of theWarner-Lambert Company, MI. Radiochemical purity was confirmedby TLC (chloroform:methanol, 10:1 v/v or 4:1 v/v) and by HPLC. Allother chemicals and solvents used were either analytical or HPLCgrade.

Animals. Male B6D2F, mice (20-25 g) were used for all experiments.They were bred and maintained in the laboratory under constanttemperature and humidity with sterile bedding and food (3) accordingto institutional guidelines. For studies with tumor-bearing mice, groupsof animals were inoculated s.c. with IO6Lewis lung cells into the right

flank and the experiments were performed after approximately 14 dayswhen the tumors were palpable (5-10-mm diameter, weight 0.42 ±0.24 g SD).

Drug Formulation and Administration. CI-921 solutions (10 HIM)were prepared by sonication in sterile water and amsacrine solutions(200 mM) were prepared in TV.A'-dimethylacetamide. Solutions werediluted with water (CI-921) or 34 mM lactic acid (amsacrine) to the

3The abbreviations used are: CI-921,9-({2-methoxy-4-((methylsulfonyl)amino]phenyl|amino)-A',5-dimethyl-4-acridinecarboxamide; GSH, glutathione: HPLC,high-performance liquid chromatography; TLC, thin-layer chromatography; tn,elimination half-life; K„,steady-state volume of distribution; Cm„,maximumconcentration; AUC, area under the concentration-time curve extrapolated toinfinity; CV, coefficient of variation.

503

Research. on January 20, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

PHARMACOKINETJCS OF CI-921

CH,0

N "Y" N'

CH, CONHCH,

Fig. 1. Chemical structures of amsacrine (left) and CI-921 (righi)

required concentration before injection. Tritiated compounds wereformulated similarly, with amsacrine being diluted with nonradiola-beled material to the required specific activity (5.8 mCi/mmol).

CI-921 and amsacrine were administered to mice i.v. at dose levelsof 14.4 (11.5 for amsacrine), 28.9, and 57.7 ^mol/kg. The lattercorresponded to the i.v. dose (30 mg/kg) providing the greatest increasein lifespan of Lewis lung tumor-bearing mice (4). It was also used forthe study of the disposition of tritiated compounds and for the comparison of the kinetics in normal and tumor-bearing mice. Normal andtumor-bearing mice treated with CI-921 were killed 5 and 30 min, l,2, 3, 4, and 5 h after administration while animals treated with amsacrine were killed after 5, 15, 30, and 45 min, 1, 1.5, 2, 3, 4, and 5 h (5per time point). Additional time points at 8, 16, 24, and 48 h wereincluded for tumor-bearing mice. Blood was collected in heparinizedtubes by ocular extrusion under ether anesthesia and the plasma separated by centrifugation. Livers and tumors were rapidly removed andblotted dry before storage. All tissues and plasma were frozen and keptat -80°Cuntil analysis.

Drug Analysis. The method of Jurlina and Paxton (11) was modifiedto determine the concentrations of unchanged CI-921 and amsacrinein liver, tumor, and plasma. Livers and tumors were homogenized inwater (4 ml) before analysis. The ethylsulfonanilide analogs of CI-921and amsacrine (synthesized by Dr. W. A. Denny) were used as internalstandards. Aliquots of plasma (100-250 ^1), liver and tumor homoge-nates (500 /¿I)were adjusted to pH 9.0 with saturated sodium tetraboratesolution (200-500 M')and extracted with freshly distilled diethyl ether(5-10 ml) for 15 min. After centrifugation at 1100 x g for 10 min theorganic solvent was removed and evaporated under a stream of nitrogen.

The residue was redissolved in mobile phase (100-200 n\) andaliquots were analyzed on a HPLC system (Waters Associates, Milford,MA) consisting of a WISP 712 sample injector, 6000A pump, Z-module, 4-^ Novapak Ct8 cartridge, 840 data control station, and aModel 400 electrochemical detector (EG&G Princeton Applied Corp.,NJ) at a potential of+500 mV. The mobile phases were l Mammoniumacetate (pH 5):acetonitrile:methanol (3:2:2 v/v/v) for CI-921 and l Mammonium acetate (pH 5):acetonitrile (7:4 v/v) for amsacrine, at a flowrate of 1.5 ml/min. The retention times for CI-921 and amsacrine were7.3 and 4.3 min, respectively. Plots of the ratio of peak heights of CI-921 or amsacrine to the internal standard against known drug concentrations were linear in the range 0.1-10 /imol/liter for tumor homoge-nates, 0.1-20 ^mol/liter for liver homogenates and 0.1-15 ^mot/literfor plasma. Theoretical recoveries for both drugs in all samples were100 ±5% and the limits of quantitation with a CV < 10% were 0.05¿iinol/literin plasma, and liver and tumor homogenates. The interassayCV for all samples analyzed was 2-13% for both CI-921 and amsacrine.

Radioactivity Assay. Total radiochemical equivalents in plasma andliver homogenates were determined by liquid-scintillation counting withquench correction by external standardization. Aliquots of plasma (20-100 pi) were diluted to 1 ml with distilled water and mixed with 10-mlscintillation fluid (Aquasol-2, New England Nuclear, Boston, MA)before counting. Aliquots (50 n\) of liver homogenates were first digested in Soluene-350 (New England Nuclear). Covalently bound radioactivity in tissues was determined after precipitation and washingwith acetonitrile as previously described (12).

Free Fraction of CI-921 and Amsacrine in Plasma. The plasma freefractions of [14C]-CI-921 and ['Hjamsacrine were determined by equi

librium dialysis using a Dianorm Dialyser (Diachema, Switzerland) anda cellulose acetate membrane with a 12.000 molecular weight cut-off(Visking, Union Carbide Corp., NY). The protein concentration in eachplasma sample was determined before and after each dialysis by the

NHSO?CH, method of Lowry et al. (13). The appropriate radiolabeled compoundwas diluted with unlabeled drug and added to plasma from normal andtumor-bearing mice to a concentration of 10 ^mol/liter. The plasmawas adjusted to pH 7.4 with CO; and dialyzed for 2 h at 37°Cagainst

isotonic phosphate buffered saline, pH 7.4. After dialysis the radioactivity in buffer and plasma was determined by liquid scintillationcounting. The free fraction (fu'), calculated as the ratio of radioactivity

in buffer to plasma, is independent of any loss of label to the apparatusor membrane but does not take into account the volume changes whichoccur due to colloidal osmotic fluid shift during dialysis (14). Thecorrected free fraction (fu) was calculated by the equation proposed byHuang (15):

fu=fu'-R/[fu'-R+ 1 -fu']

where R = (plasma protein concentration after dialysis)/(plasma proteinconcentration before dialysis).

Pharmacokinetic Methods. The AUC was determined by the trapezoidal rule up to the maximum concentration and by the log trapezoidalrule thereafter, with extrapolation to infinity by addition of C/fi, whereßis the terminal slope of the log concentration-time curve, estimatedby unweighted least-squares regression and C, is the estimated concentration at the last time point. The elimination half-life (/./,) was calculated as 0.693//1 The mean concentration-time profiles from the threedose levels were fitted to one and two compartment models with eitherlinear or Michaelis-Menten kinetics by MKMODEL, an extended leastsquares modelling system, and models were compared using theSchwarz Criterion (16). Elsewhere, data were compared by Student's t

test, with differences considered significant if P < 0.05.

RESULTS

Plasmokinetics of Unchanged CI-921 and Amsacrine in Normal Mice. The plasma concentration-time profiles for unchanged CI-921 and amsacrine are shown in Fig. 2. Dose-dependent elimination kinetics were observed for both compounds, as indicated by the increasing half-life (for CI-921) andthe overproportional increase in the AUC with increasing dose(Table 1). Four- and 5-fold increases in dose resulted in 9.6-and 13-fold increases in the AUC for CI-921 and amsacrinerespectively. In addition, the AUC 0/mol-h/liter) for CI-921was approximately 5-fold that for amsacrine. Equimolar dosesof CI-921 also produced a higher Cmax(1.5-fold) and a longertv, (4-fold) than amsacrine.

Comparison of the plasma concentration-time data (Fig. 2)fits to a one- or two-compartment model with linear or Michaelis-Menten kinetics indicated the best fit for CI-921 was toa one-compartment model (Km, 3.7 ¿imol/liter; Kraas,18 ¿¿mol/h/kg; yss, 3.3 liter/kg). The best fit for amsacrine was to a two-compartment model (Km, 3.6 ¿imol/liter; Fmax,76 ¿imol/h/kg;ym 4.8 liter/kg).

Comparison of Plasma and Liver Kinetics in Normal andTumor-bearing Mice. The plasma and liver concentration-timeprofiles of unchanged CI-921 and amsacrine for both normaland tumor-bearing mice following a dose of 57.7 /umol/kg areshown in Fig. 3. The mean plasma and liver concentrations atthe various time points for either drug did not show anysignificant difference between normal and tumor-bearing mice.Similar AUC and tv, values were observed in normal and tumorgroups (Table 2). For both drugs and both groups, the amountin the liver was considerably higher than in plasma but ?./,valueswere similar for each compound in plasma and liver.

Comparison of Tumor Kinetics. To define the AUC for unchanged drug in tumor, additional determinations were madeat 8, 16, 24, and 48 h after treatment. However, by 24 hconcentrations were at or below the limit of sensitivity of thetumor assay. Large variations in the tumor levels of unchanged

504

Research. on January 20, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

PHARMACOKINETICS OF CI-921

Fig. 2. Mean plasma concentrations of unchanged CI-921 (left) and amsacrine (right) asa function of time after i.v. administration ofdoses of 57.7 (D), 28.9 (A), and 14.4 Mmol/kg(11.5 for amsacrine) (O). Vertical bars, indicateSD. Inserts, fit of the data to the Michaelis-Menten model.

E3 1.0ÃœzO 0.5O

0.1

005

345TIME (h)

1.0 1.5TIME (h)

Table 1 Plasma pharmacokinetics of unchanged CI-921 and amsacrine innormal mice

Dose Cm„ AUCfcmol/kg) (nmol/liter) t* (h) fcmol-h/liter) AUC/dose

CI-92157.7 20 1.2 31 0.5328.9 10 0.90 11 0.3814.4 3.8 0.60 3.2 0.22

Table 2 Plasma, liver, and tumor pharmacokinetics of unchanged CI-921 andamsacrine in normal and tumor-bearing mice at 57.7 ^mol/kg dose

CI-921(a)Plasma

NormalTumor

f»(h)AUC

(^mol-h/liter) AUC(a)/AUC(b)

20 ±1.719 ±0.8

1.21.0

3131

4.95.2

Amsacrine57.728.911.5

136.52.3

0.300.180.27

6.31.90.5

0.110.070.04

LiverNormalTumor

Tumor

Amsacrine(b)Plasma

NormalTumor

250 ±54 1.2280 ±100 1.2

11 ±4.4 3.9°

13 ±1.5 0.3012 ±1.2 0.33

710830

68

6.36.0

9.69.1

1.8

LiverNormalTumor

110 ±12 0.28140 ±8 0.31

7491

i 100o

OP

UU

OU

10

1.O

0.1

Tumor 9.9 ±5.8 2.7" 37

1.0 2.0 3.0TIME (h)

4.0 5.0

Fig. 3. Mean concentrations of unchanged CI-921 and amsacrine in plasmaand liver of normal (solid lines) and tumor-bearing (broken lines) mice as afunction of time after an i.v. dose of 57.7 ><mol/kg.

" ^mol/kg for liver and tumor.* Calculated by unweighted least-squares linear regression from 2 h onwards.

drug were observed (Fig. 4), but mean concentrations were notsignificantly different for CI-921 and amsacrine except at 5 hand later. Tumor half-lives for CI-921 and amsacrine were 3.9±0.4 h and 2.7 ±0.2 h, respectively. These values weresignificantly different (P < 0.001). The AUC for CI-921 intumors was 1.8-fold that for amsacrine.

Disposition of Radioactive CI-921 and Amsacrine in NormalMice. Groups of mice were administered [3H]-CI-921 or [3H]-

amsacrine (57.7 ¿¿mol/kg).The observed profiles of total radioactivity and radioactivity bound to precipitated material, together with unchanged parent drug are shown in Figs. 5 and 6.At 5 min (the first time point) unchanged drug (as measured byHPLC) accounted for 73% (CI-921) and 48% (amsacrine) oftotal radioactivity present in plasma. In addition, 17% of CI-921 radioactivity (increasing to 30% by 30 min) and 23% ofamsacrine radioactivity (increasing to 56% by 2 h) appeared tobe covalently bound to precipitated plasma proteins.

505

Research. on January 20, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

PHARMACOKINETICS OF CI-921

ÃŽ

AMSACRINE

11 12 13 14 15 IB

TIME (h)

Fig. 4. Mean concentration of unchanged CI-921 (upper curve) and amsacrine(lower curve) in Lewis lung tumors after an i.v. dose of 57.7 ^mol/kg. Dashedlines, plasma concentrations for comparison.

1.0

Fig. 5. Disposition of [3H)-C1-921 in liver and plasma of normal mice afteran i.v. dose of 57.7 ^mol/kg. O«. total radioactivity; DB, CI-921 by HPLC;A A, bound radioactivity.

For both drugs, total radioactivity present in liver was considerably higher (approximately 10-fold) than that in plasma.Unchanged drug represented a higher proportion of total radioactivity for CI-921 (74% at 5 min) than for amsacrine (46% at

5 min), while covalently bound radiolabel comprised 21% ofCI-921 radioactivity (increasing to 40% at 5 h) and 15% of

amsacrine (increasing to 77% at 2 h).Plasma Free Fraction by Equilibrium Dialysis. The free frac

tion of CI-921 and amsacrine was determined at a drug concentration of 10 ^mol/liter in plasma from untreated normal andtumor-bearing mice (Table 3). The free fraction of amsacrinein plasma (0.067) was approximately 10-fold that of CI-921(0.006). There was no significant difference in the free amsacrine fraction in plasma from normal and tumor-bearing mice,but for CI-921 there was a significant increase (P =0.01) in thefree fraction in plasma from tumor-bearing mice (0.008).

Fig. 6. Disposition of [3H]amsacrine in liver and plasma of normal mice afteran i.v. dose of 57.7 ^mol/kg. O •¿�,total radioactivity; D •¿�amsacrine by HPLC;A A, bound radioactivity.

Table 3 Percentage free fraction for CI-921 and amsacrine in plasma fromnormal and tumor-bearing mice

MeanSDnr testNormal0.633%0.107

gCI-921Tumor-bearing0.795%0.1097

P = 0.01AmsacrineNormal6.72%

0.72gTumor-bearing7.00%

0.527NS

DISCUSSION

The plasma pharmacokinetics of CI-921 and amsacrine havebeen examined at three doses up to the maximum tolerateddose. Both compounds give reasonable fits to Michaelis-Men-ten kinetics. Correcting these kinetic parameters, taking intoaccount the free drug fraction for CI-921 (0.0063) and amsacrine (0.0672), gives the following values: CI-921 (free): K„,0.02 ^mol/liter; Fmax,0.11 ^mol/h/kg; Vsi, 520 liter/kg. Amsacrine (free): Km, 0.25 ^mol/liter; Fmax,5.1 ^mol/h/kg; Vm =71 liter/kg. The lower Km for CI-921 suggests a higher affinityfor the eliminating mechanism, while the higher Fmaxfor amsacrine suggests a much greater capacity for the removal ofamsacrine.

The reasons for the dose-dependent kinetics observed forthese agents are not known. A possible reason is saturation ofan elimination pathway. For both drugs biliary excretion is themajor route of elimination, with GSH conjugates representingthe major metabolites present in bile (17, 18, 21). Both compounds have been observed to deplete hepatic GSH after administration of high doses (17, 18) and depletion of hepaticGSH by pretreatment with buthionine sulfoximine altered theplasma kinetics of both drugs in the rabbit (19, 20). Furtherstudies on the role of GSH in the metabolism of these compounds are underway. It is also unlikely that concentration-dependent plasma protein binding plays a part in the nonlinearkinetics since saturation of plasma binding was observed forCI-921 alone (22) and not for amsacrine over the range 1-50/^mol/liter (23). At low plasma concentrations (relative to Km),

506

Research. on January 20, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

PHARMACOKINETICS OF CI-921

the VmiJKm ratio can be regarded as an index of plasmaclearance, i.e., 4.7 liter/h/kg for CI-921 and 21 liter/h/kg foramsacrine. Using a blood/plasma concentration ratio of 0.74for CI-921 and 0.82 for amsacrine (from addition of appropriateradiolabel to mouse blood), blood clearance values of 6.4 and26 liter/h/kg are obtained for CI-921 and amsacrine, respectively. The hepatic blood flow in the mouse is reported to be 5liter/h/kg (24) which suggests that both CI-921 and amsacrinemay undergo high hepatic extraction and/or other eliminationpathways play a significant role in the removal of parent drug.This is in contrast to rabbits and humans where both drugsappeared to experience low hepatic extraction (6, 25, 26). Anadditional elimination pathway which may be of major significance is the biotransformation of the compounds to covalentadducts in the blood (12, 27). This reaction appears to belargely species specific for the mouse and may partly explainthe greater weight-normalized clearance values observed inmice. Further studies on these mechanisms are in progress.

It is interesting to compare the AUC values for CI-921 in

mice with those obtained in clinical studies. A phase I study ofCI-921 (6) at a dose of 216 mg/m2 (11 ^mol/kg ±2.1 SD)provided an AUC of 75 /¿mol•¿�h/liter (±13.6SD), corresponding to a free drug AUC of 0.075 /¿mol•¿�h/liter.4 This dose is

approximately one third of the maximum phase I single dose(28). In comparison, CI-921 administered to mice at one halfof the optimal dose provides a total drug AUC of 11 /umol-h/liter and a free drug AUC of 0.069 Mmol-h/liter (Tables 1, 3).The free drug AUC values for humans and mice are thereforecomparable, even though the total drug AUC values are muchhigher for humans. Drug AUC values have been postulated tobe useful in predicting clinical dosage from mouse data (29)and the present results suggest that free drug AUC values maybe more useful parameters in this regard.

AUC values for amsacrine in human studies have previouslynot been published. The established phase I single dose ofamsacrine is 150 mg/m2 (30). A phase III study (26) of amsacrine in leukemia patients at a dose of 200 mg/m2 (12 ^mol/kg±1.5 SD) provided a total drug AUC of 52 ^mol-h/liter (±16SD), corresponding to a free drug AUC of 1.5 ¿¿mol•¿�h/liter.In comparison, amsacrine administered to mice at the optimaldose provides a total drug AUC of 6.3 ^mol •¿�h/liter and a freedrug AUC of 0.40 ¿¿mol•¿�h/liter. Again, free drug AUC valuesare more comparable for the two species than are total drugAUC values.

It is clear from Fig. 3 that the presence of a s.c. Lewis lungtumor has little effect on drug-elimination kinetics in plasmaand liver. However, this tumor does not metastasize to the liverand at the stage of growth employed, one might expect littleeffect on liver metabolism, since the tumor did not exceed 2%of the body weight (31). Indeed if these compounds undergohigh hepatic extraction as our results suggest, the eliminationkinetics would be more susceptible to changes in hepatic bloodflow rather than inhibition of hepatic metabolizing enzymes.Nevertheless the presence of the s.c. tumor slightly increasesthe plasma free drug fraction (from 0.0063 to 0.008) for CI-

921 but not for amsacrine. We observed no significant difference in the amsacrine free fraction in plasma from leukemiapatients and healthy controls (23) but a similar study has notbeen undertaken with CI-921.

The liver appears to be highly efficient in trapping both drugswith a concentration ratio of 10- to 20-fold as compared withplasma. The parallel nature of elimination curves for CI-921

4 Unpublished.

and amsacrine suggests efficient exchange between plasma andliver compartments. Drug loss from liver occurs much morerapidly than from s.c. tumor (Fig. 5), presumably because theaverage diffusion distance is considerably shorter for the well-vascularized liver than for the poorly vascularized tumor. Thecomparatively high liver drug levels and slow tumor eliminationkinetics of CI-921 and amsacrine can be compared with thosefor doxorubicin in mice with Colon 38 tumors. Liver drugconcentrations of doxorubicin are 100-fold those in plasma andtv, values in plasma and tumor are 10 and 46 h, respectively(32).

Although the half-life of CI-921 in plasma is 4-fold longerthan that for amsacrine, it is only 1.4-fold longer in tumors,indicating that plasma kinetics do not predict tumor kinetics.The question of whether this difference in tumor kinetics canexplain the superiority of CI-921 against a number of mousetumors including Lewis lung is of obvious interest in terms ofpredicting the clinical potential of CI-921. It is noteworthy thatthe free drug plasma AUC for CI-921 is 45% that of amsacrine,whereas the tumor-associated AUC for CI-921 (68.3 ¿imol-h/liter) is 1.8-fold higher than that for amsacrine. This indicatesmore efficient uptake and retention of CI-921 by the tumor.However, superiority of CI-921 over amsacrine may also existat the level of cell selectivity, since CI-921 has similar in vitrotoxicity to amsacrine against mouse and human bone marrowcultures, but superior toxicity against some tumor cells including L1210 murine leukemia and Lewis lung cells (33). Bothpharmacokinetic and cell selectivity mechanisms may apply tohumans as well as mice, suggesting that CI-921 may have agreater clinical efficacy.

ACKNOWLEDGMENTS

Thanks are extended to Professor W. A Denny for synthesizing oneof the compounds used in this study, to Dr. N. H. G. Holford forwriting the models, to Chris Thoreau for excellent technical assistance,and to Lynden Hull for preparing the manuscript.

4 REFERENCES

1. Cain, B. F.. and Atwell. G. J. The experimental antitumour properties ofthree congeners of the acridinyl methanesulphonanilide (AMSA) series. Eur.J. Cancer, 10: 539-549, 1974.

2. Arlin. Z. Current status of amsacrine combination chemotherapy programsin acute leukemia. Cancer Treat. Rep., 67: 967-970, 1983.

3. Baguley, B. C, Kernohan, A. R., and Wilson, W. R. Divergent activity ofderivatives of amsacrine (m-AMSA) towards Lewis lung carcinoma and P388leukaemia in mice. Eur. J. Cancer Clin. Oncol. 19: 1607-1612, 1983.

4. Baguley, B. C., Denny, W. A., Atwell, G. J., Finlay, G. J.. Rewcastle, G. W.,Twigden, S. J., and Wilson, W. R. Synthesis, antitumor activity and DNAbinding properties of a new derivative of amsacrine yv,5-dimethyl-9-((2-methoxy-4-methylsuIfonylamino)phenylamino)-4-acridinecarboxamide.Cancer Res., 44: 3245-3251, 1984.

5. Hardy, J. R., Harvey, V. J., Paxton, J. W., Evans, P. C, Smith, S., Grillo-Lopez, A., Grove, W., and Baguley, B. C. A phase 1 trial of the amsacrineanalog 9-[(2-methoxy-4-methylsulfonylamino)-phenylamino]-/V,5-dimethyl-4-acridinecarhoxamide (CI-921). Cancer Res., 48:6593-6596, 1988.

6. Paxton, J. W., Hardy, J. R., Evans, P. C., Harvey, V. J., and Baguley, B. C.The clinical pharmacokinetics of A'-5-dimethyl-9-[(2-methoxy-4-methyl-sul-fonylamino)phenylamino]-4-acridinecarboxamide (CI-921) in a phase I trial.Cancer Chemother. Pharmacol., 22: 235-240, 1988.

7. Covey, J. M., Kohn, K. W., Kerrigan, D., Tilchen, E. J., and Pommier, Y.Topoisomerase Il-mediated DNA damage produced by 4'-(9-acridinyla-mino)-methanesulfon-m-anisidide and related acridines in L1210 cells andisolated nuclei: relation to cytotoxicity. Cancer Res., 48: 860-865, 1988.

8. Finlay, G. J., and Baguley, B. C. The use of human cancer cell lines as aprimary screening system for antineoplastic compounds. Eur. J. Cancer Clin.Oncol., 20: 947-954, 1984.

9. Finlay, G. J., Wilson, W. R., and Baguley, B. C. Comparison of in vitroactivity of cytotoxic drugs towards human carcinoma and leukaemia celllines. Eur. J. Cancer Clin. Oncol., 22: 655-662, 1986.

10. Baguley, B. C., Finlay. G. J., and Wilson, W. R. Cytokinetic resistance ofLewis lung carcinoma to cyclophosphamide and the amsacrine derivative CI-

507

Research. on January 20, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

PHARMACOKINETICS OF CI-921

921. in: T. C. Hall (ed.). Cancer Drug Resistance. Progress in Clinical andBiological Research, Vol. 23, pp. 47-61. New York: Alan R. Liss. Inc., 1986.

11. Jurlina, J. I ., and Paxton, J. W. A high-performance liquid Chromatographieassay for plasma concentrations of A'-5-dimethyl-9-[(2-methoxy-4-methylsul-fonylamino)phenylamino]-4-acridinecarboxamide (CI-921, NSC 343 499). J.Chromatogr., J«: 431-435, 1985.

12. Kestell, P., Paxton, J. W., Robertson, I. G. C., Evans, P. C., Dormer, R. A.,and Baguley, B. C. Thiolytic cleavage and binding of the antitumor agent CI-921 in blood. Drug. Metab. Drug Interact., in press, 1990.

13. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. Proteinmeasurement with the Folin phenol reagent. J. Biol. Chem., 193: 265-275,1951.

14. Tozer, T. N., Gambertoglio, J. G., Fürst,D. E., Avery, D. S., and Holford,N. H. G. Volume shifts and protein binding estimates using equilibriumdialysis: application to prednisolone binding in humans. J. Pharm. Sci., 72:1442-1448, 1983.

15. Huang, J. D. Errors in estimating the unbound fraction of drugs due tovolume shift in equilibrium dialysis. J. Pharm. Sci., 72: 1368-1369, 1983.

16. Holford, N. H. G. MKMODEL, a modeling tool for microcomputers—apharmacokinetic evaluation and comparison with standard computer programs (Abstract). Clin. Exp. Pharmacol. Suppl., 9: 95, 1985.

17. Cysyk, R. L., Shoemaker, D. D., and Admanson, R. H. The pharmacologicaldisposition of 4'-(9-acridinylamino)methanesulfon-m-anisidide in mice andrats. Drug Metab. Disposition, J: 579-590, 1977.

18. Robertson, I. G. C, Kestell, P., Dormer, R. A., and Paxton, J. W. Involvement of glutathione in the metabolism of the anilinoacridine antitumor agentsCI-921 and amsacrine. Drug Metab. Drug Interact., in press, 1990.

19. Paxton, J. W., Foote, S. E., and Singh, R. M. The effect of buthioninesulphoximine, cimetidine and phenobarbitone on the disposition of amsacrine in the rabbit. Cancer Chemother. Pharmacol., 18: 208-212, 1986.

20. Paxton, J. W., Evans, P. C., and Hardy, J. R. The effect of cimetidine,phenobarbitone and buthionine sulphoximine on the disposition of N-5-dimethyl-9-|(2-methoxy-4-methyl(sulphonylamino)phenylamino]-4-acridine-carboxamide (CI-921). Cancer Chemother. Pharmacol., 23: 291-295, 1989.

21. Shoemaker, D. D., Cysyk, R. L., Padmanabhan, S.. Bhat, H. B., and Mal-speis, L. Identification of the principal biliary metabolite of 4'-(9-acridinyl-amino)methanesulfon-m-anisidide in rats. Drug Metab. Dispos.. 10: 35-39,1982.

22. Andress, L. D., Whitfield, L. R.. and Chang, T. Characterisation of CI-921,the amsacrine analog, binding to human plasma and plasma proteins (Abstract). Proc. Am. Assoc. Pharm. Sci., PP1460, 1988.

23. Paxton, J. W., Jurlina, J. L., and Foote, S. E. The binding of amsacrine tohuman plasma proteins. J. Pharm. Pharmacol., 38: 432-438, 1986.

24. Boxenbaum, H. Interspecies scaling, allometry, physiological time, and theground plan of pharmacokinetics. J. Pharmacokinet. Biopharm., 10: 201-227, 1982.

25. Paxton, J. W., and Jurlina, J. L. Comparison of the kinetics and protein binding of the anticancer drug, amsacrine and a new analogue, N-5-dimethyl-9-[(2-methoxy-4-methylsulfonylamino)phenylamino]-4-acridine-carboxamide,in rabbits. Cancer Chemother. Pharmacol., 16: 253-256, 1986.

26. Jurlina, J. L., Varcoe, A. R., and Paxton, J. W. Pharmacokinetics of amsacrine in patients receiving combined chemotherapy for treatment of acutemyelogenous leukemia. Cancer Chemother. Pharmacol., 14: 21-25, 1985.

27. Wilson, W. R., Cain, B. F.. and Baguley, B. C. Thiolytic cleavage of theantitumour compound 4'-(9-acridinylamino)methanesulfon-m-anisidide (m-AMSA, NSC 156303) in blood. Chem. Biol. Interact., 18: 163-178, 1977.

28. Robert, F., Mignucci, M., Grove, W., Javier, J., Asmar, S., and Grillo-Lopez,A. Phase I study of CI-921: single dose schedule (Abstract). Proc. Am. Soc.Clin. Oncol., 6: 26, 1987.

29. Davis, L. E., Alberts, D. S., Plezia, P. M., Roe, D. J., and Griswold, D. P.Predictive model for plasma concentration-versus time profiles of investiga-tional anticancer drugs in patients. J. Nati. Cancer Inst., 80: 815-819, 1988.

30. Legha, S. S., Gutterman, J. U., Benjamin, R. S., Burgess, M. A., Valdivieso,M., and Bodey, G. P. Phase I clinical investigation of 4'-(9-acridinyl-amino)-methanesulfon-m-anisidide (NCS 249992), a new acridine derivative. CancerRes., 38: 3712-3716, 1978.

31. Powis, G., Harris, R. N., Basseches, P. S., and Santone, K. S. Effects ofadvanced leukemia on hepatic drug-metabolising activity in the mouse.Cancer Chemother. Pharmacol., 16: 43-49, 1986.

32. Formelli, F., Carsana, R., and Pollini, C. Comparative pharmacokinetics andmetabolism of doxorubicin and 4-demethoxy-4'-O-methyldoxorubicin in tumor-bearing mice. Cancer Chemother. Pharmacol., 16: 15-21, 1986.

33. Ching, L-M. Finlay, G. J. Joseph, W. R., and Baguley, B. C. Comparison ofthe cytotoxicity of amsacrine and its analogue CI-921 against cultured humanand mouse bone marrow and tumour cells. Eur. J Cancer Clin. Oncol., inpress, 1990.

508

Research. on January 20, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

1990;50:503-508. Cancer Res   Philip Kestell, James W. Paxton, Pandora C. Evans, et al.   and Lewis Lung Tumors in Mice,5-dimethyl-4-acridinecarboxamide (CI-921) in Plasma, Liver,

N9-({;2-Methoxy-4-[(methylsulfonyl)-amino]phenyl}amino)-Disposition of Amsacrine and Its Analogue

  Updated version

  http://cancerres.aacrjournals.org/content/50/3/503

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/50/3/503To request permission to re-use all or part of this article, use this link

Research. on January 20, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from