(CANCER RESEARCH 50, 4417-4422. July 15. 1990]

Evaluation of the Antitumor Activity of Gemcitabine(2 ',2 ' -Difluoro-2 ' -deoxycytidine)

Larry W. Hertel,1 George B. Boder, J. Stan Kroin, Sharon M. Rinzel, Gerald A. Poore,

Glen C. Todd, and Gerald B. GrindeyLilly Research Laboratories, Eli Lilly and Company, Lilly Corporate Center, Indianapolis, Indiana 46285

ABSTRACT

A new pyrimidine antimetabolite, 2',2'-difluorodeoxycytidine, Gem

citabine (LY188011, dFdCyd) has been synthesized and evaluated inexperimental tumor models. dFdCyd is a very potent and specific deoxycytidine analogue. The concentration required for 50% inhibition ofgrowth is 1 ng/ml in the CCRF-CEM human leukemia cell culture assay.Concurrent addition of deoxycytidine to the cell culture system providesabout a 1000-fold decrease in biological activity. The inhibition of growthof human leukemia cells in culture led to the in vivo evaluation of thiscompound as a potential oncolytic agent. Maximal activity in vivo wasseen with dFdCyd when administered on an every third day schedule. 1-/3-D-Arabinofuranosylcytosine, administered on a daily for 10-day schedule, was directly compared to dFdCyd in this evaluation. dFdCyd demonstrated good to excellent antitumor activity in eight of the eight murinetumor models evaluated. l-/3-D-Arabinofuranosylcytosine was substantially less active or had no activity in these same tumor models. This invivo activity against murine solid tumors supports the conclusion thatdFdCyd is an excellent candidate for clinical trials in the treatment ofcancer.

INTRODUCTIONThe biological activities of some 2'-deoxyribonucleosides and

arabinonucleosides have been known for several years (1, 2).Interest in arabinosyl pyrimidines as potential chemotherapeu-tic agents was stimulated by the observation that ara-C2 was

found to have some selective antiviral activity versus DNAviruses such as herpes simplex and vaccinia (3-5).

ara-C is a synthetic compound chosen for clinical trials inacute leukemias because of its activity in leukemia L1210 (6).ara-C is known as the most effective drug in adult acute leukemia (7); however, it is rapidly inactivated by enzymatic deami-nation. Substitution with certain functional groups at the 2'-position of both 2'-deoxyribo- as well as 2'-deoxyarabinopy-

rimidine nucleosides not only alters susceptibility to enzymaticdeactivation, but also modifies the biological activity (e.g., 2'-fluoro-5-iodoarabinosylcytidine and 2'-0-nitro-l-/3-D-arabino-

furanosylcytosine) (8, 9).This report, on a series of l-(2-deoxy-2,2-difluororibofu-

ranosyl)pyrimidines, is part of a program to design, synthesize,and evaluate nucleosides of potential value as anticancer and/or antiviral agents. The synthesis of these novel pyrimidinenucleosides began with a simple and stereocontrolled synthesisof 2-deoxy-2,2-difluoro-D-ribose (10). Condensation of the appropriately substituted difluoro sugar and nucleic acid basesyielded, after removal of the protecting groups, 2-deoxy-2,2-difluororibofuranosyl nucleosides (10). One of these is thepyrimidine antimetabolite, 2',2'-difluorodeoxycytidine, Gem

citabine (LY188011, dFdCyd). Initially this compound was

Received 8/28/89; revised 4/2/90.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.

1To whom requests for reprints should be addressed.2The abbreviations used are: ara-C, 1-fi-D-arabinofuranosylcytosine; dFdCyd,

2',2'-difluorodeoxycytidine; dFdCTP, the triphosphate of dFdCyd; ara-CTP, thetriphosphate of ara-C; ILS, increase in life span; CCRF-CEM, human leukemiacells; L5178Y, murine leukemia cells.

synthesized as a potential antiviral agent. dFdCyd was veryactive in the cell culture antiviral screen, inhibiting both RNAand DNA viruses (11). However, the compound proved to havea narrow therapeutic index when it was administered dailyduring the in vivo evaluation of antiviral activity. The presentstudy describes the in vivo antitumor activity associated withdFdCyd.

MATERIALS AND METHODS

Cell Culture Systems. CCRF-CEM, a human leukemia cell line (12),was grown as previously described (13). Dose-response curves weregenerated for the various compounds as shown in Fig. 2 to determinethe concentration required for 50% inhibition of growth. Cells in RPMI1640 supplemented with 10% dialyzed fetal bovine serum (Grand IslandBiological Co.), 16 mM 4-(2-hydroxyethyl)-l-piperazineethanesulfonicacid, and 8 mM 3-(A'-morpholino)propanesulfonic acid buffers wereadded to the wells at a final concentration of 4.8 x IO4cells/well in atotal volume of 2.0 ml. After 72 h of incubation at 37°C(95% air, 5%

CO2), cell numbers were determined on a ZBI Coulter Counter. Cellnumber for indicated controls at the end of incubation is usually 4-6 xIO5cells/well. Cells were periodically checked for adventitious agents

including Mycoplasma. Recovery from seed stocks frozen in liquidnitrogen was done on a 6- to 9-month schedule in order to maintainlow passage number.

Flow Cytometric Analysis of CCRF-CEM and LSI 78Y Cells in Vitro.Stock cultures of human T-cell leukemia (CCRF-CEM) and L5178Ycells were maintained in RPMI 1640 (Grand Island) supplemented with10% dialyzed fetal bovine serum (Grand Island). Subcultures wereestablished by simple dilution (1:10) into complete medium. dFdCydor ara-C was dissolved in medium, diluted serially to 100 times thefinal concentration in complete growth medium, and added to 4 ml ofmedium containing the cells to achieve the final concentrations. Theinoculum was 3-4 x IO5cells/ml. Incubation of the cells in triplicate,

including controls, was carried out in closed culture tubes (Corning,16-125 mm) at 36.5°Cfor 16 h. At this point, 7 ml hypotonie saline-

propidium iodide solution were added to the cell suspension. Thepreparation was held at 4°Cprior to analysis of DNA content, using a

Coulter EPICS V flow cytometer tuned to 488 nm. This procedure wasmodified after the method of Krishan (14).

Acute Toxicity Testing. The acute toxicity of dFdCyd was evaluatedwhen administered i.v. to young adult male and female ICR mice. Eachanimal (5/sex) received a single maximum dose possible per kg of bodyweight. Control animals (5/sex) received a single i.v. dose of 0.9%saline solution equivalent to the maximum volume of test solutionadministered (25 ml/kg). All animals were observed for 2 weeks afterdosing for mortality and signs of toxicity. Necropsies were conductedon all test animals.

Subchronic Toxicological Evaluation. The study was designed toexpose mice to repeated dose levels of the test compound for 3 months.Groups of 15 male and 15 female C57BL/6 x C3H F, (hereafter calledB6C3Fi) mice were given i.p. dose levels of the test compound. Thefollowing data were obtained: signs of toxicity, mortalities, bodyweights, food consumption, hematology, clinical chemistry, organweights, gross pathology, and histopathology.

Solid Tumor Systems. The adenocarcinoma 755, 6C3HED lympho-sarcoma, M5076, X-5563 myeloma, P-388 leukemia, and LI210 leukemia were obtained from the Division of Cancer Treatment, NationalCancer Institute (15). The B-16 melanoma and P1534J lymphocytic

4417

Research. on February 21, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

ANTITUMOR ACTIVITY OF dFdCyd

leukemia were obtained from The Jackson Laboratory, Bar Harbor,ME (16). First passage tumor was stored in liquid nitrogen usingstandard techniques. Passage or recipient mice (C57BL/6, C3H, DBA/2) were obtained from Charles River Laboratories, Portage, MI.

Solid Tumor Drug Testing. For drug testing, the tumor is removedfrom animals and minced into 1- to 3-mm2 fragments using sterile

techniques. Tumor pieces are checked for sterility using both antibioticMedium I and brain heart infusion medium (Difco, Detroit, MI).Recipient mice are shaved and tumor pieces are implanted s.c. in theaxillary region by trocar. Drug therapy on the appropriate schedule isinitiated on the day after tumor implantation, except where noted.Drugs were dissolved in 2.5% Emulphor for all experiments. All animals were weighed at the beginning and end of drug treatment. Foodand water were provided ad libitum. On days 10 to 12, all tumors weremeasured 2-dimensionally (length and width), using vernier calipers.From these measurements tumor weights could be calculated by usingthe following formula (15):

tumor weight (mg) =tumor length (mm) x tumor width (mm)2

There were 10 mice/treatment group with 20 mice/control group. Theonly groups included in the analysis for therapeutic activity for ara-Cwere those in which drug toxic deaths did not exceed 10% of the treatedgroup. There were no toxic deaths with the dFdCyd-treated groups.Compounds were considered active if more than 60% inhibition oftumor growth was achieved at maximally tolerated doses. No significantweight loss was noted for dFdCyd-treated groups at 80 mg/kg or belowin any of the experiments.

Leukemia Drug Testing. For drug testing, DBA/2 mice (CharlesRiver Laboratories) are inoculated i.p. with 1 x IO6 cells and drug

therapy is initiated 24 h later. Drug activity for these leukemia modelsis calculated by the ILS as compared to the control group. No treatedanimals survived 60 days postinoculation.

RESULTS

Initial evaluation of the antitumor activity of a variety ofdifluorodeoxynucleoside analogues (Fig. 1) in addition todFdCyd was conducted by using CCRF-CEM human leukemiacells in culture. dFdCyd is a potent inhibitor of cellular replication, as the concentration of drug required for 50% inhibitionof growth is approximately 1 ng/ml (Fig. 2). The comparable5-methyl analogue is about 300-fold less potent and the uridineanalogue is about 5,000-fold less active. The 5-fluorouridineanalogue was completely inactive up to 20 pg/ml. The thymi-dine analogue was active only at 180 pg/m\, which is equivalentto a 180,000-fold decrease in biological activity as compared todFdCyd (17).

ara-C is also a potent inhibitor of CCRF-CEM cells inculture, with 50% inhibition of growth being observed at about1 ng/ml (Fig. 3). Very low concentrations of deoxycytidine (3Mg/ml) substantially reduce the activity of ara-C, and 10 ng/m\

NH2

HO

R=H,F,CH3

Fig. 1. Structures of the various 2'.2'-difluorodeoxynucleosides evaluated forantiviral and anticancer activity. 1. R = H. 2'-deoxy-2',2'-difluorocytidine: 1.R =F, 2'-deoxy-2'.2'-difluoro-5-fluorocytidine: 1. R=CH3, 2'-deoxy-2',2'-di-fluoro-5-methylcytidine; 2. R=H, 2'-deoxy-2',2'-difluorouridine: 2, R=F, 2'-deoxy-2',2'-difluoro-5-fluorouridine: 2. R=CH,, 2',2'-difluorothymidine.

10CH

I 80-co^ 60-

o—-a.

oI40-Boo.

20-0-r\

\\\\\0.001

0.01\\nVVta»0.1 1\\\VT*i 10

Dose (/ig/ml)Fig. 2. Activity of various difluorodeoxynucleoside analogues against CCRF-

CEM cells. The compounds tested were: dFdCyd (x x); 2'-deoxy-2',2'-difluoro-5-methylcytidine (O O); 2'-deoxy-2',2'-difluorouridine (X X);2'-deoxy-2',2'-difluoro-5-fluorouridine (O O). The concentration of 2',2'-

difluorothymidine to achieve 50% inhibition of growth was 180 i,.u mh

100-,

O 60-

40-

Q- 20-

0-0.1 10.0001 0.001 0.01

ARA-C (/-g/ml)

Fig. 3. Ability of deoxycytidine to decrease growth inhibition of CCRF-CEMcells induced by ara-C. Deoxycytidine concentrations: none (x); 100 ^g/rnl (O);30 Mg/ml (•);10 ¿ig/ml(A); 3 ^g/ml P).

100-,

80-

60-

20-

0.0001 0.001 0.01 0.1 1 10dFdC (¿ig/ml)

Fig. 4. Ability of deoxycytidine to decrease growth inhibition of CCRF-CEMcells induced by dFdCyd. Deoxycytidine concentrations: none (x); 100 jig/ml (O);30 fig/ml (•):10 jig/ml (A); 3 >ig/ml (D).

produces about a 1000-fold decrease in growth inhibitory activity. While a similar decrease of dFdCyd growth inhibition isobserved with deoxycytidine, about 3-fold higher concentrations of compound are required to achieve a comparable degreeof protection (Fig. 4). This decrease in activity is probably dueto competitive substrate interactions of the two compounds fordeoxycytidine kinase, the initial enzyme involved in the activation of both drugs (18, 19). The flow cytometric analysis inCEM cells (Fig. 5) shows a minimal effective concentration ofdFdCyd at a concentration of 0.02 /¿g/ml,whereas the minimaleffective concentration of ara-C is 10-fold higher at 0.2 pg/m\.The histograms clearly demonstrate that dFdCyd accumulatescells at the early S phase of the division cycle. Flow cytometricanalysis of dFdCyd and ara-C on L5178Y cells reveals greatersensitivity of Gj-S cells to dFdCyd (data not shown). A G.-Sblock in response to high concentrations of ara-C has beenreported (20, 21). We interpret this to reflect greater accumu-

4418

Research. on February 21, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

ANTITUMOR ACTIVITY OF dFdCyd

O"5

I

A = Control= dFdCO-.02/tg/ml

Relative Fluorescence

A = Control

B= Ara-C 0.2/ig/ml

B

Relative FluorescenceFig. 5. Flow cytometric analysis of DNA content of CEM cells treated with

dFdCyd (A) or ara-C (A); 16-h exposure.

dramatic differences in activity were observed with murine solidtumor models. As shown in Fig. 7, dFdCyd produced completeinhibition of tumor growth in the X-5563 plasma cell myelomafrom 5 up to 40 mg/kg/day. No toxicity or weight loss occurredin the mice treated with the dose of 40 mg/kg/day. Substantialactivity was also observed at 1.25 and 2.5 mg/kg/day as well.Since the maximum dose of dFdCyd on this intermittent schedule is in excess of 125 mg/kg/day, therapeutic activity can beobserved over a 2-log range of dFdCyd concentrations. Inmarked contrast, ara-C is completely inactive in this tumormodel (Fig. 7). In the C3H mice used for this study, a dose of40 mg/kg/day of ara-C administered daily for 10 days was toxicwith 7 of 10 mice dead by the end of the 2-week period.

Much higher doses of dFdCyd (160 mg/kg/day, every thirdday for 4 doses) were required before significant antitumoractivity (93% inhibition) was observed against B-16 melanoma(Fig. 7). No toxic deaths were observed with dFdCyd in thisexperiment. As expected, less than 50% inhibition of tumorgrowth was observed with ara-C at the maximum dose of 80mg/kg/day. Excellent antitumor activity was also observed fordFdCyd against the CA-755 adenocarcinoma with completeinhibition of tumor growth observed at 20, 40, and 80 mg/kg/day every third day for 4 doses (Fig. 7). No toxic deaths orweight losses were observed in the mice treated with dFdCyd.

160-1

140-

120-

100-

80-

60-

40-

20-

0-1.25 2.5 5.0 10 20 40 80

Dose (mg/kg)

Fig. 6. Comparison of antitumor activity of dFdCyd with ara-C against L1210leukemia in DBA/2 mice. dFdCyd (x) was dosed on days 1, 4, 7, and 10, whileara-C (O) was dosed daily for 10 days. Average survival for vehicle-treated controlswas 8.6 ±1.2 (SD) days. Survival (days) of drug-treated groups at maximumdoses were: dFdCyd, 22.7 ±1.2; ara-C, 19.6 ±1.3.

lation of the triphosphate of dFdCyd (21, 22).Simultaneous treatment of Chinese Hamster ovary or

L5178Y cells with dFdCyd and vinblastine results in decreasedaccumulation of G2+ M cells consistent with a specific block inS phase (21).

The initial in vivo evaluation of dFdCyd was conducted withL1210 leukemia. While activity can be observed on a dailyschedule of dFdCyd administration with this tumor model, aless chronic schedule of administration is more effective. Asshown in Fig. 6, excellent antitumor activity was observed whenthe compound was administered on days 1,4, 7, and 10. Onthis treatment schedule, dFdCyd was substantially more activethan ara-C administered on a daily schedule of administration.While 80/kg/day is the maximum tolerated dose of ara-C inthis experiment, the maximum tolerated dose of dFdCyd onthis staggered schedule of administration is above 125 mg/kg.The initial evaluation of the antitumor activity of dFdCyd useda maximum dose of 40 mg/kg due to a limited supply ofcompound. ara-C was somewhat more active against P-388leukemia (ILS, 111 %, 40 mg/kg/day) than dFdCyd (ILS, 93%,20 mg/kg/day), using the same treatment schedules.

While no significant differences in antitumor activity wereobserved between dFdCyd and ara-C in these two leukemias,

100-1

1.25 40 80 16010 20

Dose (mg/kg)

Fig. 7. Comparison of antitumor activity against three murine solid tumormodels. dFdCyd and ara-C against X-5563 myeloma in C3H mice. dFdCyd (•)was dosed on days 1, 4, 7, and 10, while ara-C (O) was dosed daily for 10 days.Tumor weight of vehicle-treated controls was 12.1 ±3.4 (SD) g. ara-C was toxicat 40 mg/kg with 7 of 10 mice dead by day 14. Tumor weight (g) of drug-treatedgroups at maximum doses were: dFdCyd, 0 ±0; ara-C, 9.3 ±4.7. dFdCyd andara-C against B-16 melanoma in C57BL/6 mice. dFdCyd (•)was dosed on days1, 4, 7, and 10, while ara-C (D) was dosed daily for 10 days starting on day 1.Tumor weight of vehicle-treated controls was 4.4 ±2.8 g. Tumor weight (g) ofdrug-treated groups at maximum doses were: dFdCyd, 0.1 ±0.1; ara-C, 2.2 ±0.7g. dFdCyd and ara-C against CA-755 adenocarcinoma in C57BL/6 mice. dFdCyd(A) was dosed on days 5, 8, 11, and 14, while ara-C (A) was dosed daily for 10days starting on day 5. Tumor weight of vehicle-treated controls was 6.8 ±2.0 g.Tumor weight (g) of drug-treated groups at maximum doses were: dFdCyd, O ±0; ara-C, 6.0 ±2.3.

4419

Research. on February 21, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

ANTITUMOR ACTIVITY OF dFdCyd

This solid tumor model was also refractory to ara-C.ara-C has good antitumor activity against M5076 with greater

than 95% inhibition of tumor growth observed at both 40 and80 mg/kg/day (Fig. 8). The C57BL/6 mice used in this studywere able to tolerate a dose of 80 mg/kg/day of ara-C in contrastto the C3H mice. A much broader therapeutic index wasobserved with dFdCyd in this tumor model as compared to ara-C with complete inhibition of tumor growth observed at 20 and40 mg/kg/day. While less activity was observed at the lowerdoses for dFdCyd, the slope of the dose-response curve was notas steep as that observed for ara-C (Fig. 8). Good antitumoractivity with dFdCyd was also observed against the 6C3HEDlymphosarcoma (95% inhibition, 20 mg/kg/day) and theP1534J leukemia (92% inhibition, 20 mg/kg/day), using anevery fourth day for 3 doses schedule.

The acute toxicity of dFdCyd was evaluated when administered i.v. to young adult male and female ICR mice. Eachanimal received a single dose of 500 mg/kg of body weight.There were no deaths in vehicle control or treated animals.Signs of toxicity in the dFdCyd-treated animals were limited toleg weakness in males and females which occurred immediatelyafter dosing and was not observed after 1 h. Hair loss in maleswas observed on test days 5 through 10. Mean body weightgains were depressed in males and females on test day 8 andwere normal on test day 15 relative to vehicle controls. NodFdCyd-related lesions were found at necropsy. The medianlethal dose of dFdCyd in 0.9% saline solution when administered i.v. to mice was estimated to be greater than 500 mg/kg.

In a toxicological evaluation of dFdCyd given i.p. to 15 maleand 15 female B6C3Fi mice per treatment group for 3 months,using doses of 1 mg/kg daily, 5 or 20 mg/kg twice a week, and40 mg/kg weekly, all but 7 of the mice survived the duration ofthe study and the deaths were not related to the action ofdFdCyd. There were no physical signs of toxicity or effects onbody weights. Hemoglobin values were markedly decreased inthe group given 1 mg/kg daily and only slightly reduced in theother treatment groups. Erythrocyte counts and packed cellvolumes followed the same trend. Leukocyte counts were markedly decreased in the group given 1 mg/kg daily and onlyslightly reduced in males of the other treatment groups. Spleenweights were markedly increased due to increased erythropoie-sis in the group given 1 mg/kg daily. Spleen weights anderythropoiesis were changed to a lesser degree in both sexesgiven 20 mg/kg twice a week and males given 40 mg/kg once aweek. The splenic erythropoiesis was in response to the decreasein erythrocytes. The spleen of mice is a major organ of erythropoiesis. The testes weights were decreased in all treatmentgroups due to hypospermatogenesis. The testes effect was most

100n

I 60-

40-

20-

O-'r

1.25 40 805.0 10 20

Dose (mg/kg)Fig. 8. Comparison of antitumor activity of dFdCyd with ara-C using M-5076

in C57BL/6 mice. dFdCyd (x) was dosed on days 5, 8, 11, and 14, while ara-C(O) was dosed daily for 10 days starting on day 5. Tumor weight of vehicle-treatedcontrols was 5.0 ±1.1 (SD) g. Tumor weight (g) of drug-treated groups atmaximum doses were: dFdCyd, 0.1 ±0.3; ara-C, 0.4 ±0.4.

severe in the group given 1 mg/kg daily. There were no dFdCyd-related changes in the clinical chemistry evaluations. Otherthan changes noted in the spleen and testes, there were no otherpathological findings related to the administration of dFdCyd.The minimal toxic dose level was 40 mg/kg given once a weekfor 3 months or 120 mg/m2 body surface area.

DISCUSSION

dFdCyd (Gemcitabine, LY188011) is a new and unique de-oxycytidine analogue. The compound was initially synthesizedas a potential antiviral drug and had excellent activity againstboth RNA and DNA viruses using cell culture assays. AlthoughdFdCyd demonstrated low visual cytotoxicity on preformedmonolayer cells in cell culture assays, the in vivo therapeuticindex, using daily administration of dFdCyd, was insufficientfor further development of this compound as a useful antiviraldrug (11).

dFdCyd was also evaluated for anticancer activity using cellculture assays and in vivo murine tumor models. dFdCyd is avery potent and quite specific deoxycytidine analogue with 1ng/ml inhibiting growth of CCRF-CEM human leukemia cellsby 50%. Concurrent addition of deoxycytidine to the cell culturesystem provides about a 1000-fold decrease in biological activity. Other difluoropyrimidine analogues were dramatically lesspotent (17).

While a daily schedule of administration produced minimalantitumor activity, a staggered schedule was more rewarding.Excellent antitumor activity and reduced toxicity were observedwhen the compound was administered on an every third dayschedule for four doses. Both acute and subchronic toxicologicalevaluation of dFdCyd demonstrated minimal toxicity to mice.A decrease in RBC and WBC parameters were observed; however, no histopathological effects were seen in any tissues excepttestes. Dose schedule was important in defining toxicity. Smalldose levels of 1 mg/kg given daily was more toxic than 40 mg/kg given once a week.

ara-C is also a deoxycytidine analogue with proven clinicalactivity. The compound is most useful for the treatment ofadult acute leukemia and has utility in other hematologicalmalignancies. Although little or no useful activity has beenobserved against human solid tumors, ara-C has excellent antitumor activity against both leukemias L1210 and P-388. ara-C demonstrates limited antitumor activity against murine solidtumor models comparable to the clinical experience. In markedcontrast, dFdCyd has broad spectrum antitumor activity. Thecompound was active in 6 of 6 murine solid tumor models and2 of 2 murine leukemia models. dFdCyd also has broad spectrum antitumor activity against 6 human carcinoma xenografttumor models.3 The compound also displayed a very broad

therapeutic index in many of these tumor models. Preclinicaltest results support the proposal that dFdCyd may prove ofclinical utility for the treatment of both leukemias and manyhuman carcinomas.

Cell cycle kinetic studies with dFdCyd (Fig. 5) clearly showspecificity for proliferation in the S phase of the cell cycle withno effect on progress through early G,, G2, or M phases of thecell cycle. Time lapse video microscopy of Chinese hamsterovary cells following 16-h exposure to dFdCyd indicates apermanent block in S phase (Fig. 9). This results in enlargementof the cells, indicating an unbalanced growth as there is noprogression into mitosis. Unbalanced growth is a phenomenon

3G. B. Grindey, unpublished results.

4420

Research. on February 21, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

ANTITUMOR ACTIVITY OF dFdCyd

associated with the prolonged suppression of macromolecularsynthesis. The unbalanced growth due to ara-C treatment hasbeen reported to be due to inhibition of DNA synthesis withoutconcomitant inhibition of RNA or protein synthesis, resultingin a marked increase in cell volume and subsequent cell death(20, 21, 23). Similar data including reduced number of mitoticcells and enlarged cells (Fig. 9) coupled to previous data (22)showing inhibition of DNA synthesis without affecting proteinand RNA synthesis support unbalanced growth in response todFdCyd treatment.

Extensive cellular pharmacology studies of dFdCyd have beenconducted to provide an explanation for this marked improvement in antitumor efficacy (19, 24, 25). dFdCyd is convertedto the triphosphate metabolite (dFdCTP) in a manner similarto the metabolism of ara-C. In several different cell types theaccumulation of dFdCTP was more rapid and greatly exceededthe concentrations of ara-CTP achieved under similar conditions (25). Moreover, the tumor cells were able to eliminate theara-CTP much more rapidly than the dFdCTP (25). The pharmacological mechanism of this improved anticancer activityappears related to the long retention of dFdCTP in tumortissue.

The clearance of dFdCTP from CCRF-CEM cells is linear(tv, = 3.3 h) at low intracellular concentrations, while a biphasicelimination was observed at higher dFdCTP levels (100-600/XM).Almost complete inhibition of dFdCTP elimination wasobserved at 300 fiM (22). dFdCyd also induces a pronouncedreduction in dCTP pools. This reduction in dCTP coupled withhigh intracellular concentrations of dFdCTP account for theprolonged inhibition of DNA synthesis and resultant cellular

Fig. 9. Effect of dFdCyd on morphology of CHO cells: control (A); 30 ng/mldFdCyd for 16 h (B). Phase contrast, x 188.

cytotoxicity observed (22). This may explain the potent toxicityobserved on a daily schedule of administration and the unusualschedule dependency observed with dFdCyd. This long half-lifeof dFdCTP may also be responsible for the observed efficacy inthe slow-growing murine and human carcinoma tumor models.

Clinical importance of the accumulation and retention of ara-CTP has been established (26-28). The therapeutic effect ofara-C was significantly better in leukemia blasts that had longretention times of intracellular ara-CTP in vitro (27) and duringtherapy (28). The activity demonstrated by high-dose ara-C inleukemic patients refractory to ara-C suggests that resistancemay be overcome by increased intracellular triphosphate levels(29). Taken together, the higher concentrations and slowerelimination of intracellular dFdCTP appear important in determining the superior antitumor activity observed with dFdCydas compared to ara-C.

In conclusion, dFdCyd has broad spectrum antitumor activityagainst murine leukemias, murine solid tumors, and humantumor xenografts. The cellular pharmacology studies clearlyindicate that the compound has unique properties for an anti-metabolite. The compound is cell cycle specific, blocking thecells at the S and late G, phase and is retained in human tumorcells for long periods. These characteristics may be responsiblefor the observed efficacy in the slow-growing murine and humancarcinoma tumor models. Such results support the conclusionthat dFdCyd is an excellent candidate for clinical trials in thetreatment of cancer.

REFERENCES

1. DeClercq, E., Descamps, J., Verhelst, G., Walker, R. T., Jones, A. S.,Torrence, P. F., and Shugar, D. J. Comparative efficacy of antiherpes drugsagainst different strains of herpes simplex virus. Infect. Dis., 141: 563-574,1980.

2. DeClercq, E. Antiviral activity of 5-substituted pyrimidine nucleoside analogues. Pure Appi. Chem., 55: 623-636, 1983.

3. Ronis, H. E., and Johnson, H. G. Inhibition of plaque formation of vacciniavirus by cytosine arabinoside hydrochloride. Bacterio!. Proc., 45:140, 1962.

4. Underwood, G. E. Activity of l-/3-D-arabinofuranosylcytosine hydrochlorideagainst herpes simplex keratins. Proc. Soc. Expl. Biol. Med., Ill: 660, 1962.

5. Kaufman, H. E., and Maloney, E. O. IDU and cytosine arabinoside inexperimental herpetic keratitis. Arch. Ophthalmol., 69:626, 1963.

6. Livingston, R. B., and Carter, S. K. Cytosine arabinoside (NSC-63878)-Clinical brochure. Cancer Chemother. Rep., Part 3, Vol. 1, No. 1, 179-205,1968.

7. Keating, M. J., McCredie, K. B., Bodey, G. P., Smith, T. L., Gehan, E., andFreireich, E. J. Improved prospects for long-term survival in adults withacute myelogenous leukemia. JAMA, 248: 2481-2486, 1982.

8. Watanabe, K. A., Reichman, V., Hirota, K., Lopez, C, and Fox, J. J.Nucleosides. 110. Synthesis and antiherpes virus activity of some 2'-fluoro-2'-deoxy arabinofuranosylpyrimidine nucleosides. J. Med. Chem., 22: 21-

24, 1979.9. Chwang, T. L., Williams, R. D., and Schieber, J. E. Synthesis of 2'-O-nitro-

9-ß-D-arabinofuranosyladenine and 2'-O-nitro-9-/3-D-arabinofuranosylhy-poxanthinc. Tetrahedron Lett., 24: 3183-3186, 1983.

10. Hertel, L. W., Kroin, J. S., Misner, J. W., and Tustin, J. M. Synthesis of 2-deoxy-2,2-difluoro-D-ribose and 2-deoxy-2,2-difluoro-D-ribofuranosyl nucleosides. J. Org. Chem., S3: 2406-2409, 1988.

11. DeLong, D. C., Hertel, L. W., Tang, J., Kroin, J. S., Wilson, J. D., Terry,J., and Lavender, J. F. Antiviral activity of 2',2'-difluorodeoxycytidine.Abstracts of Meeting of American Society of Microbiology, March 24-28,Washington, DC, 1986.

12. Foley, G. E., Lazarus, H., Farber, S., Uzman, B. G., Boone, B. A., andMcCarthy, R. E. Continuous culture of human lymphoblasts from peripheralblood of a child with acute leukemia. Cancer (Phila.), 18: 522-529, 1965.

13. Grindey, G. B., Wang, M. C., and Kinahan, J. J. Thymidine inducedperturbations in ribonucleoside and deoxyribonucleoside triphosphate poolsin human leukemic CCRF-CEM cells. Mol. Pharmacol., 16:601-606, 1979.

14. Krishan, A. Rapid flow cytofluorometric analysis of mammalian cell cycle bypropidium iodide staining. J. Cell Biol., 66: 188-193, 1975.

15. Geran, R. I., Greenberg, N. H., MacDonald, M. M., Schumacher, A. M.,and Abbott, B. J. Protocols for screening chemical agents and naturalproducts against animal tumors and other biological systems (Third Edition).Cancer Chemother. Rep. 3, (Part 3), 1, 1972.

16. Green, E. L. (ed.), Handbook on Genetically Standardized Jax Mice. BarHarbor, ME: The Bar Harbor Times Publishing Co., 1968.

4421

Research. on February 21, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

ANTITUMOR ACTIVITY OF dFdCyd

17. Grindey, G. B., Boder, G. B., Hertel, L. W., Kroin, J. S., Poore, G. A.,Rinzel, S. M., and Todd, G. C. Antitumor activity of 2',2'-difluorodeoxycy-

tidine (LYI88011). Proc. Am. Assoc. Cancer Res.. 27: 296, 1986.18. Gandhi, V. V., Nowak, B., and Plunkett, W. Regulation of deoxycytidine

kinase (dCK) in K562 cells. Proc. Am. Assoc. Cancer Res., 28: 408, 1987.19. Chubb, S., Heinemann, V., Novotny, L., Hertel, L. W., Nowak, B., Mineishi,

S., Grindey, G. B., and Plunkett, W. Metabolism and action of 2',2'-

difluorodeoxycytidine (dFdC) in human leukemia cells. Proc. Am. Assoc.Cancer Res., 28: 324, 1987.

20. Jackson, R. C. Biological effects of folie acid antagonists with antineoplasticactivity, in: L. A. Dethlefsen (ed.). Cell Cycle Effects of Drugs. New York:Pergamon Press, 1986.

21. Pallavicini, M. G. Cytosine arabinoside: molecular, pharmacokinetic andcytokinetic considerations. In: L. A. Dethlefsen (ed.), Cell Cycle Effect ofDrugs, pp. 136-162. New York: Pergamon Press, 1986.

22. Plunkett, W., Gandhi, V., Chubb, S., Nowak, B., Heinemann, V., Mineishi,S., Sen, A., Hertel, L. W., and Grindey, G. 2',2'-Difluorodeoxycytidine

metabolism and mechanism of action in human leukemia cells. NucleosidesNucleotides, 8: 775-785, 1989.

23. Darzynkiewicz, Z. Cell growth and division cycle. In: L. A. Dethlefsen (ed.),

Cell Cycle Effect of Drugs, pp. 1-30. New York: Pergamon Press, 1986.24. Heinemann, V., Hertel, L. W., Grindey, G. B., and Plunkett, W. Comparison

of the cellular pharmacokinetics and toxicity of 2',2'-difluorodeoxycytidine(dFdC) and Arabinosylcytosine (Ara-C). Proc. Am. Assoc. Cancer Res., 28:324, 1987.

25. Heinemann, V., Hertel. L. W., Grindey, G. B.. and Plunkett, W. Comparisonof the cellular pharmacokinetics and toxicity of 2',2'-difluorodeoxycytidineand 1-if-D-arabinofuranosylcytosine. Cancer Res., 48: 4024-4031, 1988.

26. Rustum, Y. M., Henderson, E. S., and Grindey, G. B. Biochemical subclassifications of acute myelocytic leukemia (AML) cells. Proc. Am. Assoc.Cancer Res., 17: 128, 1976.

27. Rustum, Y. M., and Preisler, H. D. Correlation between leukemic cellretention of l-/i-D-arabinofuranosylcytosine-5'-triphosphate and response totherapy. Cancer Res., 39:42-49, 1979.

28. Plunkett, W., lacoboni, S., Estey, E., Liliemark. J. O.. and Keating. M. J.Pharmacologically directed Ara-C therapy for refractory leukemia. Semin.Oncol., 12 (Suppl. 3): 20-30, 1985.

29. Herzig, R. H.. Herzig, G. P., Lazarus, H. N., Wolff, S. N., and Phillips, G.L. Successful treatment of patients with refractory acute non-lymphocyticleukemia (ANLL) using high-dose cytosine arabinoside (HD Ara-C) with andwithout anthracycline. Blood, 58: 14Ia. 1981.

4422

Research. on February 21, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from

1990;50:4417-4422. Cancer Res   Larry W. Hertel, George B. Boder, J. Stan Kroin, et al.  

-deoxycytidine)′-Difluoro-2′,2′Evaluation of the Antitumor Activity of Gemcitabine (2

  Updated version

  http://cancerres.aacrjournals.org/content/50/14/4417

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/50/14/4417To request permission to re-use all or part of this article, use this link

Research. on February 21, 2020. © 1990 American Association for Cancercancerres.aacrjournals.org Downloaded from