[CANCER RESEARCH 42, 2081-2086, May 1982]0008-5472/82/0042-0000$02.00

Optimal Scheduling of Methotrexate and 5-Fluorouracil in Human BreastCancer1

Chris Benz, Tina Tillis, Ellen Tattelman, and Ed Cadman2

Departments of Medicine and Pharmacology, YaleSchool of Medicine, New Haven, Connecticut 06510

ABSTRACT

We have shown previously that methotrexate pretreatmentof murine leukemia and human colon carcinoma cell culturesresults in augmented intracellular accumulation of 5-fluorour-

acil metabolites. Both of these drugs are commonly used forthe treatment of women with breast cancer; thus, sequencingof methotrexate before 5-fluorouracil was evaluated in vitrousing a human mammary carcinoma cell line, 47-DN. Intracellular 5-fluorouracil accumulation was maximally increased 4-

fold in cultures pretreated with 10 /ÕMmethotrexate for 24 hr.This enhancement of 5-fluorouracil metabolism was associatedwith increased intracellular levels of 5-phosphoribosyl 1-pyro-

phosphate, resulting from the antipurine effect of methotrexate.Brief exposure to exogenous hypoxanthine at physiologicalconcentrations reversed the biochemical synergism betweenmethotrexate and 5-fluorouracil. Other antimetabolites associated with elevations of 5-phosphoribosyl 1-pyrophosphate enhanced intracellular accumulation of 5-fluorouracil up to 2.5-

fold. In cloning assays, 18 hr of methotrexate pretreatmentfollowed by 5-fluorouracil resulted in optimal synergistic cyto-

toxicity, which could be prevented if high concentrations ofleucovorin were given between methotrexate and 5-fluorouracil

administration.Since these results indicated that optimal breast tumor tox-

icity in vitro was achieved by 18- to 24-hr sequencing ofmethotrexate and 5-fluorouracil, a clinical toxicity study was

carried out to assess whether this drug schedule could betolerated. Seven patients with advanced cancer were treatedwith 21 courses of sequential therapy. No toxicity occurredwith 38% of treatment courses; mild to moderate leukopeniaand mucositis occurred with 29 and 38% of courses, respectively. Toxicity was related to retreatment interval and notcumulative drug dose or elevated serum methotrexate levels.These clinical results suggest that Phase II studies evaluating24-hr-sequenced methotrexate and 5-fluorouracil in breast

cancer are warranted.

INTRODUCTION

Studies of adjuvant chemotherapy in postmastectomy patients with positive axillary nodes suggest that combinationdrug therapy can reduce the rate of recurrence by 20 to 40%and significantly increase survival for women with operablebreast cancer (10, 18, 25). Despite these optimistic projections, present incidence rates show that one of every 13 womenwill develop breast cancer, and more than one-half will even-

1Supported by Grants CA-24187, CA-27130, and CA-08341 from the National Cancer Institute and Grant CH-145 from the American Cancer Society.

2 Recipient of a Cancer Research Award from the American Cancer Society.To whom requests for reprints should be addressed.

Received September 29, 1981; accepted January 29, 1982.

tually die of disseminated disease (11, 25). The use of combination chemotherapy in disseminated breast cancer has improved objective response rates over single-agent therapy by

about 40%, yet there has been no improvement in overallsurvival and no clear superiority in palliative benefit over theuse of sequential single-agent therapy (9, 18). These discour

aging facts reflect the brevity of response durations followingsystemic chemotherapy, 7 to 11 months, and low completeresponse rates of about 15% (7). Perhaps more knowledgeablescheduling of multiple drugs given in combination will improveour therapeutic impact on breast cancer; basic laboratorystudies may be able to provide the necessary rationale fordevising such synergistic combinations.

MTX3 and FUra are 2 of the most commonly used drugs in

the treatment of breast cancer. These 2 antimetabolites areusually administered simultaneously and often in combinationwith cyclophosphamide and prednisone, both in the adjuvantsetting and as treatment for disseminated disease (18). Ourlaboratory's biochemical studies in mouse leukemia and human

colon carcinoma cell lines have shown that the antitumor effectof combined MTX and FUra is synergistically enhanced whenMTX precedes FUra administration in vitro (1, 5, 6). By inhibiting de novo purine synthesis and elevating intracellular poolsof PRPP, MTX enhances the intracellular accumulation andmetabolism of FUra, resulting in synergistic tumor cell kill. Thissynergism does not occur when MTX and FUra are givensimultaneously or when FUra treatment precedes MTX. Furthermore, results in the human colon carcinoma cell line suggest that the MTX pretreatment interval for optimal synergismwith FUra is dependent on cellular growth rates (1). Thesestudies have now been extended to a hormone-dependenthuman breast carcinoma cell line, 47-DN, and applied in thedesign of a clinical trial assessing toxicity to optimally se-quenced MTX and FUra.

MATERIALS AND METHODS

Cell Line, Drug, and Clonal Growth Assay. The hormone-dependenthuman mammary carcinoma, 47-DN, is a well-characterized (19) con

tinuously growing monolayer cell line that under present culture conditions doubles in 30 hr. Cells were grown in Roswell Park MemorialInstitute Tissue Culture Medium 1640 supplemented with 10% of fetalcalf serunrneonatal calf serum (1:1) (Grand Island Biological Co.,Grand Island, N. Y.). For indicated experiments, cells were grown inmedium containing dialyzed fetal calf serum (Grand Island BiologicalCo.). Maximal 47-DN growth rate was obtained when 1 nw estradiol

(Sigma Chemical Co., St. Louis, Mo.) and 0.2 IU insulin (Eli Lilly andCo., Indianapolis, Ind.) per ml were added to the culture medium.Stocks were passaged weekly, and single-cell suspensions were pre

pared using a trypsin (0.05%):EDTA (0.02%) solution. Stock cultures

3The abbreviations used are: MTX. methotrexate; FUra, 5-fluorouracil; PRPP,5-phosphoribosyl 1-pyrophosphate; LV, leucovorin (5-formyltetrahydrofolate).

MAY 1982 2081

Research. on August 11, 2021. © 1982 American Association for Cancercancerres.aacrjournals.org Downloaded from

C. Benz et al.

and cloning studies were grown in 75-sq cm sterile plastic culture

flasks (Costar, Data Packaging, Cambridge, Mass.) with 25 ml ofmedium in 5% CO2 incubators at 37°.Cell counts were performed on

a Model ZBI Coulter Counter (Coulter Electronics, Inc., Hialeah, Fla.);clonal growth and drug sensitivity assays were performed using amonolayer technique and an automated colony counter (Biotran II; NewBrunswick Scientific Co., Inc., Edison, N. J.) as described in detailelsewhere (1, 2). All drugs were purchased from Sigma with theexception of [6-3H]FUra (20 Ci/mmol) which was obtained from Mo-

ravek Biochemicals (City of Industry, Calif.), and 6-diazo-5-oxo-i_-nor-leucine and u-alanosine, which were obtained from the Division of

Cancer Treatment, National Cancer Institute (Bethesda, Md.).Biochemical Assays. The intracellular accumulation of FUra was

measured by our previously described microfuge method (1). Briefly,cell cultures were exposed to 100 JIM [6-3H]FUra and at the indicated

times harvested by rapid trypsinization (0.05% solution) and counted.A 0.1-ml suspension of these cells was placed in a 0.5-ml plastic

microfuge tube and immediately centrifuged at 10,000 rpm for 15 secto sediment the cells through a silicone-oil interphase into 0.04 ml of5% perchloric acid. The drug-containing medium remained above thissilicone-oil interphase. Triplicate samples in microfuge tubes were thenfrozen quickly in an ethylene glycol:dry ice bath and cut at the inter-

phases of perchloric acid, oil, and medium. The radioactivity containedin the perchloric tip represented the total amount of intracellular FUraand metabolities accumulated, recorded as nmol FUra per 106 cells.

The oil fraction contained no radioactivity; the third fraction was themedium. The amount of radiolabeled FUra incorporated into cellularRNA was quantitated by standard alkaline hydrolysis of perchlorate-insoluble cell precipitates; and soluble mono-, di-, and triphosphateribonucleotides of FUra were quantitated by high-pressure liquid chro-

matography (Partisil SAX column with a linear phosphate buffer gradient), as described previously (1 ).

Intracellular pools of PRPP in control and MTX-treated 47-DN cells

were quantitated by a standard enzymatic assay using the enzymeadenine phosphoribosyltransferase which was isolated from suspension cultures of L1210 (1 ). Lysed cell extracts of 47-DN cells containingmeasurable amounts of PRPP were incubated with [3H]adenine (Sigma)and the isolated L1210 enzyme. The amount of [3H]AMP formed, which

was dependent upon the available PRPP (which provided the phos-

phorylated ribose for conversion of adenine to AMP), was quantitatedby performing simultaneous control assays with known quantities ofPRPP, and values were reported as ng PRPP per 106 cells.

Clinical Study. A limited study to determine toxicity of 24-hr-se-

quenced administration of MTX and FUra was carried out in 7 patientswith advanced cancer (Table 5). Schedule and dose of drugs for the21 courses of therapy was: P.O. MTX (50 mg/sq m) every 6 hr 5 times;i.v. bolus FUra (600 mg/sq m) 1 hr after fifth dose of MTX; and P.O. LV(10 mg/sq m) every 6 hr 6 times beginning 6 hr after the fifth MTXdose. Patients receiving more than one course were initially retreatedat 7 to 14 days. If toxicity developed, the retreatment interval and notdrug dosage was altered. All patients had received prior treatment noless than 3 weeks before study entry, and 6 of the 7 had previouslyreceived 1-hr-sequenced MTX-FUra as part of an earlier study (13).

No patient was entered with preexisting neutropenia or creatinineclearance <65 ml/min. Toxicity and performance status were evaluated by the Eastern Cooperative Oncology Group criteria (24). SerumMTX levels were measured by the radioimmunoassay technique duringall 21 courses of therapy.

Statistics. All experimentally derived charts and tables representmean values of triplicate samples in single experiments, with S.D. <5%

or as indicated. All experiments were repeated at least once for validity.Calculations were performed on a Hewlett-Packard 67 programmable

calculator.

RESULTS

Clonal growth of 47-DN shows dose-dependent sensitivity to

6 hr of FUra and 24 hr of MTX (Chart 1) over the indicatedconcentration ranges. When dialyzed serum is used in theculture medium (free of hypoxanthine and thymidine), FUra andMTX toxicity are enhanced by approximately 20 and 50%,respectively. Since 100 /¿MFUra is both therapeutically achievable (20) and necessary to completely inhibit DNA synthesis in47-DN (measured by incorporation of [3H]deoxyuridine into

DNA), this concentration was used in all biochemical assays.Although 100 /IM FUra resulted in virtually complete growthinhibition of 47-DN by 6 hr, Chart 2 shows that FUra at thisconcentration accumulated intracellularly in a near-linear fash-

3 "o -

Chart 1. Clonal growth of treated human mammary carcinoma, 47-DN. Duplicate cultures were seeded with 5 x 104 cells as described (2), and the mean

colony counts were recorded. The effects of FUra and MTX on the number ofmonolayer colonies were measured and expressed as a percentage of controlgrowth. Bars, range of values from multiple experiments.

3.0

u>O 2.0

I.O

800

600 tt

o

§400 ^

200

O IO 20 30

Time (hr)

Chart 2. Intracellular accumulation (O) and incorporation into RNA (•)of[3H]FUra into human breast cancer cells, 47-DN, exposed to 100 JIM FUra at

Time 0.

2082 CANCER RESEARCH VOL. 42

Research. on August 11, 2021. © 1982 American Association for Cancercancerres.aacrjournals.org Downloaded from

Optimal Scheduling of MTX and FUra in Human Breast Cancer

¡onfor over 30 hr in these cells as described in other humanmonolayer cultures (1). This intracellular accumulation wasalso accompanied by a proportionate increase in FUra incorporation into cellular RNA. After 6 hr of exposure to 100 /IMFUra, the total amount of FUra metabolites concentrated intra-cellularly is greater than 20-fold that of FUra in the medium,

suggesting that these biochemical results are not aberrantconsequences of dying cells. For convenience in subsequentexperiments, the 6-hr time point was used as a measure of the

rate of intracellular FUra accumulation.The effect of MTX on the rate of intracellular FUra accumu

lation is represented in Chart 3A. A 24-hr exposure to 10 fiM

MTX resulted in the greatest enhancement of intracellular FUraaccumulation. Longer and shorter MTX pretreatment periodswere less effective. This enhancement of FUra accumulationwas also dependent on MTX concentration with maximumenhancement observed after 10 JUMMTX, a dose which completely inhibits clonal growth after 24 hr (Chart 1). The enhancement of intracellular FUra accumulation following MTXexposure was observed at concentrations of FUra ranging from1 tO 100/iM.

Biochemical studies in L1210 and HCT-8 cells have shown

that enhanced FUra accumulation is associated with increasedamounts of intracellular PRPP, which are the result of de novopurine synthesis inhibition by MTX (1, 5, 6). In these cell lines,>90% of the accumulated FUra is present as ribonucleotides.MTX-pretreated 47-DN cells also had increased PRPP levels,

which were maximal by 24 hr and paralleled the increasedrates of intracellular FUra accumulation (Chart 36). In bothcontrol and MTX-pretreated 47-DN, 70 to 75% of accumulatedFUra was present as soluble 5-fluorouridine triphosphate and>98% as total FUra-containing ribonucleotides; MTX-pretreated cells increased incorporation of FUra into RNA by 2- to3-fold. Other drugs, known to inhibit de novo purine synthesisand increase intracellular pools of PRPP in L1210 and HCT-8cells (1, 4), also enhanced FUra accumulation in 47-DN cellsfrom 1.3- to 2.5-fold (Table 1).

Hypoxanthine is capable of consuming intracellular pools ofPRPP and reducing FUra accumulation in MTX-pretreatedL1210 cells (5). It is normally present in both human and bovineserum (16) and can be removed by dialysis. When 47-DN cellsare grown in hypoxanthine-free medium, optimal enhancement

A. [3HJFUro AccumulationB Intracellular PRPP

11102 12 24

Time (hr)

m

4000

0 3 18 24

Time (hr)

Chart 3. The effect of MTX exposure on intracellular FUra accumulation andPRPP levels of human breast cancer cells, 47-DN. Monolayer cell cultures wereexposed to 10 /IM MTX for the indicated time periods before adding 100 ¡ÕM16-3H]FUra (20 Ci/mmol) and determining the intracellular FUra accumulation. The

intracellular PRPP levels were also measured at the indicated times.

of FUra accumulation occurs after 24-hr pretreatment with 10

JIMMTX (Table 2). Adding hypoxanthine back to cells optimallypretreated with MTX totally reversed this augmentation in FUraaccumulation (Table 3).

Cloning studies were performed to demonstrate synergisticgrowth inhibition in 47-DN cells treated sequentially by MTX

and FUra. In studies describing our monolayer cloning technique, we showed that HCT-8 and 47-DN cells required greater

than 6 and 12 hr of MTX pretreatment, respectively, prior toFUra exposure to produce greater than additive cytotoxicity(2). These results were reproduced in 47-DN cells cultures

using several concentrations of MTX and FUra, chosen tomimic possible clinical use of these agents. Table 4 illustratesthis syngergistic cytotoxicity, as well as the influence of LV onsequenced MTX-FUra synergism in 47-DN cells. MTX was

present in the cell cultures for 24 hr, and FUra was addedduring each of four 6-hr intervals during the MTX exposure.

Greater than additive cytotoxicity was observed only whenFUra was added after 18 hr of exposure to MTX. When LV wasadded during exposure to MTX, MTX-FUra synergism was

observed only when the LV followed administration of FUra. LVat 100 U.Mtotally reversed the inhibition of thymidylate synthesisby 0.1 ¡IMMTX (as measured by incorporation of [3H]deoxyu-

ridine into DNA) but did not prevent the biochemical or growth-inhibitory effects of higher concentrations of MTX.

Table 1

Enhancement of intracellular accumulation of FUra (100fnm)inpretreated 4 7-DN

Pretreatment (24 hr) % of control*

Methylmercaptopurine riboside (0.1 UM)AzaserinedOfiM)6-Diazc-5-oxc-L-norleucine (10 MM)L-Alanosine (10 KM)

178195246136

Control = 2.26 ±0.61 (S.D.) nmol/10e cells/6 hr.

Table 2

Enhancement of intracellular accumulation of FUra (100 pu) inpretreated 47-DN

Cultures were grown in Roswell Park Memorial Institute Tissue Culture Medium1640 supplemented with 10% dialyzed fetal calf serum, insulin (0.2 ID/ml), andestradici (1 UM)

MTX concentration(MM)0.01

0.101.00

10.0%

of control8 at following pretreatmentInterval1

2 hr 24hr93

100 149150 209178 249

a Control = 2.34 nmol/10' cells/6 hr.

Table 3

Effect of hypoxanthine on intracellular accumulation of FUra inpretreated 4 7-DN

Cultures were grown in Roswell Park Memorial Institute Tissue Culture Medium1640 supplemented with 10% dialyzed fetal calf serum, insulin (0.2 III/ml), andestradici (1 nw)

Pretreatment(JIM)Hypoxanthine,

3 ru

mo10.0

1.00.10MTX,

24hr0

1010

1010Accumulation

(% of control0)55

63214425350

' Control = 1.43 nmol/10" cells/6 hr.

MAY 1982 2083

Research. on August 11, 2021. © 1982 American Association for Cancercancerres.aacrjournals.org Downloaded from

C. Benz et al.

Seven patients were treated with a schedule of MTX, FUra,and LV (see "Materials and Methods") designed according to

the sequencing schedule that produced optimal tumor toxicityin vitro. In addition, drug doses were chosen which could beadministered conveniently and would achieve cytotoxic serumconcentrations. The toxicity experienced from 24-hr-se-quenced MTX and FUra is shown in Table 5. There were noepisodes of thrombocytopenia or deterioration in renal function. Four of 7 patients tolerated more than 2 courses at anaverage treatment interval of 23 days. Two of 21 courses wereassociated with severe toxicity (Eastern Cooperative OncologyGroup Grades III to IV), and this occurred with the secondtreatment course in both patients. Mild to moderate toxicity(Eastern Cooperative Oncology Group Grades I to II) occurredin 11 courses, usually after the first or second treatment, and8 treatment courses were associated with no toxicity whatsoever. Mean serum MTX levels 1 hr prior to and 1 hr after thefifth dose of MTX were 1.07 ±0.74 S.D. and 2.10 ±0.92¡IM,respectively. Occurrence of toxicity was unrelated to eithercumulative drug doses or greater than average serum MTXlevels.

Table 4Effect of MTX, FUra, and LV on donai growth of 4 7-DN

% of control growth

TreatmentMTX,

0.1 fiM, 24hrLV.100/iM,6hrFUra,

6hrMTX-.FUra61st2nd3rd4thMTX

-»LV +FUracBefore

FUraWithFUraAfter

FUra1

JIMFUra90±9a1

00 ±1090±990

±960±673±742

±479

±869±649±5lOjiMFUra24

±324

±510±120

* Mean ±S.D.b FUra added during each of 4 6-hr intervals during a 24-hr 0.1 UM MTX

exposure.c LV (100 /IM, 6 hr) added before, with, or after FUra. The FUra was given

only during the fourth interval of MTX exposure in the experiment with LV.

DISCUSSION

Chart 4 schematically illustrates the mechanism of enhancedFUra accumulation following MTX pretreatment. MTX inhibitsdihydrofolate reducíase preventing regeneration of the reduced folate (tetrahydrofolate) pool necessary for one-carbontransfer in thymidylate synthesis and in de novo purine synthe-

de novo Pyrimidine

Synthesis!

PRPPOÎofot«P»01«®Furo/^^

^J^»PUMP!FUDP

»FdUMPOMP.I

->/FUMP

| dUMP ¿r*1

RNA / X/JUDP

^dUDP (i\Jf1*3/tj^vx'uRNAJHP

dTMP1dTTPde

novo Purine

Synthesis^^©iij^**^"^

""^xlf?!,

^^^~~"~- . ^

41K)!s

11ö ^x^^b^iiIMPAMP

GMP1

1AOPGOP/

' 1\P--»dCTP«-DNA /l\ATP

' .» CdATP dGTP GTP

RNA DNA RNA

Chart 4. Proposed interaction of MTX and FUra. Broken arrows, multipleenzymatic steps. Enzymes (circled numbers): 1, amidophosphoribosyltransfer-ase; 2, phosphoribosyl glycineamide formyltransferase; 3, phosphoribosyl ami-noimidazole carboxamide formyltransferase; 4, thymidylate synthetase; 5, dihy-drofolate reducíase;6, orotate phosphoribosyttransferase; and 7,phosphoribosylpyrophosphate synthetase. MTX inhibits Enzyme 5, and dTMP synthesis continues until the tetrahydrofolate pools no longer support the methyl transfer todUMP. Because of this reduction in tetrahydrofolate pools, purine synthesis isalso inhibited. 5-Fluorodeoxyuridine monophosphate directly inhibits Enzyme 4in the presence of tetrahydrofolate. FUMP, 5-fluorouridine monophosphate;FUDP, 5-fluorouridine diphosphate; FdUMP, 5-fluorodeoxyuridine monophos-phate; FUTP, 5-fluorouridine triphosphate; OMP, orotidine monophosphate; FH4.tetrahydrofolate; FH¡,dihydrofolate.

Table 5Clinical toxicity sfudy of 24-hr sequence^ MTX and FUra with LV rescue

PatientToxicitygradea/courseance

sta-PatientPrimarycancerBreastBreastBreastBreastMycosis

fungoidesMycosisfungoidesHead

and necktus"

courses1221124Total

courses462323121LeukopeniaII.

0, 0,00,IV, I, 0, 0,I0,

III0,0,0I,

III,0,00Nausea-Emesis0,

0, 0,0II,I, 0. 0, 0,00,00,0,0II,

00,0,00Diarrhea0,

0, 0,00,I, 0, I, 0,00,01,0,00.00,0,00Mucositis0,

0, 0,00,II, 1, II, 0,00,0II.

1.0II.0II.

II,00

No. of courses

Grade8Leukopenia0

13I 4II 2III 1IV 1Nausea-Emesis18

1200Diarrhea18

3000Mucositis13

2600

Eastern Cooperative Oncology Group criteria (24).

2084 CANCER RESEARCH VOL. 42

Research. on August 11, 2021. © 1982 American Association for Cancercancerres.aacrjournals.org Downloaded from

sis. The consequence of the reduction in de novo purinesynthesis is increased accumulation of the cosubstrate, PRPP,required for the initial biochemically committed step in thepurine pathway. The increased availability of PRPP in MTX-

treated cells can be used by orotate phosphoribosyltransferaseto transform FUra to 5-fluorouridine monophosphate, which

can in turn be converted to other intracellular nucleotides.There are at least 2 toxic intracellular metabolites of FUra;

fluorodeoxyuridylate, which inhibits thymidylate synthetaseand therefore DNA synthesis, and 5-fluorouridine triphosphate,

which is incorporated into cellular RNA. Intracellular accumulation and phosphorylation of FUra was enhanced up to 4-foldin MTX-pretreated 47-DN cell cultures and was associated withsynergistic cytotoxicity. The 18- to 24-hr MTX pretreatment

interval necessary to maximally enhance FUra accumulationand toxicity is longer than that observed in both HCT-8 (6 to

12 hr) and L1210 (3 hr) cells which have shorter doublingtimes. The critical role of this pretreatment interval has notbeen reported previously (12, 22, 27) and would be expectedto be important in the design of clinical protocols.

When regeneration of the tetrahydrofolate pool is blocked byMTX, continued synthesis of thymidylate and DNA depletes thetetrahydrofolate pool and subsequently reduces de novo purinesynthesis. The fact that MTX substantially alters folate poolsand reduces purine synthesis only in cells which are synthesizing DNA has important therapeutic implications. Tumors withlonger doubling times might have proportionately fewer cellssynthesizing DNA during any given interval of MTX exposure,resulting in less opportunity for synergism with FUra. Variationsgreater than 10-fold have been observed among growth rates

of primary human breast tumors and their metastatic lesions(29). This could influence clinical response rates to sequentialMTX-FUra therapy.

Clinical trials of MTX and FUra sequenced 1 hr apart inpatients with advanced breast, colorectal, and head and neckcancers have shown encouraging responses with little if anyhost toxicity (13,17, 26, 30). The long doubling times of breasttumors (29) and our experimental data with 47-DN cells sug

gest that pretreatment intervals >18 hr are necessary to produce optimal cytotoxic synergism between MTX and FUra.Studies trying to demonstrate the clinical utility of prolongingMTX pretreatment intervals might show that increased hosttoxicity negates the therapeutic advantage of enhanced tumorcell kill. One small uncontrolled study which gave patientscomparable doses of MTX and FUra sequenced 4 hr apartreported unacceptable myelosuppression and attributed thistoxicity to the longer MTX pretreatment interval (28). Anotherstudy, comparing MTX-FUra schedules in tumor-bearing mice,found that 24-hr sequencing was maximally tumoricidal butalso increased early toxic deaths more than 6-fold over the

other 2 drug schedules (23).In light of these studies, we conducted a Phase I study to

determine whether 24-hr-sequenced MTX-FUra could be tol

erated. Although the dose of drugs administered was identicalto that used in previous studies with 1-hr drug sequencing (13,

26), MTX was administered p.o. in divided doses over 24 hr forpatient convenience to sustain serum MTX levels above 1 /IM.The administration of LV was scheduled 6 hr after i.v. bolusadministration of FUra, in accordance with our experimentalresults (Table 4). No toxicity occurred with 38% of treatmentcourses; mild to moderate leukopenia and mucositis occurred

Optimal Scheduling of MTX and FUra in Human Breast Cancer

with 27 and 38% of courses, respectively (Table 5).Since LV can rapidly reverse the biochemical effects of MTX,

decreasing both intracellular pools of PRPP and rate of FUraaccumulation (5), it is important to establish the optimal clinicaltiming of LV administration. Our experiments (Table 4) supportthe theoretical concept that LV should be given after FUraadministration.

Circulating levels of "salvage" purines in human plasma can

potentially prevent the antitumor effects of de novo purineinhibitors and abrogate the sequence-dependent synergism ofMTX-FUra (5, 15, 16). The kinetic characteristics of the enzymes necessary to salvage hypoxanthine, inosine, or adeno-

sine vary considerably from tumor to tumor (3, 8, 21, 31 ); it isof interest that the normal physiological range for plasmahypoxanthine is between 0.1 and 10 ¡IM(16), the same concentration range that reverses FUra accumulation in MTX-pretreated 47-DN (Table 3). Thus, agents known to decreaseexcretion or increase circulating levels of hypoxanthine suchas allopurinol, probenecid, sulfinpyrazone, and salicylates (14)might also reduce the likelihood of antitumor activity by MTX-

FUra therapy. Stem cell assays have been used to try todetermine the plasma hypoxanthine concentrations that mightselectively protect bone marrow precursors from MTX toxicity(16). Studies comparing the selectivity of sequential MTX-FUra

toxicity by stem cell assay of normal human marrow and freshlyresected breast cancer specimens would be of considerableinterest and further help in the rational design of clinical treatment for breast cancer utilizing these 2 agents.

ACKNOWLEDGMENTS

We would like to thank Dr. Rabindranth Nayak at Columbia University forproviding the 47-DN cells, Barbara Stanley and Joan Gesmonde for their tissueculture assistance, and the following people for their aid in their work andpreparation of this manuscript: Robert Heimer, Lee Newcomer, Arlene Cashmore.and Hillary Raeffer.

REFERENCES

1. Benz. C., and Cadman. E. Modulation of 5-fluorouracil metabolism andcytotoxicity by antimetabolite pretreatment in human colorectal adenocar-cinoma, HCT-8. Cancer Res., 41: 994-999, 1981.

2. Benz, C., Schoenberg, M., Choti. M., and Cadman, E. Schedule-dependentcytotoxicity of methotrexate and 5-fluorouracil in human colon and breasttumor cell lines. J. Clin. Invest., 66. 1162-1165, 1980.

3. Berlino, J., and Allaudeen, H. Biochemical abnormalities in some humanneoplasms; the leukemias. In: T. Symington and R. Carter (eds.), ScientificFoundations of Oncology, pp. 77-85. Chicago: Year Book Medical Publish

ers, Inc., 1976.4. Cadman, E., Benz, C., Heimer, R., and O'Shaughnessy, J. The effect of de

novo purine synthesis inhibitors on 5-fluorouracil metabolism and cytotoxicity. Biochem. Pharmacol., 30: 2469-2472. 1981.

5. Cadman, E., Heimer, R., and Benz, C. The influence of methotrexate pretreatment on 5-fluorouracil metabolism in L1210 cells. J. Biol. Chem., 256:1695-1704, 1981.

6. Cadman, E., Heimer, R., and Davis, L. Enhanced 5-fluorouracil nucleotideformation after methotrexate administration: explanation for drug synergism.Science (Wash. D. C.), 205: 1135-1137, 1979.

7. Carbone, P. P. Chemotherapy in the treatment strategy of breast cancer.Cancer (Phila.), 36 (Suppl.): 633-637, 1975.

8. Chang, C., Brockman, R., and Bennett, L. Adenosine kinase from Li 210cells. J. Biol. Chem., 255: 2366-2371, 1980.

9. Chlebowski, R. T., Irwin, L. E., Pugh, R. P., Sadoff, L.. Hestorff, R.. Wiener,J. M., and Bateman, J. R. Survival of patients with metastatic breast cancertreated with either combination or sequential chemotherapy. Cancer Res.,39:4503-4506, 1979.

10. Cooper, R. G., Holland, J. F., and Glidewell. O. Adjuvant chemotherapy ofbreast cancer. Cancer (Phila.). 44: 793-798, 1979.

11. Davis, T. E., and Carbone, P. P. Drug treatment of breast cancer. Drugs, 16:441-464, 1978.

12. Donehower, R. C.. Allegra. J. C., Uppman, M. E., and Chabner, B. A.

MAY 1982 2085

Research. on August 11, 2021. © 1982 American Association for Cancercancerres.aacrjournals.org Downloaded from

C. Benz et al.

Combined effects of methotrexate and 5-fluoropyrimidine on human breastcancer cells in serum-free culture. Eur. J. Cancer, 16: 655-661, 1980.

13. Gewirtz. A. M., and Cadman, E. Preliminary report on the efficacy ofsequential methotrexate and 5-fluorouracil in advanced breast cancer. Cancer (Phila.), 47: 2552-2555. 1981.

14. Goldfinger, S., Klinenberg, J., and Seegmiller, J. The renal excretion ofoxypurines. J. Clin. Invest., 44: 623-628, 1965.

15. Houghton, J., and Houghton, P. 5-Fluorouracil in combination with hypoxan-thine and allopurinol: toxicity and metabolism in xenographs of humancolonie carcinomas in mice. Biochem. Pharmacol.. 29. 2077-2080, 1980.

16. Howell, S.. Mansfield. S., and Taetle, R. Thymidine and hypoxanthinerequirements of normal and malignant human cells for prevention againstmethotrexate cytotoxicity. Cancer Res., 41: 954-950, 1981.

17. Isacoff, W. H., Tisman, G.. and Block, J. B. Treatment of advanced colorectalcarcinoma with high dose methotrexate given alone and in combination with5-fluorouracil. Proc. Am. Assoc. Cancer Res., 20. 381. 1979.

18. Kardinal, L. G., Donegan. W. L., and Spratte. J. S., eds. Chemotherapy. In:Cancer of the Breast, pp. 405-447. Philadelphia: W. B. Saunders Co.,

1979.19. Keydar, I., Chen, L., Karby, S., Weiss, F., Delarea, J., Radu, M., Chartcik,

S.', and Brenner, H. Establishment and characterization of a cell line of

human breast carcinoma origin. Eur. J. Cancer, 15: 659-670, 1979.20. Kirkwood, J.. Ensminger, W., Rosowsky, A., Papathanasopoulos, N., and

Frei, E. Comparison of pharmacokinetics of 5-fluorouracil and 5-fluorouracilwith concurrent thymidine infusions in a Phase I trial. Cancer Res., 40:107-113, 1980.

21. Kizaki. H., Morris, H., and Weber. G. Behavior of inosine phosphorylaseactivity in normal, differentiating, and regenerating liver and in hepatomas.Cancer Res.. 40: 3334-3344. 1980.

22. Kufe, D. Metabolic enhancement of 5-fluorouracil (5-FU) incorporation into

human tumor cell RNA. Proc. Am. Assoc. Cancer Res., 21: 265, 1980.23. Mulder, J., Smink, T., and Van Putten, L. 5-Fluorouracil and methotrexate

combination chemotherapy: the effect of drug scheduling. Eur. J. CancerClin. Oncol., 17: 831-837, 1981.

24. Oken, M., Creech, R., Tormey, D., Morton, J., Davis, T., McFadden, E., andCarbone. P. Toxicity and response criteria of the Eastern CooperativeOncology Group. Cancer Clin. Trials, in press. 1982.

25. Osborne, C. K., Knight. W. A., Ill, Yochonowitz, M. G., and McGuire, W. L.Modern approaches to the treatment of breast cancer. Blood, 56: 745-7521980.

26. Pitman, S. W., Kowal, C. S., Papac, R. J., and Berlino, J. R. Sequentialmethotrexate-5-fluorouracil: a highly active drug combination in advancedsquamous cell carcinoma of the head and neck. Proc. Am. Assoc. CancerRes., 21. 473, 1980.

27. Sawyer, R. C., Stolfi, R. L., and Marin, D. S. Uridine reversal of thymidine-fluorouracil-induced host toxicity in the CD8 murine mammary tumor system.Proc. Am. Assoc. Cancer Res., 21: 14, 1980.

28. Solan, A., Vogl, S. E., Kaplan, B. H., Lanham, R., and Berenzwieg, M. 5-Fluorouracil and methotrexate in combinations for colorectal cancer: unacceptable toxicity when intermediate dose MTX precedes FU by 4 hours.Proc. Am. Assoc. Cancer Res., 21: 335, 1980.

29. Spratt, J. S., and Spratt, J. A. Growth rates. In: W. L. Donegan and J. S.Spratt (eds.), Cancer of the Breast, pp. 197-220. Philadelphia: W. B.Saunders Co., 1979.

30. Tisman, G., Isacoff, W. H., and Drakes, T. Salvage of breast cancer adjuvanttreatment failures with high dose methotrexate and 5-FU. Proc. Am. Assoc.Cancer Res., 20. 140, 1979.

31. Weber, G. Enzymatic strategy of the cancer cell. In: J. Schultz and F. Ahmad(eds.), Cancer Enzymology, pp. 63-99. New York: Academic Press Ine1976.

2086 CANCER RESEARCH VOL. 42

Research. on August 11, 2021. © 1982 American Association for Cancercancerres.aacrjournals.org Downloaded from

1982;42:2081-2086. Cancer Res   Chris Benz, Tina Tillis, Ellen Tattelman, et al.   Human Breast CancerOptimal Scheduling of Methotrexate and 5-Fluorouracil in

  Updated version

  http://cancerres.aacrjournals.org/content/42/5/2081

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/42/5/2081To request permission to re-use all or part of this article, use this link

Research. on August 11, 2021. © 1982 American Association for Cancercancerres.aacrjournals.org Downloaded from