[CANCER RESEARCH 44, 3278-2284, June 1984]

Improved Therapeutic Index with High-Dose Methotrexate: Comparison ofThymidine-Purine Rescue with Citrovorum Factor Rescue in Mice1

Lawrence L. Samuels2 and James A. Straw3

Department of Pharmacology, The George Washington University Medical Center, Washington, DC 20037

ABSTRACT

Two biochemically different rescue agents, citrovorum factor(CF) and thymidine-inosine-allopurinol (TIA), were compared in

an attempt to identify the mechanism for the increased therapeutic index achieved with high-dose methotrexate (MTX) plus res

cue.Both CF and TIA were capable of protecting mice from MTX

dosages up to 2000 mg/kg. Treatment of L1210-bearing mice

with 2000 mg/kg MTX plus CF or TIA produced a 70 and 100%increase in life span, respectively, compared with 29% increasein life span achieved with the maximally tolerated dose of MTXalone. Bioassay of surviving peritoneal L1210 cells showed thata 4.5-log tumor kill occurred 24 hr after 2000 mg/kg MTX, while400 mg/kg MTX produced only a 2-log cell kill. This differentialtumor kill in the 24-hr period after MTX and prior to the onset of

rescue accounted for the observed increase in animal survivaltimes. In addition, treatment with 2000 mg/kg MTX resulted in aone-log-greater tumor kill of cells metastasized to the brain than

did treatment with 400 mg/kg MTX.Following 2000 mg/kg MTX, additional tumor kill, as measured

by bioassay, occurred during the period of TIA rescue but notduring CF rescue, which was consistent with the observeddifferences in survival times between CF- and TIA-rescued mice.

DNA synthesis in tumor and host tissue, as measured by therate of [3H]dCyd incorporation into DNA, was cyclic after TIA

administration but not after CF administration. The cyclic natureof DNA recovery in TIA-treated mice paralleled plasma kineticsof thymidine. It is postulated that "thymineless" intervals created

by the rapid disappearance of thymidine resulted in inhibition ofDNA synthesis and additional tumor cell kill during TIA rescue.Normal tissue did not appear to be adversely affected by exposure to these "thymineless" intervals.

INTRODUCTION

The antimetabolite MTX,4 like most anticancer drugs, has

limited selectivity and, as a result, its effectiveness is limited bytoxicity to normal proliferating cells, particularly intestinal epithelium and bone marrow. One approach towards improving thetherapeutic index of MTX is to administer potentially lethal doses

1This research was supported by NIH Grant CA-22866.

Presented in part at the 71st and 72nd Annual Meetings of the AmericanAssociation for Cancer Research, 1980 and 1981 (18,19).

From a dissertation presented to the Department of Pharmacology, The Graduate School of Arts and Sciences, The George Washington University, 1981, byLawrence L. Samuels, in partial fulfillment of the requirements for the Ph.D.

* Present address: Laboratory of Molecular Therapeutics, Memorial Sloan-Ket-

tering Cancer Center, New York, NY 10021.3To whom requests for reprints should be addressed.4The abbreviations used are: MTX, methotrexate; CF, citrovorum factor; dThd,

thymidine; ILS, increase in life span; TIA, thymidine-inosine-allopurinol; dCyd,deoxycytidine: PCA, perchloric acid; t.i.d., 3 times/day.

Received July 2,1982; accepted February 24,1984.

of the drug, followed by a "rescue" agent to prevent serious

toxicity to normal cells (1). MTX is thought to produce its anti-

tumor effects by inhibiting the enzyme dihydrofolate reductase(EC 1.5.1.3.) and depleting intracellular levels of reduced folatecofactors required for de novo purine, thymidylate, and proteinsynthesis (15). The metabolic block produced by MTX can becircumvented by providing either a reduced folate (i.e., CF) orthe end products of folate metabolism, namely, dThd and apurine source. Goldin ef al. (6-8) first demonstrated that simul

taneous or delayed administration of CF could protect normalmice from toxic doses of MTX. These same investigators showedthat delayed administration of CF in tumor-bearing animals couldprevent MTX-induced host toxicity without diminishing the anti-tumor activity of the drug (6-8). Since the original observationsof Goldin et al., high-dose MTX plus CF rescue has been used

with some success in the treatment of certain human cancers(4, 13, 14). Despite many studies, however, the mechanism bywhich CF enhances the therapeutic index of MTX is not wellunderstood.

Protection of cells from MTX toxicity using dThd and a purinesource was first shown in vitro by Hakala et al. (9, 10). Studiesby Straw ef al. (23) show that MTX toxicity can be prevented inmice by administration of hypoxanthine:allopurinol:dThd. In tumor-bearing animals, Harrap ef al. (11) demonstrated an improved antitumor activity when MTX was followed by hypoxan-

thine:allopurinol:dThd rescue.Despite differences in their respective biochemical and phar-

macokinetic properties, both CF and dThd:purine are capable ofincreasing the therapeutic index of MTX when given in theappropriate dose-schedule. This ability of both CF and dThd:

purine to enhance the antitumor activity of MTX suggests that afactor common to both rescue agents may be responsible forthis event. An understanding of this common factor would helpto elucidate the basis for the selectivity of high-dose MTX with

rescue.In the present studies, we confirm the ability of CF or TIA to

enhance the therapeutic efficacy of MTX in L1210-bearing mice.

The findings suggest that cell kinetic differences between hostand neoplastic cells are responsible for the selective actions ofhigh-dose MTX plus rescue. In addition, we observed that nu-

cleoside rescue is superior to rescue with CF in terms of increasing the survival times of tumor-bearing mice. The basis for this

improved therapeutic efficacy of MTX using nucleoside rescueis discussed.

MATERIALS AND METHODS

Chemicals. MTX, CF, and dThd were provided by Dr. V. Narayana,National Cancer Institute (Bethesda, MD). Inosine, thymine, and dCydwere purchased from Sigma Chemical Co. (St. Louis, MO). |3H|dCyd(28.3

Ci/mmol) was obtained from New England Nuclear (Boston, MA). Allo-purinol was a gift from Burroughs-Wellcome Laboratories (Research

2278 CANCER RESEARCH VOL. 44

on May 23, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Therapeutic Efficacy of High-Dose Methotrexate plus Rescue

Triangle Park, NC). All other chemicals were of reagent grade, exceptmethanol, which was high-pressure liquid chromatography grade.

Animals. Male C57BL/6 x DBA/2 (hereafter called BD2F,) micepurchased from Sprague-Dawley Laboratories (Madison, Wl) were used

in all experiments. Mice ranged in weight from 22 to 25 g. In the bioassaystudies, male DBA/2 mice (20 to 24 g) were used as recipient mice.Animals were fed standard lab chow and water ad libitum. L1210 ascitescells were maintained in the peritoneal cavity of DBA/2 mice by weeklypassage of 10s cells.

Survival Studies. BD2F, mice were treated i.p. with MTX (dissolvedin (2 to 4% NaHCOa) on Day 4 after inoculation i.p. withlO6 L1210 cells.

At 24 hr after MTX, mice were given i.p. injections of either CF (6 mg/kg) or TIA (dThd, 500 mg/kg; inosine, 50 mg/kg; allopurinol, 10 mg/kg)t.i.d. for 5 consecutive days. (Allopurinol, a xanthine oxidase inhibitor,was included to block purine catabdism.) Unrescued controls received0.9% NaCI solution (saline).

The number of surviving animals was determined daily, and mediansurvival time of each treatment group was calculated as the day aftertumor inoculation, when the number of surviving animals was (A/ - 1)/2.

Data from several survival experiments were pooled, and statisticalanalysis of the median survival times was carried out using the nonpar-ametric Mann-Whitney U test.

Tumor-bearing and non-tumor-bearing mice were weighed daily, and

the body weights were expressed as the percentage of Day 0 (day ofMTX administration) body weight.

Bioassay Method for Ouantitating L1210 Cells. A bioassay technique, based on the procedure of Skipper ef a/. (22), was used toquantitate the number of L1210 cells which survive MTX treatment. Theadvantage of the bioassay compared with dye exclusion and electroniccounting is that the bioassay measures only tumor cells which retain theability to proliferate and ultimately kill the host.

Mice were inoculated i.p. with 106 L1210 cells and treated on Day 4

with either saline (untreated controls), MTX (400 mg/kg), or MTX (2000mg/kg). At 24 hr after MTX, mice were sacrificed, and tumor cells wereflushed from the peritoneal cavity with 0.1 mM CF in 0.9% NaCI solution(at room temperature). CF was used to accelerate the loss of MTX fromdrug-treated tumor cells to more completely wash MTX from cells. L1210cells were washed again and resuspended in ice-cold 0.9% NaCI solution.One-tenth of the total volume of the cell suspension (0.4 ml) was

réinjectée!into a group of recipient DBA/2 mice. Since the L1210 tumorline grows best in the DBA/2 strain, these mice were used to maximizethe sensitivity of the assay.

The brain was removed from the donor mice and macerated by flushingthrough a 20-gauge needle and then a 25-gauge needle. One-tenth of

the total volume of brain homogenate (0.4 ml) was reinjected into groupsof recipient mice. Survival times of all recipient mice were determined asdescribed previously.

A standard curve was generated by inoculating known quantities ofL1210 cells into the recipient mice and determining their respectivesurvival times. Tumor cells from untreated leukemic mice were harvestedand counted electronically. Groups of 8 recipient DBA/2 mice werereinoculated i.p. with either 102, 103, 104, or 10s L1210 cells, and their

respective life spans were determined. Life spans of recipient miceinoculated with brain and L1210 cells from MTX-treated mice were

compared to the standard curve, which plotted host survival time versustumor cell number, to determine the number of surviving L1210 cells.

Incorporation of [3H]dCyd into DMA. [5-3H]dCyd was chosen as the

radiolabeled precursor to study the effect of high-dose MTX plus rescue

on the rate of DNA synthesis, because it is incorporated into DNAwithout passing through a folate-dependent step. This characteristicallows dCyd to be used to compare the effects of CF and TIA on MTX-induced changes in the rate of DNA synthesis. In addition, since the 3H

group is in position 5 of the pyrimidine ring, the label would be lost ifdCyd was incorporated into DNA via deamination to dUMP and conversion to TMP. A large carrier dose of unlabeled dCyd was used tocircumvent the effects of MTX on intracellular pool sizes of dCyd.

Preliminary studies showed that endogenous pools were saturated by adCyd dose of 50 mg/kg, and incorporation of [3H]dCyd was linear for 30

min (data not shown).Mice were killed by cervical dislocation 30 min after inoculation with

[3H]dCyd (2 mCi/kg; 50 mg/kg), and L1210 cells were flushed from the

peritoneal cavity with 0.9% NaCI solution. The cells were centrifuged,and the pellet was resuspended in 5 ml of hypotonie saline (0.2% NaCIsolution) to lyse RBC and returned to isotonicity with 1 ml of 4.5% NaCIsolution. After the supernatant was discarded, the cell pellet was suspended in 5 ml of 0.2 N PCA and stored at 4°overnight.

Small intestine was obtained by cutting 1 to 2 cm distal to the pylorusand removing 18 cm of intestine. Fatty tissue was removed, and thelumen was flushed clean with saline. The tissue was homogenized in 6ml of 0.2 N PCA, and the homogenate was stored at 4°until further

analysis.The femur from each mouse was removed, the ball joint was clipped,

and the marrow was flushed from the femur with 1.5 ml of 0.2 N PCA.The marrow from two femurs was pooled, and the samples were storedat 4°.

DNA from the tissue samples was extracted by a modified Schmidt-

Thanhouser procedure (23). All samples were centrifuged at 1000 x gfor 5 min. The precipitated material was suspended in 4 ml of 0.2 N PCA,centrifuged at 2000 x g for 10 min, and then washed twice with 4 ml of0.2 N PCA (bone marrow, tumor) or 8 ml of 0.2 N PCA (small intestine)in order to remove acid-soluble materials, üpidswere extracted once

(bone marrow, tumor) with 4 ml of ethanol:ether (3:1, v/v) or twice (smallintestine) with 8 ml of ethanohether. After each washing, the pellets werecentrifuged at 500 x g for 5 min. RNA was hydrolyzed by suspendingthe precipitate in 4 ml of 0.5 N KOH and heating in a water bath for 1 hrat 37°. Following incubation, 1.2 ml of 50% trichtoroacetic acid was

added to give a final concentration of 5% trichloroacetic acid, and thesample was chilled for 20 min. The DNA pellet was washed once with 4ml of 0.2 N PCA, resuspended in 0.5 N PCA, and hydrolyzed for 20 minat 90°.The samples were centrifuged, and the DNA-containing super

natant was saved for determination of radioactivity and DNA concentration.

Duplicate 500-jtl aliquots of the supernatant were analyzed by the

method of Burton (3) to determine DNA concentration. Radioactivity wasmeasured using liquid scintillation spectrometry.

Determination of the Number of Nucleated Bone Marrow Cells. Thefemurs of sacrificed mice were dissected from the mice. The marrowwas flushed twice with 2.5 ml of Isoton II (Coulter Diagnostics, Hialeah,FL), and the marrow from two femurs was pooled. Three drops ofZapisoton (Coulter Diagnostics) were added to lyse the membranes ofnucleated and nonnudeated cells. After the cell suspension was gentlymixed, nuclei were counted electronically using a Model 2F CoulterCounter (100-/im aperture tube).

Separation and Quantitation of dThd, Thymine, Inosine, and Hy-

poxanthine. Blood was obtained from anesthetized mice from the orbitalsinus following removal of the eye. The blood was collected in 1.5-ml

microcentrifuge tubes. Plasma was diluted 1:1 with distilled water anddeproteinized by boiling for 15 min at 90° and with 1 N PCA. The

supernatant was neutralized with an appropriate volume of 2 N KOH andstored at -6° until further analysis.

Purine and pyrimidine nudeosides and bases were separated by ahigh-pressure liquid chromatography method modified from the methodof Rustum (17). The solvent delivery system consisted of a Waters M-

45 and 6000A pump controlled by a Model 660 Solvent Programmer(Waters Associates, Mitford, MA). Samples were injected onto a C-18reverse-phase Column (üChrosorb RP 18,10 urn; Altex Scientific, Inc.,

Berkley, CA) and eluted with 5% methanol in water. Samples weredetected using a Model 440 UV absorbance detector (254 nm; WatersAssociates). Peak areas were integrated using either a Hewlett-Packard3381 A Integrator (Hewlett-Packard, Corvallis, OR) or a Waters Data

Module (Water Associates), and concentrations were calculated basedon the external standard method.

JUNE 1984 2279

on May 23, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

L. L Samuelsand J. A. Straw

RESULTS

Protection from MIX Toxicity. Animals treated with a singledose of MIX (400 mg/kg) lost 25% of their body weight (Chart1) and died within 6 days. However, when either CF or TIA wasadministered, serious weight loss and animal deaths were prevented. These results (Chart 1) demonstrate that CF and TIAcould each protect mice from drug toxicity following MTX dosesas high as 2000 mg/kg. An equimolar concentration of thymine(260 mg/kg) was found to substitute for dThd in the TIA rescueregimen and protect mice from MTX toxicity. However, dThdalone (500 mg/kg) given t.i.d. for 5 days failed to prevent animaldeaths after a single injection of MTX (2000 mg/kg) (results not

shown).The effect of CF and TIA on reversing MTX toxicity to bone

marrow was determined in leukemic mice treated with MTXfollowed by either CF or TIA rescue. In untreated control mice,there were 2.03 x 107 nucleated cells/femur (Chart 2). Treatment

with MTX (400 mg/kg) and MTX (2000 mg/kg) produced a nadirof 9.6 and 4.6 x 106 cells, respectively. When CF or TIA was

given after MTX (2000 mg/kg), the total number of nucleatedmarrow cells at the nadir was slightly greater than that inunrescued animals. Recovery following MTX plus CF occurredafter 4 days, whereas no apparent recovery of marrow cells wasobserved by Day 6 in TIA-treated animals. However, recovery ofbone marrow cells was sufficient to prevent host lethality, sincethe mice given TIA rescue survived potentially lethal doses of

MTX.High-Dose versus Conventional-Dose MTX. Animals were

treated with various doses of MTX 4 days after inoculation with106 L1210 cells. Beginning 24 hr after MTX, animals were giveneither no rescue, CF, or TIA t.i.d. for 5 days. The maximum doseof MTX alone which caused no toxic deaths was 200 mg/kg. Asdescribed earlier (Chart 1), animals could tolerate MTX dosages

105-

100—I-

Iuj 95-

o>Q 90-U.o

85-

80-

400MTX+TIA

2000MTX+CF

400MTX+CF

2000 MTX+TIA

400 MTX alone

DAYS AFTER MTX

Chart 1. Effects of MTX plus rescue on body weights of normal BD2F, mice.Mice were given a single i.p. injection of MTX, followed 24 hr later with either CFor TIA given t.i.d. for 5 days. »,MTX (400 mg/kg) alone; »,MTX (400 mg/kg) +CF; O, MTX (2000 mg/kg) + CF; A, MTX (400 mg/kg) + TIA; A, MTX (2000 mg/kg) + TIA. In the MTX (400 mg/kg) alone group, all 5 mice were dead by Day 5.

Å’

111U.te.ilio.

111o

ooc

2-

10' -

8-

6-

3ozu,ooc

iz

DAYS AFTER MTX

Chart 2. Effect of MTX plus rescue on the number of nucleated bone marrowcells per femur of BD2F, mice. Animals received either no drug (Day 0 control), asingle i.p. injection of MTX (400 mg/kg) (*), or MTX (2000 mg/kg) 4 days after i.p.inoculation of 1 X 10* 1.1210cells. Twenty-four hr after MTX, mice were giveneither 0.9% NaCIsolution (unrescuedcontrols) (•),CF (6 mg/kg) (LI),or TIA (dThd,500 mg/kg; inosine,50 mg/kg; allopurinol,10 mg/kg) (A) t.i.d. for 5 days. At varioustimes, bone marrow cells were obtained, and the number of nucleated cells wasdetermined as described in "Materials and Methods." Each point represents the

mean of 4 paired femurs; bars, S.E.

up to 2000 mg/kg without serious toxicity when CF or TIArescue was given. As seen in Chart 3, a dose-dependent increase

in life span occurred which appeared to plateau between 400and 1000 mg/kg, and which then dramatically increased followingMTX (2000 mg/kg). The percentage of ILS achieved with themaximum dose of MTX (2000 mg/kg) with rescue was significantly greater (p < 0.001) than that achieved with the maximumdose of MTX alone (200 mg/kg). Furthermore, TIA rescue wassuperior to rescue with CF. In mice given MTX (2000 mg/kg),rescue with CF and TIA resulted in a 70 and 100% ILS, respectively (p < 0.001). The difference between CF and TIA rescuewas not apparent at MTX doses of less than 2000 mg/kg.

Quantitation of L1210 Cell Kill. Chart 4 shows the number ofi.p. clonogenic cells present at various times after MTX administration. Twenty-four hr after MTX and prior to the start ofrescue, MTX at 400 mg/kg produced about a 2-log tumor cellkill, whereas the 2000-mg/kg dose produced about a 4.5-log

tumor cell kill. Based on the tumor burden present at 24 hr anda doubling time of 13 hr, the day of death (when tumor burdenreached 1.5 x 109 cell) was predicted to be 8.5 days for control,

13 days for MTX (400 mg/kg), and 17 days for MTX (2000 mg/kg). The observed survival times were 7.8 days for control, 9.6days for MTX (400 mg/kg) plus rescue with CF, and 9.9 daysfor MTX 400 mg/kg) plus rescue with TIA. After MTX (2000 mg/kg), mice survived for 13.4 days when rescued with CF and 16.8days when rescued with TIA. These data show that the major

2280 CANCER RESEARCH VOL. 44

on May 23, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Therapeutic Efficacy of High-Dose Methotrexate plus Rescue

way in which the high-dose regimen increased survival time was

by producing a greater tumor cell kill in the 24 hr prior to rescue.Chart 4 also shows that resumption of tumor cell growth

occurred in both rescued and unrescued animals. During the first48 to 72 hr of rescue, the apparent doubling times in CF- andTIA-rescued mice were 15 and 36 hr, respectively. After Day 8,the rate of tumor regrowth was similar in both CF- and TIA-

rescued animals, and it paralleled the growth rate in untreatedmice. Delayed resumption of growth in the TIA-rescued mice

resulted in a lower number of clonogenic cells at Day 10 ascompared to the CF-rescued mice. Furthermore, this differencein tumor burden at Day 10 was sufficient to account for thedifference in survival times (13.5 days with CF rescue and 16.8days with TIA rescue).

Effect of MTX Dose on L1210 Cells That Metastasized tothe Brain. A bioassay was used to test if a relationship existsbetween the dose-related increase in tumor kill and the capacity

to eradicate L1210 cells which migrate to the brain. Mice weretreated on Day 4 after i.p. inoculations of 106 L1210 cells with

MTX at either 400 or 2000 mg/kg. Twenty-four hr after MTX,

mice were sacrificed, and the number of clonogenic tumor cellswas determined by bioassay. There was an approximately one-

log difference in tumor kill between mice given MTX (400 mg/kg)and mice given MTX (2000 mg/kg) (data not shown). Since thegrowth kinetics of L1210 cells in the brain is the same as in theperitoneal cavity and since there is a specific tumor burdenassociated with animal death (22), this 1-log difference corresponds to a 1.5-day increase in animal survival time. This differ

ential tumor kill of L1210 cells which have metastasized to thebrain appears to account for only a portion of the total (4 to 7days) increase in median survival times between animals givenMTX (400 mg/kg) and those given MTX (2000 mg/kg).

Ul

2Uloc

100

90

80

70

60

50

40

30

20

CF0

100

60

60

40

20

100 200 400 1000

mg/kg MTX

2000

Chart 3. Survival of L1210-bearing mice following treatment with MTX with orwithout rescue. Mice were given inoculations i.p. with 1 x 10* L1210 cells and

were treated 4 days later with increasing doses of MTX. Untreated controls weregiven saline in place of MTX. Twenty-four hr after MTX, mice received either saline

(no rescue) (•),CF (6 mg/kg) (U). or TIA (dThd, 500 mg/kg; inosine, 50 mg/kg;allopurinol. 10 mg/kg) (A) t.i.d. for 5 days. In each individual experiment, 8 to 10mice per treatment group were used. The percentage of ILS was determined (asdescribed in "Materials and Methods") for each treatment group within each

experiment, and the mean percentage of ILS from all survival experiments wascalculated. 'Increase in life span greater than MTX (1000 mg/kg) (p < 0.001);"increase in life span greater than MTX (2000 mg/kg) + CF (p < 0.001 ). Bar. S.E.

10' -

10- -

10«

104-

10" •

10«

10'

3CONTROL

MTX 400 mg/kg

MTX2000 mg/Kg

10

DAYS AFTER L1210 INOCULATION

Chart 4. Bioassay of i.p. clonogenic L1210 cells following treatment with MTXplus rescue. BD2F, mice were inoculated with 10* L1210 cells i.p. on Day 0. OnDay 4, mice were treated with a single i.p. injection of MTX (2000 mg/kg). Twenty-four hr after MTX, mice received either CF (6 mg/kg) or TIA (dThd, 500 mg/kg;inosine, 50 mg/kg; allopurinol, 10 mg/kg) t.i.d. for the duration of the experiment.Tumor cells from one group of 10 donor mice were harvested from the peritonealcavity and reinoculated into one group of 10 recipient DBA/2 mice as described in"Materials and Methods," and the median life span of each group of recipient mice

was determined. L1210 cells from control mice were harvested and countedelectronically, and groups of 8 mice were inoculated with 102,10s, 10*. or 10s cells

to generate a standard curve by relating tumor inoculum to median life span. O,untreated controls, *, MTX (400 mg/kg); •,MTX alone (2000 mg/kg); D, MTX(2000 mg/kg) + CF; A, MTX (2000 mg/kg) + TIA.

Effects of High-Dose MTX plus Rescue on DMA Synthesis.Incorporation of [3H]dCyd into DMA was used to calculate the

rate of DNA synthesis in small intestine, bone marrow, and L1210cells. In the absence of rescue, MTX inhibited DNA synthesis inal three tissues for at least 68 hr (Charts 5 and 6). Recovery ofDNA synthesis was dependent on the rescue agent used. DuringTIA rescue, a marked resumption of DNA synthesis was observed in bone marrow and small intestine. As seen in Chart 5,the rate of [3H]dCyd incorporation into bone marrow and intes

tinal epithelial DNA was cyclic, peaking 2 hr after each TIAinjection and returning to prerescue rates within 4 hr. As a result,intervals between TIA injections existed during which DNA synthesis remained inhibited. Although the magnitude of recoveryof L1210 synthesis was less than that seen in intestinal epithelium and bone marrow, the pattern of resumption of L1210 cellswas also cyclic from 24 to 48 hr.

In contrast to the cyclic recovery of DNA synthesis followingTIA administration, there was a gradual resumption of DNAsynthesis in small intestine and bone marrow following CF (Chart6). During the first 2 days of rescue (24 to 48 hr after MTX), therate of gut DNA synthesis increased from 25% of control (at 24

JUNE 1984 2281

on May 23, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

L. L Samuelsand J. A. Straw

hr) to 50% of control (at 57 hr). During Day 3 of rescue (72 to88 hr), the rate of DNA synthesis overshot control levels. DMAsynthesis in bone marow showed little recovery until the end ofthe second day (62 hr) of CF rescue. Unlike the case with thesmall intestine, the degree of recovery of synthesis of bonemarrow DNA failed to reach 100% of control during 3 days ofrescue. Little resumption of DNA synthesis was observed inL1210 cells. L1210 DNA synthesis was 10% of control 24 hrafter MTX, and it remained less than 10% of control during 3days of CF rescue.

Plasma Kinetics of TIA. In order to determine if the kineticsof TIA could account for the cyclic resumption of DNA synthesis,plasma levels of dThd and inosine and their metabolites thymineand hypoxanthine were determined using high-pressure liquid

30 35 40 45 50 55 60 65

HOURSAFTERMTX ADMINISTRATION

Chart 5. Effect of MTX plus TIA on DNA synthesis. BD2F, mice were treatedwith MTX (2000 mg/kg) ¡.p.at f = 0 hr. Twenty-four hr after MTX, TIA (dThd, 500mg/kg; inosine, 50 mg/kg; altopurinol, 10 mg/kg) was given t.i.d. for the durationof the experiment. At various time points, mice were sacrificed, and the rate ofincorporation of pHJdCyd into DMAof gut (•),bone marrow (•),and L1210 cells(»)was determined as described in "Materials and Methods.' N = 4 for each point;

bars. S.E. Open symbols, animals treated with MTX alone (24, 48, and 67 hr).Arrows, time of TIA injections.

50 55 «0 66 70

HOURS AFTER MTX

Chart6. Effect of MTX plus CF on DNA synthesis. BD2F, mice were treatedwith MTX (2000 mg/kg) i.p. at f = 0 hr. Twenty-four hr after MTX, CF (6 mg/kg)was given t.i.d. for the duration of the experiment. At various times, mice weresacrificed, and the rate of incorporation of pH]dCyd into DNA of gut (•),bonemarrow (•).and L1210 cells (»)was determined as described in "Materials andMethods." N = 4 for each point; bars. S.E. Open symbols, mice treated with MTX

atone.Arrows, time of CF injections.

10-

10-

10-' -

0 1 2 3 4 5 e 7

HOURS

Chart 7. Plasmalevelsof dThd and thyminefollowingTIA administration.Tumor-bearingmice were given a single i.p. injection of TIA (dThd, 500 mg/kg; inosine, 50mg/kg; altopurinol, 10 mg/kg) 24 hr after MTX. At various times after TIA, Woodwas collected, and plasma concentrations of dThd and thymine were determinedby high-pressure liquid chromatography as described in "Materials and Methods."

•,dThd; O, thymine. Eachpoint represents the mean of 4 mice; oars, S.E.

chromatography. Following TIA, dThd plasma levels increased100-fold within 10 min and then declined, with an apparent half-

life of 16 min (Chart 7). Plasma concentrations of dThd returnedto endogenous levels 4 to 5 hr after TIA administration. Thymine,a major metabolite of dThd, increased 10-fold, peaking 30 minafter TIA. Within 4 hr, thymine plasma concentrations also returned to endogenous levels. A 30-fold increase in inosine plasmalevels was observed within 30 min of TIA administration (datanot shown). Inosine declined rapidly and reached endogenouslevels by 4 hr. Consistent with previous studies showing catab-olism of inosine to hypoxanthine, plasma levels of hypoxanthineincreased 20-fold within 5 min. Although hypoxanthine has areported half-life of 13 min (12), in the presence of altopurinol,the rate of catabolism was noticeably decreased, such that levelsremained above endogenous concentrations for almost the entire7-hr period between TIA injections.

These data show that there were intervals during TIA rescuewhen unbalanced growth could occur due to lack of thymine attimes when aequate purines were present.

DISCUSSION

The results presented in this study demonstrate that, in thesame model system, an improved therapeutic index of MTX canbe achieved using either CF or TIA rescue. Treatment of L1210-bearing mice with MTX (2000 mg/kg) followed either by CF orTIA resulted in an increase in life span that was significantlygreater than that achieved with the maximally tolerated dose ofMTX alone (Chart 3). These findings corroborate earlier studiesshowing that CF (6-8, 20) or dThd:purine (11, 23) can improve

2282 CANCER RESEARCH VOL. 44

on May 23, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

the therapeutic efficacy of MIX. More importantly, these resultsestablish that 2 rescue agents which differ in their respectivebiochemical and pharmacokinetic properties can achieve thesame end point, namely, an increase in the therapeutic indexMIX.

Two important conclusions can be derived from this data.First, the major difference between low-dose MIX and high-

dose MTX is the degree of tumor cell kill prior to rescue. As seenin Chart 4, treatment with MTX (2000 mg/kg) produced approximately a 2.5-log greater tumor kill than did treatment with MTX(400 mg/kg) during the 24-hr period following MTX, when virtually

all tumor kill occurred. Second, this increase in tumor kill occuredwithout a proportional increase in host toxicity. No significanttoxic effects on animal body weight or survival were seen innormal mice given potentially lethal doses of MTX followed byeither CF or TIA rescue. This suggests that, while tumor cellsare sensitive to increased levels of MTX, host tissue can toleratelarger MTX doses if rescue is begun within 24 hr of MTXtreatment.

The prolonged survival times achieved with high-dose MTX

may result in part from an increased ability to kill metastasizingcells, particularly cells migrating to the brain. Since most tumor-

bearing hosts die as a consequence of disseminated disease(and the brain often serves as a protected sanctuary for tumorcells), eradication of metastasizing tumor can significantly affectthe host's response to chemotherapy. One disadvantage of MTX

is that, like other polar compounds, it has a reduced ability topenetrate the blood-brain barrier and achieve sufficient concen

trations to kill tumor cells in the brain. Results from this studyshow that a single i.p. injection of MTX (2000 mg/kg) producedapproximately a 1-log greater tumor kill of cells in the brain than

did treatment with MTX (400 mg/kg).High-dose MTX may have pharmacological actions different

from low-dose MTX which could explain the differential tumor

cell kill observed between MTX (400 mg/kg) and MTX (2000 mg/kg). Assuming that survival times reflect the extent of tumor cellkill, the dose-response survival curve illustrated in Chart 3 shows

a plateau for tumor kill between MTX (400 mg/kg) and MTX(1000 mg/kg), after which an abrupt increase in life span occurswith MTX (2000 mg/kg). This dose-response relationship isuncharacteristic for a self-limiting, phase-specific drug like MTX

(2), with which increasing drug doses produce greater tumor killuntil a plateau is reached, after which no additional kill occurs.Also, since 65 to 75% of L1210 cells are in S phase (21), thegreater than 99.99% tumor kill following MTX (2000 mg/kg)(Chart 4) suggests that either high-dose MTX is not a self-limiting

agent or it can kill cells in other parts of the cell cycle. Althoughthe precise mechanism for the enhanced tumor kill remains tobe determined, results from this study suggest that high-doseMTX has sites of action additional to those of lower-dose MTX.

The observed increases in tumor cell kill following MTX (2000mg/kg) were not accompanied by a proportional increase intoxicity to normal cells. The increase in therapeutic index of high-

dose MTX rescue may result from cell kinetic differences between normal and malignant cells, namely, the presence of anormal stem cell population. Pinedo ef al. (16) demonstrated thatstem cell recruitment occurs in the presence of MTX levels whichinhibit RNA and DNA synthesis, indicating that RNA and DNAare not required for recruitment in the cell cycle. Thus, increasesin MTX concentration would not necessarily increase the depletion of the normal stem cell population. Administration of rescue

Therapeutic Efficacy of High-Dose Methotrexate plus Rescue

24 hr after MTX reverses host toxicity, suggesting that the stemcell reserve was not depleted prior to rescue intervention.

Survival times of leukemic mice given MTX (2000 mg/kg) weresignificantly different, depending on whether CF (70% ILS) orTIA (100% ILS) rescue was administered. The different effectsof CF and TIA on tumor kill are illustrated in Chart 4, and theyshow that the number of surviving L1210 cells in the peritonealcavity of MTX CF-treated mice was greater than that in mice

given MTX/TIA. Using data on tumor burden that were generatedfrom the bioassay experiments (Chart 4), the survival times oftreated mice were calculated and were found to corroborateobserved survival times. Harrap ef al. (11), using a dose schedulesimilar to that used in this lab, have also reported that TIA orHAT rescue was superior to CF rescue in terms of prolongingsurvival times of leukemic mice. However, no explanation for thisobservation was offered by these investigators.

Results from this study suggest that tumor cell kill can occurduring TIA rescue but not during CF rescue and that it accountsfor differences between CF and TIA on leukemic animal survival.We postulate that, during TIA rescue, periodic intervals occurduring which tumor cells, which were not killed prior to rescue,can undergo thymineless death, a condition of unbalanced cellgrowth which occurs when DNA synthesis is blocked yet RNAand protein synthesis remain undisturbed (5). If, however, RNAand protein synthesis are inhibited along with DNA, the cell doesnot die but enters a quiescent state and is insensitive to phase-or cycle-specific agents such as MTX. Cells can be forced into

the cell cycle again if RNA and protein synthesis are restored. Insupport of this hypothesis, we found that: (a) recovery of MTX-

inhibited DNA synthesis was cyclic following TIA administration(Chart 5), with periods between TIA injections when DNA synthesis was blocked; (b) plasma levels of dThd paralleled therecovery of DNA synthesis, increasing after TIA and then returning to endogenous levels withint 4 hr. During intervals betweenTIA injections, when dThd levels approximated endogenouslevels, purine concentrations remained elevated, creating thecondition for unbalanced growth; and (c) when clonogenic tumorcells were measured by bioassay, the rate of tumor regrowthduring TIA rescue was reduced, suggesting that additional tumorkill occurred. In contrast, regrowth of peritoneal tumor cellsduring CF rescue paralleled tumor regrowth in untreated controlanimals.

It is unclear why normal tissue is not adversely affected bythymineless intervals during TIA rescue. Based on observationof cell kinetic parameters, it would be predicted that host cellswould be subject to the toxic actions of MTX during the periodicthymineless intervals. However, any toxic effects which mayoccur did not significantly affect the host. This observation isimportant, because it does not follow a predicted pattern ofresponse. If, indeed, normal cells are selectively spared duringthymineless intervals, this would have important implications inthe design of future rescue protocols.

The findings presented in this study demonstrate that theincrease in therapeutic index of MTX results from an enhancedtumor kill prior to rescue and requires a dose of MTX above athreshold level. In addition, nucleoside rescue had some therapeutic advantage over CF when given at periodic intervals. Thisobservation may have clinical implications; however, cautionusing nucleoside rescue in humans must be followed becauseof potential toxicities. Although no serious toxicities were observed in animals rescued with TIA, we did find that recovery of

JUNE 1984 2283

on May 23, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

L. L Samuels and J. A. Straw

nucleated bone marrow Å“il growth (Chart 2) was delayed, andthis could be significant when the limiting toxicity is bone marrow,as it is in humans.

REFERENCES1. Berlino, J. R. 'Rescue* techniques in cancer chemotherapy: use of leucovorin

and other rescue agents after methotrexate treatment. Semin. OncoL 4:302-

316,1977.2. Bruce, W. R., Meekers, B. E., and Valeriote, F. A. Comparison of the sensitivity

of normal hematopoetic and transplanted lymphoma colony-forming cells tochemotherapeutic agents administered in vìvo.J. Nati. Cancer Inst., 37: 233-

245.1966.3. Burton, K. A study of the conditions of the diphenylamine reaction for the

colorimetrie estimation of deoxynbonucleic acid. Biochem. J. 62: 315-323,

1956.4. Capizzi, R. S., DeConti. R. C., Marsh, J. C.. and Berlino, J. R. Methotrexate

therapy of head and neck cancer improvement in therapeutic index by use ofleucovorin "rescue." Cancer Res., 30:1782-1788,1970.

5. Cohen, S. S. One the nature of thymineless death. Ann. NY Acad. Sci., 86:292-301,1971.

6. Goldin. A., Mantel, N , Greenhouse, S. W., Venduti. J. M., and Humphreys, S.R. Estimation of the antileukemic potency of the antirnetabolite aminopterin,administered alone and in combination with citrovorum factor or folk; acid.Cancer Res., 13: 843-850,1953.

7. Goldin, A., Mantel, N , Greenhouse, S. W., Vendetti, J. M., and Humphreys,S. R. Effect of delayed administration of citrovorum factor on the antileukemiceffectiveness of aminopterin in mice. Cancer Res., 14: 43-48,1954.

8. Goldin, A., Venduti. J. M., Kline, I., and Mantel, N. Eradication of cell (L1210)by methotrexate and methotrexate plus citrovorum factor. Nature (Lond.), 272:1548-1550,1966.

9. Hakala, M. T. Prevention of toxicity of amethopterin for Sarcoma 180 cells intissue culture. Science (Wash DC), 726: 255,1957.

10. Hakala, M. T., and Taylor, E. The ability of purine and thymine derivatives andof glyoine to support the growth of mammalian cells in culture. J. Biol. Chem.,234:126-128,1969.

11. Harrap. K. R., Taylor, G. A., and Browman, G. P. Enhancement of thetherapeutic effectiveness of methotrexate and protection of normal proliferating

tissues with purines and pyrimidmes. Chem.-Bioi. Interact., 78:119-128,1977.

12. Harrap, K. R., Taylor, G. A., Talbot, D. C., Calvert, A. H., Stock, J. A., Stringer,M., Browman, G. P., Dady, P. J., and Cobley, T. Purine and pyrimidine"salvage" and its significance in methotrexate therapy. In: W. Davis and K. R.

Harrap (eds), Advances in Tumor Protection, Detection, and Characterization,Vol. 4, Characterization and Treatment of Human Tumors, pp. 93-105. Amsterdam: Excerpta Medica, 1978.

13. Jaffe, N., Frei, E., Ili, Traggis. D., Cassidy, J. R., Watts, H., and Filler, R. M.High-dose methotrexate with citrovorum factor in osteogenic sarcoma: Progress Report II. Cancer Treat. Rep., 67: 675-679,1977.

14. Jaffe, N.. Fret, E., Ill, Watts, H., and Traggis, D. High-dose methotrexate inosteogenic sarcoma: a 5-year experience. Cancer Treat. Rep., 62: 259-264,1978.

15. Johns, D. G. and Benino, J. R. Folate antagonists, in: J. F. Holland and E. FreiIII (eds.), Cancer Medicine, pp. 739-754. Philadelphia: Lea and Febiger, 1973.

16. Pinedo. H. M., Chabner. B. A., Zaharko, D. S., and Bull, J. M. Evidence forearly recruitment of granulocyte precursors during high-dose MTX infusions inmice. Blood, 48: 301-307,1976.

17. Rustum, Y. M. High-pressure liquid chromatography. L. Quantitative separationof purine and pyrimidine nucleosides and bases. Anal. Biochem., 90:289-299,1978.

18. Samuels, L. L. Possible mechanisms for the improved therapeutic activity ofhigh-dose methotrexate (HDMTX) with rescue. Proc. Am. Assoc. Cancer Res.,27: 273, 1980.

19. Samuels, L. L., and Straw, J. A. Effect of citrovorum factor (CF) and thymidine/inosine/allopurmol (TlA) rescue on antitumor activity of methotrexate (MTX).Proc. Am. Assoc. Cancer Res., 22:214,1981.

20. Sirotnak, F. M., Moccio, D., and Dorick, D. Optimization of high-dose methotrexate with leucovorin rescue therapy in L1210 leukemia and Sarcoma 180murine tumor models. Cancer Res., 38: 345-353,1978.

21. Skipper. H. E., Schabet, F. M., Jr., Mellen. L. B., Montgomery. J. A., Wilkoff,L. J., Lloyd. H. H., and Brockman, R. W. Implications of biochemical, cytoki-netic, pharmacologie, and toxicologie relationships in the design of optimaltherapeutic schedules. Cancer Chemother. Rep. 54:431-449,1970.

22. Skipper, H. E., Schabel, F. M., Jr., and Wilcox. W. S. Experimental evaluationof potential anticancer agents. XIII. On the criteria and kinetics associated with"curability" of experimental leukemia. Cancer Chemother. Rep. 35: 1-111,

1964.23. Straw, J. A., Talbot. D. C., Taylor, G. A., and Harrap, K. R. Some observations

on the reversibility of methotrexate toxicity in normal proliferating tissues. J.Nati. Cancer Inst., 58:91-97,1977.

2284 CANCER RESEARCH VOL. 44

on May 23, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

1984;44:2278-2284. Cancer Res   Lawrence L. Samuels and James A. Straw  Factor Rescue in MiceComparison of Thymidine-Purine Rescue with Citrovorum Improved Therapeutic Index with High-Dose Methotrexate:

  Updated version

  http://cancerres.aacrjournals.org/content/44/6/2278

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/44/6/2278To request permission to re-use all or part of this article, use this link

on May 23, 2018. © 1984 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from