comparison of enzyme-linked immunosorbent assays. J Clin Microbiol 29:1635-1639, 1991 (24) KRENITZ W: Uber das Auftreten von Mageninhalt bei Carcinoma Ventriculi. Dtsch Med Wochenschr 22:872,1906 (25)

LAMBERT JR,

HANSKY J, EAVES ER,

ET AL:

Campylobacter-\]kt organisms (CLO) in human stomach. Gastroenterology 88:1463, 1985 (26)

JIANG SJ, U U WZ, ZHANG DZ, ET AL:

Cam-

pylobacrer-Yike organisms in chronic gastritis, peptic ulcer, and gastric carcinoma. Scand J Gastroenterol 22:553-558,1987 (27)

VON WULFFEN H, GROTE HJ, GATERMANN S, ET

(31)

MORRIS AJ,

ALJ MR,

NICOLSON GI, ET AL:

Long-term follow-up of voluntary ingestion of Helicobaclerpylori. Ann Intern Med 114:662663,1991 (52)

FORMAN D, NEWELL DG, FULLERTON F, ET AL:

Association between infection with Helicobacter pylori and risk of gastric cancer Evidence from a prospective investigation. BMJ 302:1302-1305,1991

Lack of Ranitidine Effects on Cyclophosphamide Bone Marrow Toxicity or Metabolism: A PlaceboControlled Clinical Trial David S. Alberts * Nancy Mason-Liddil, Patricia M. Plezia, Denise J. Roe, Robert T. Dorr, Robert F. Struck, J. Gregory Phillips

We previously reported that cimetidine but not ranitidine significantly enhances cyclophosphamide-induced bone marrow toxic effects and the appearance of cyclophosphamide alkylating species in a murine leukemia mouse model, and we advised caution in the use of cimetidine with microsomally metabolized anticancer drugs. Both drugs have been used for the treatment Vol. 83, No. 23, December 4, 1991

Cyclophosphamide is a commonly used cytotoxic drug with useful clinical activity against a variety of human can-

cers. It must be metabolized in vivo to its alkylating species to exhibit antitumor activity in animals and humans (7,2). The initial step of the complex metabolic process is mediated by hepatic microsomal mixed-function oxidases linked with cytochrome P-450, which are involved in the metabolism of numerous noncytotoxic drugs (3-7). Cimetidine and ranitidine are H2 histamine antagonists used extensively in the management of acute and chronic gastric and duodenal ulcers and in pathological hypersecretory conditions. Cimetidine is known to bind to the cytochrome P-450 enzyme complex and thereby inhibit the metabolism of numerous commonly used drugs (8-14). Additionally, several more recent investigations have demonstrated that cimetidine can inhibit the clearance of various cytotoxic agents, including fluorouracil (75), doxorubicin (16), cyclophosphamide (17,18), and carmustine (79). Ranitidine, unlike cimetidine, does not appear to influence significantly the elimination of cytotoxic drugs metabolized by the mixed-function oxidase system in the liver (18-20). We have reported previously that, in leukemic mice given cyclophosphamide, cimetidine but not ranitidine significantly enhances the appearance of alkylating species as well as cyclophosphamide-induced bone marrow toxic effects (17,18).

Received December 17, 1990; revised June 27, 1991; accepted September 10, 1991. Supported by Public Health Service grants CA17094, CA-23074, and CA-46951 from the National Cancer Institute, National Institutes of Health, Department of Health and Human Services; and by a grant from Glaxo Pharmaceutical Co., Research Triangle Park, N.C. D. S. Alberts, R. T. Dorr, Section of Hematology and Oncology, Department of Medicine and Department of Pharmacology, Arizona Cancer Center, College of Pharmacy, University of Arizona, Tucson, Ariz. N. Mason-Liddil, Arizona Cancer Center, University of Arizona. P. M. Plezia, Section of Hematology and Oncology, Department of Medicine, and Arizona Cancer Center, College of Medicine and the Department of Pharmacy Practice, University of Arizona. D. J. Roe, Department of Family and Community Medicine, Arizona Cancer Center, University of Arizona. R. F. Struck, J G. Phillips, Southern Research Institute, Birmingham, Ala. 'Correspondence to: David S. Alberts, M.D., Arizona Cancer Center, 1515 N. Campbell Ave., Tucson, AZ 85724.

REPORTS 1739

Downloaded from http://jnci.oxfordjournals.org/ at Florida Atlantic University on July 25, 2015

AU Immunoblot analysis of immune response to Campyiobaaer pylori and its clinical associations. JClin Pathol 41*53-659, 1988 (2S) BERKOWICZ J, LEE A: Person-to-person transmission of Campylobacter pylori (letter). Lancet 2:680-681, 1987 (29) ROTHMAN KJ: Modern Epidemiology. Boston: Little, Brown, 1986, pp 64-68 (50) MiETTlNEN O: Estimability and estimation in case-referent studies. Am J Epidemiol 103:226-235,1976

of gastric complications of chemotherapy. Using a randomized, doubleblind, crossover study design, we have now evaluated the potential interaction of ranitidine with cyclophosphamide in seven cancer patients, who received two courses of cyclophosphamide, one with ranitidine and one with placebo. Four patients received ranitidine in the first course, and three received placebo. Ranitidine or placebo was started 3 days before a single dose of cyclophosphamide and given for 17 consecutive days. Ranitidine or placebo was given orally (300 mg/d), and cyclophosphamide (600 mg/m2) was given intravenously with [3H]cyclophosphamide (1000 jiCi). Cyclophosphamide treatment was repeated at 4 weeks plus or minus 4 days. Blood samples were collected at intervals from 5 minutes to 24 hours after cyclophosphamide treatment and analyzed by thin-layer chromatography and radioassay for the drug and its metabolites. On days 0, 7, 14, and 21 after cyclophosphamide administration, complete blood cell counts, white blood cell differential counts, platelet counts, and SMA-17 were determined. The differences in mean nadir white blood cell counts, granulocyte counts, hemoglobin levels, and hematocrit values during ranitidine versus placebo treatment were not statistically significant In a statistical but not a clinical sense, mean nadir platelet counts were significantly lower with ranitidine. There was a statistically significant increase in area under the curve for drug concentration in plasma x time (AUC) with ranitidine as well as a statistically significant decrease in the total-body clearance rate of the cyclophosphamide molecule. However, the effect on AUC for the major oncolytic metabolites 4-hydroxycyclophosphamide and phosphoramide mustard was not statistically significant The lack of toxicologic or metabolic interaction between ranitidine and cyclophosphamide suggests that ranitidine can be used safely with cyclophosphamide. [J Natl Cancer Inst 83:1739-1743,1991]

Stress ulceration, gastritis, and peptic ulcer disease are common and debilitating complications in patients receiving cancer chemotherapy. Thus, H2 histamine antagonists are commonly prescribed for these patients, both as therapy and prophylaxis (21). Since cimetidine but not ranitidine appears to affect the metabolism of cyclophosphamide to its active alkylating species in mice, caution has been advised concerning the administration of cimetidine but not ranitidine with microsomally metabolized anticancer agents (17,18). In this study, we have evaluated the potential interaction of ranitidine with cyclophosphamide in cancer patients.

Patient selection. Criteria for entry in the study were histopathological diagnosis of advanced metastatic cancer, subject aged 18 years or older, normal hepatic and renal functions, current intravenous treatment with cyclophosphamide, and no administration of other cytotoxic agents within 28 days prior to administration of this alkylating agent. Study design. The study employed a randomized, double-blind, crossover design with seven patients receiving cyclophosphamide while receiving either ranitidine or placebo. Patients were randomly assigned to receive their first course of cyclophosphamide with either ranitidine or placebo, which was administered starting 3 days before a single dose of cyclophosphamide and was continued for 17 consecutive days. Ranitidine at a dose of 300 mg/d was given orally at 10 PM each day, and 600 mg/m2 of cyclophosphamide mixed with 1000 (J.Ci of [3H]cyclophosphamide in 125 mL of normal saline was administered intravenously for a 15-minute period at approximately 10 AM on day 4 of the course. The cyclophosphamide dose was repeated at 4 weeks plus or minus 4 days. During the 17-day period of ranitidine or placebo treatment, concomitant medications remained the same, and the only antiemetic allowed prior to cyclophosphamide administration was a fixed dose ofdexamethasone. Clinical laboratory studies. Complete blood cell counts, white blood cell differential counts, platelet counts, and

1740

plasma of cyclophosphamide and of metabolite x time (AUC). When plasma concentration levels were greater than 0 after 24 hours, the AUC from 24 hours to infinity was computed using the terminal slope of the curve for plasma concentration versus time. Curves were plotted for cyclophosphamide, 4-hydroxycyclophosphamide (as aldophosphamide semicarbazone), and phosphoramide mustard from appropriate plasma level values ± SD (obtained from triplicate analyses of each plasma sample). Paired / tests were used to compare data for patients treated with ranitidine and patients treated with placebo, since each patient served as his or her own control.

Results Clinical Toxicity Data Toxicity data were obtained for seven patients during two courses of cyclophosphamide, one with placebo and the other with ranitidine treatment. Four patients received ranitidine and three received placebo during the first course of cyclophosphamide therapy. The statistical analysis compared the nadir values for platelet, white blood cell, and granulocyte counts, hemoglobin levels, and hematocrit values obtained during the two cyclophosphamide courses with ranitidine with those obtained during the two cyclophosphamide courses with placebo, using a paired / test. The results are shown in Table 1. Nadir white blood cell, granulocyte, and platelet counts occurred on approximately day 21. The mean nadir white blood cell counts, granulocyte counts, hemoglobin levels, and hematocrit values observed during the ranitidine treatment were not significantly different from those observed during the placebo treatment (P>.\ for each comparison). The mean nadir platelet count during ranitidine administration (231 OOO/fiL), however, was significantly lower (P = .03) than that during placebo administration. Because of the large number of individual blood chemistries (SMA-17), statistical comparison was not appropriate. Nevertheless, there were no obvious changes in the mean values for these 17 blood chemistry parameters between the first and second courses of cyclophosphamide administration. No patient had

Journal of the National Cancer Institute

Downloaded from http://jnci.oxfordjournals.org/ at Florida Atlantic University on July 25, 2015

Methods

SMA-17 (semiautomated chemistry panel) were obtained just prior to each dose of cyclophosphamide. Complete blood cell counts, white blood cell differential counts, and platelet counts were also obtained on days 7, 14, and 21 following cyclophosphamide dosing. Pharmacokinetics and metabolism studies. Blood (10 mL) was collected from an indwelling needle or catheter at various times from 5 minutes to 24 hours after cyclophosphamide treatment, mixed immediately with 1 mL of 0.4 M semicarbazide (pH 7.4) in capped tubes, and centrifuged (2000 rpm for 3 minutes). Plasma was separated, immediately frozen, and stored at -20 *C until analysis. Extraction of cyclophosphamide and/or 4-hydroxycyclophosphamide, stabilized as aldophosphamide semicarbazone, was accomplished with chloroform as described previously by Struck et al. (22). The extracted plasma was lyophilized, and the residue was triturated with methanol. Centrifugation yielded a supernatant, which was treated with excess ethereal diazomethane to stabilize and isolate phosphoramide mustard as its methyl ester. These extraction procedures previously have been reported to result in cyclophosphamide, 4-hydroxycyclophosphamide, and phosphoramide mustard recoveries from plasma of 85% or greater (22^3). The two plasma extracts, containing aldophosphamide semicarbazone or phosphoramide mustard methyl ester, were fractionated separately in triplicate by thin-layer chromatography with appropriate synthetic standards as reported previously (522). Briefly, this procedure involved the co-chromatography with cyclophosphamide and aldophosphamide semicarbazone for chloroform extracts and with the methyl ester of phosphoramide mustard for the methanol extracts. Detection of synthetic standards was accomplished by use of 4-(/>-nitrobenzyl)pyridine, which allows visualization of the alkylating species as blue spots. Individual alkylating spots were then collected by scraping the thin-layer chromatography plates and were transferred to scintillation vials, treated with 1 mL methanol, and radioassayed to quantitate the parent drug and metabolites. Pharmacokinetic analysis. The linear trapezoidal rule was used to determine area under the curve for concentrations in

Table 1. Hematologic toxicity in seven patients treated with cyclophosphamide plus ranitidine or cyclophosphamide plus placebo* Pretreatment Blood cell count, Platelets White blood cells Granulocytes

Ranitidine nadir Placebo nadir

453.601189.71 231.00168.42 275.29194.27 9.2811.07 6.4511.69

3.2112.37 1.7011.95

3.4112.26 1.7411.57

Hemoglobin, g/dL

12.9011.47

11.1011.30

11.2311.00

Hematocrit,*

38.7714.84

32.7013.48

33.5913.13

Difference in nadir

P value

-44.29142.40' -0.2012.21 -0.041 1.72 -0.1310.79 -0.8911.38

.03 .82 .95 .68 .14

•Unless otherwise indicated, values = means 1SD for seven patients.

evidence of significant hepatic or renal impairment.

Fig. 1. Plasma concentration versus time curves for all courses of cyclophosphamide administered with ranitidine or with placebo in seven cancer patients. Values for data points are means 1 SD. Wide SE bars refer to curve for cyclophosphamide plus ranitidine; narrow bars refer to curve for cyclophosphamide plus placebo.

Vol. 83, No. 23, December 4, 1991

Discussion There are three important results of this double-blind, randomized, placebo-controlled, crossover clinical trial. 1) Ranitidine administration did not change the pattern or degree of cyclophosphamideinduced leukopenia or granulocytopenia. 2) Ranitidine administration had no significant effect on the AUC values for the two major oncolytic cyclophosphamide metabolites 4-hydroxycyclophosphamide and phosphoramide mustard. 3) Ranitidine administration was associated with significantly prolonged plasma terminalphase half-life and increased AUC for the parent (unchanged) drug. These results are consistent with those we reported earlier in a murine tumor model. In that study, cimetidine but not ranitidine increased cyclophosphamideinduced cytotoxicity and the AUC values for cyclophosphamide metabolites {18). Since the parent compound cyclophosphamide must be converted to 4-hydroxycyclophosphamide and phosphoramide mustard to express its cytotoxicity (/), a change in the pharmacokinetics of the parent compound without a corresponding change in the pharmacokinetics of its metabolites would not be expected to result in a corresponding change in the cytotoxicity of cyclophosphamide in vivo. The data on the mean plasma AUC shown in Table 3 do reveal a 35.1% increase in the mean plasma AUC of the cyclophosphamide metabolite 4-hydroxycyclophosphamide and a 15.9% increase in the AUC of phosphoramide mustard REPORTS 1741

Downloaded from http://jnci.oxfordjournals.org/ at Florida Atlantic University on July 25, 2015

patients (Table 2). Additionally, the totalbody clearance (CI-TB) of cyclophosphamide was decreased in six of seven Pharmacokinetic Data patients during ranitidine exposure. As Plots of the mean data on AUC for all shown in Tables 2 and 3, there was apcourses of cyclophosphamide plus raniti- preciable interpatient variation in cyclopharmacokinetics, but dine and cyclophosphamide plus placebo phosphamide considerably less intrapatient variation. are shown in Fig. 1. Pharmacokinetic data Ranitidine administration significantly infor the parent cyclophosphamide comcreased the mean half-life from 6.26 to pound and its two oncolytic metabolites, 10.65 hours (P = .0344; 67% average in4-hydroxycyclophosphamide and phoscrease), increased the mean plasma phoramide mustard, are shown in Tables AUCo_ from 157.24 to 264.45 (Mg/mL) x 2 and 3. The values for cyclophosphahour (P = .0242; 70% average increase), mide terminal-phase half-life in plasma (h/fii) and for AUC for plasma concentra- and decreased the mean plasma clearance tion of drug x time from time of adminis- from 7.42 to 5.19 L/h (P = .0254; 29% tration to infinity (AUQ^.) were greater average decrease). There were no difwith ranitidine administration in all seven ferences in apparent volume of distribu-

tion at steady state (VDss) for cyclophosphamide in the presence or absence of ranitidine administration. In contrast to the effects of ranitidine administration on cyclophosphamide pharmacokinetics, there appeared to be no significant effect of ranitidine treatment on the plasma AUC0.12 of the oncolytic cyclophosphamide metabolites 4-hydroxycyclophosphamide and phosphoramide mustard (Table 3). Because of their relatively low concentrations and transient presence in plasma after cyclophosphamide administration, it was not possible to model mathematically the pharmacokinetics (e.g., plasma ty$ and CI-TB) of these important cyclophosphamide metabolites.

Table 2, Pharmacolunetic data on cyclophosphamide and metabolites in plasma from seven patients treated with cyclophosphamide plus placebo or cyclophosphamide plus ranitidine* Cyclophosphamide Patient No.

Ranitidine administration

AUCo-12, (HE/mL) x h

Cyclophosphamide

AUCo~, (Ug/mL) x h

4-Hydroxycyclophosphamide

Phosphoramide mustard

VDss, L

12.6

165.3 171.2

0.59 0.35

0.49 0.42

NA 93.5

6.3 6.8

2

5.6 7.1

123.5 178.0

0.40 1.06

0.62 0.66

81.9 62.3

9.2 8.5

3

4.3 5.9

78.7 160.3

0.28 0.34

0.56 0.33

65.1 43.4

9.9 5.1

4

6.1 7.5

101.8 132.2

0.17 0.32

0.20 0.30

84.9 90.5

11.9 7.7

5

7.6 21.1

201.2 423.7

0.25 0.39

0.35 0.71

43.6 67.4

5.6 29

6

5.0 9.3

160.5 412.4

0.54 0.54

0.56 0.51

43.7 20.2

5.9 2.7

7

6.3 11.1

269.7 373.4

0.33 0.54

0.32 0.63

NA 37.0

3.2 27

67 ±55

70 ±55

9.0

1

-29 ±24

*- = without ranitidine; + = with ranitidine; NA = data not available. For definition of pharmacoldnetic parameters, see "Results" section. tValues = means ± SD.

Table 3. Mean pharmacoldnetic data for seven patients treated with cyclophosphamide plus ranitidine or cyclophosphamide plus placebo* Cyclophosphamide + ranitidine

Cyclophosphamide + placebo

Difference

P value

Cyclophosphamide

tvfi,h AUCo~, (ng/mL) x h VDss, L O r e , L/h

10.65 ±5.15 264.45 ± 131.41 59.18 ±27.35 5.19±2.50

6.26 157.24 60.74 7.42

±1-58 ±64.62 ±21.47 ±3.01

4.39 107.21 -1.56 -2.23

±4.26 ±94.70 ±19.79 ±1.99

.0344 .0242 .84 .0254

4-Hydroxycyclophosphamide: AUCo-12, (Hg/mL) x h

0.50 ±0.26

0.37 ±0.15

0.13 ±0.27

.23

Phosphoramide mustard: AUCo-i2,(ng/mL)x h

0.51 ±0.17

0.44 ±0.15

0.07 ±0.21

.43

•Unless otherwise indicated, values = means ± SD for seven patients. For definition of pharmacokinetic parameters, see "Results" section.

with administration of ranitidine. Although these increases do not represent statistically significant changes, with a sample size of only seven, we would have been able to detect only a doubling of the placebo-associated mean plasma AUC values of these two active cyclophosphamide metabolites. Thus, the relatively small patient sample size of this study may not have allowed us to determine a statistically significant change in the plasma AUC values of 4-hydroxycyclophosphamide and/or phosphoramide mustard as a result of ranitidine administration. Although ranitidine administration was associated with cyclophosphamide-induced platelet count nadirs that were significantly lower than those observed in

1742

patients treated with placebo, the mean nadir count was only 231 000 ± 68 420/fiL. These slight reductions in platelet counts were, of course, clinically insignificant. Thus, concurrent administration of ranitidine does not appear to affect the toxicity of cyclophosphamide. Brenner et al. (16) and Harris et al. (20) have reported that both cimetidine and ranitidine administered prior to doxorubicin caused a significant prolongation of the plasma half-life of the parent compound, an increase in plasma AUC, and a decrease in total-body clearance. Despite these cimetidine- and ranitidine-induced changes in the pharmacokinetics of the parent compound doxorubicin, neither cimetidine nor

ranitidine pretreatment had a significant effect on the plasma AUC of doxorubicin's major metabolites, doxorubicinol and 7-deoxydoxorubicinol aglycone. Ranitidine pretreatment may alter cyclophosphamide's plasma half-life and clearance through H2 receptor antagonism and a subsequent decrease in hepatic blood flow. Reduced hepatic clearance of cyclophosphamide secondary to decreased hepatic blood flow could lead to a prolongation of cyclophosphamide's plasma half-life as well as an increased AUC value, as has been documented previously for cimetidine's effects on propranolol pharmacokinetics (24-26). As pointed out by Brenner et al. (16), hepatic blood flow changes induced by cimetidine or ranitidine H2 receptor antagonism mainly would affect the pharmacokinetics of drugs that are highly extracted by the liver. Although cyclophosphamide fits into this category, a ranitidine-induced reduction in hepatic blood flow would still not explain the absence of ranitidineassociated changes in cyclophosphamide metabolism as documented in the present study. It is well established that cimetidine can bind to the cytochrome P-450 enzyme complex in hepatic cells and thereby inhibit the metabolism of other drugs, including warfarin, theophylline, propranolol, diazepam, chlordiazepoxide, phen-

Journal of the National Cancer Institute

Downloaded from http://jnci.oxfordjournals.org/ at Florida Atlantic University on July 25, 2015

% changet

CITB, L/h

References

(8) SERLJN MJ, SIBEON RG, MOSSMAN S, ET AL:

Cimetidine: Interaction with oral anticoagulants in man. Lancet 2:317-319, 1979 (9) O'REILLY RA: Comparative interaction of cimetidine and ranitidine with racemic warfarin in man. Arch Intern Med 144:989-991, 1984 (10) HEAGERTY AM, DONOVAN MA, CASTLEDEN

CM, ET AL; Influence of cimetidine on pharmacokinetics of propranolol. Br Med J [Clin Res] 282:1917-1919,1981 (11) KIRCH W, SPAHN H, KOHLERH, ETAL: Interac-

tion of metoprolol, propranolol and atenolol with concurrent administration of cimetidine. Klin Wochenschr 60:1401-1407, 1982 (12) HEAGERTY AM, DONOVAN MA, CASTLEDEN

CM: The influence of histamine (Hi) antagonists on propranolol pharmacokinetics. Int J Clin Pharmacol Res 2:203-205,1982 (13) NEUVONEN

PJ,

TOPKOLA

RA,

KASTE

M:

Cimetidine-phenytoin interaction: Effect on serum phenytoin concentration and antipyrine test.EurJClinPharmacol21:215-220, 1981 (14) BARTLE WR, WALKER SE, SHAPERO T: Dose-

dependent effect of cimetidine on phenytoin kinetics. Clin Pharmacol Ther 33:649-655, 1983 AL: The influence of cimetidine on the pharmacokinetics of 5-fluorouracil. Br J Clin Pharmacol 18:421-430,1984 (16) BRENNER DE, COLLINS JC, HANDE KR: The ef-

fects of cimetidine upon the plasma pharmacokinetics of doxorubicin in rabbits. Cancer Chemother Pharmacol 18:219-222, 1986 (17) DORR RT, ALBERTS DS: Cimetidine enhance-

ment of cyclophosphamide antitumour activity. Br J Cancer 45:35-43, 1982 (18) DORR RT, SOBLE MJ, ALBERTS DS: Interaction

of cimetidine but not ranitidine with cyclophosphamide in mice. Cancer Res 46:17951799,1986 (19) DORR RT, SOBLE MJ: Hj-Antagonists and car-

mustine. J Cancer Res Clin Oncol 115:41-46, 1989 (20) HARRIS NL, BRENNER DE, ANTHONY LB, ET

AL: The influence of ranitidine on the pharmacokinetics and toxicity of doxorubicin in rabbits. Cancer Chemother Pharmacol 21:323-328, 1988

Cimetidine: Prophylaxis against upper gastrointestinal haemorrhage after renal transplanvation of cyclophosphamide in man and tation. Br Med J 1:398-400, 1978 animals. Cancer 27:1512-1529, 1971 (2) SLADEK NE: Therapeutic efficacy of cyclo- (22) STRUCK RF, ALBERTS DS, HORNE K, ET AL: phosphamide as a function of its metabolism. Plasma pharmacokinetics of cyclophosphaCancer Res 32:535-542, 1972 mide and its cytotoxic metabolites after intravenous versus oral administration in a (3) STRUCK RF, ALBERTS DS: Blood levels of randomized, crossover trial. Cancer Res alkylating metabolites of cyclophosphamide in 47:2723-2726,1987 the mouse after iv or oral administration. Cancer Treat Rep 68:765-770, 1984 (23) STRUCK RF: Isolation and identification of a stabilized derivative of aldophosphamide, a (4) CONNORS TA, COX PJ, FARMER PB, ET AL: major metabolite of cyclophosphamide. CanSome studies of the active intermediates cer Res 34:2933-2935,1974 formed in the microsomal metabolism of cyclophosphamide and isophosphamide. Bio(24) FEELY J, WILKINSON GR, MCALLISTER CB, ET chem Pharmacol 23:115-129,1974 AL: Increased toxicity and reduced clearance of lidocaine by cimetidine. Ann Intern Med (5) STRUCK RF, KIRK MC, MELLETT LB, ET AL: 96:592-594, 1982 Urinary metabolites of the aniitumor agent cyclophosphamide. Mol Pharmacol 7:519-529, (25) KIRCH W, KOHLER H, SPAHN H, ET AL: Interac1971 tion of cimetidine with metoprolol, propranolol, or atenolol. Lancet 2:531-532, 1981 (6) BAGLEY CM JR, BOSTICK FW, DEVFTA VT JR: Clinical pharmacology of cyclophosphamide. (26) REIMANN IW, KLOTZ U, SIEMS B, ET ALJ Cancer Res 33:226-233,1973 Cimetidine increases steady state plasma levels of propranolol. Br J Clin Pharmacol (7) FRIEDMAN OM, WOWNSKY I, MYLES A: Cy12:786-790,1981 clophosphamide (NSC-26271)-related phosphoramide mustards—recent advances and historical perspective. Cancer Treat Rep 60:337-346,1976

Vol. 83, No. 23, December 4, 1991

Noriyuki Masuda, Masahiro Fukuoka* Kaoru Matsui, Yoko Kusunoki, Shinzoh Kudoh, Shunichi Negoro, Nobuhide Takifuji, Mamoru Fujisue, Hideo Morino, Kazuhiko Nakagawa, Masayuki Nishioka, Minoru Takada

(15) HARVEY VJ, SLEVTN ML, DILLOWAY MR, ET

(2/) JONES RH, RUDGE Cl, BEWICK M, ET AL: (/) BROCK N, GROSS R, HOHORST HJ, ETAL: Acti-

Establishment of Tumor Cell Lines as an Independent Prognostic Factor for Survival Time in Patients With Small-Cell Lung Cancer

We studied tumor samples from 39 patients, who entered our study from January 1989 to May 1990, to assess whether the ability to establish a continually growing tumor cell line from fresh tumor specimens can be associated with decreased survival times in patients with small-cell lung cancer. The tumor samples were used to establish cell lines in culture using a serum-free medium supplemented with hydrocortisone, insulin, transferrin, estrogen, and selenium (HITES). Thirty-three of these specimens were obtained by Fiberoptic bronchoscopy from primary sites during routine diagnostic procedures. A total of 11 (28%) cell lines were established: seven (21%) from 33 primary tumors and four (80%) from five peripheral lymph nodes. Survival times of the 11 patients whose tumor cell specimens continually grew in culture at any time during their clinical course were significantly shorter than those of the 28 patients whose tumor cell specimens did not grow in vitro (median survival time of 26 weeks versus 73 weeks; P = .0068). Cox's proportional hazards model, including sex, age, Eastern Cooperative Oncology Group performance status, stage, source of specimen, treatment, and in vitro tumor cell growth in the overall patient group, showed that cell line establishment (P = .0017) and no therapy (P = .0015) were the most important factors indicating poor survival time.

REPORTS 1743

Downloaded from http://jnci.oxfordjournals.org/ at Florida Atlantic University on July 25, 2015

ytoin, and metoprolol (8-14). We have documented previously that pretreatment with cimetidine but not ranitidine significantly enhances cyclophosphamide's antileukemic activity and bone marrow toxic effects by increasing plasma levels of its alkylating species (17,18). Since the same pharmacokinetic and toxicologic effects of cyclophosphamide may occur in cancer patients treated concomitantly with cimetidine, we have suggested caution with respect to the administration of this H2 blocking agent in the presence of cyclophosphamide. We would have preferred that the design for our pharmacokinetic and toxicologic trial involve a toxicologic and pharmacokinetic comparison of ranitidine versus cimetidine in combination with cyclophosphamide. Because of the unpredictable nature of a cimetidine-cyclophosphamide interaction, however, we decided that this risk was unwarranted in the setting of a clinical pharmacokinetic trial. Instead, we focused on a comparison of ranitidine with placebo in a controlled trial. The lack of a clinically important toxicologic or metabolic interaction between ranitidine and cyclophosphamide in cancer patients, documented by theresultsof this study, suggests that ranitidine can be safely administered to patients during therapy with cyclophosphamide.