- Email: [email protected]
Decreased incidence of spontaneous mammary gland neoplasms in female F344 rats treated with amphetamine, methylphenidate, or codeine
CANCER LETTERS
ELSEVIER
Cancer Letters 102 (1996) 77-83
Decreased incidence of spontaneous mammary gland neoplasms in female F344 rats treated with amphetamine, methylphenidate, or codeine June K. Dunnick*, Michael R. Elwell, Joseph K. Haseman Nutional Institute
of Environmental Health Sciences, PO. Box 12233, Research Triangle Park, NC 27709, USA
Received 26 December 1995; revision received 19 January 1996;accepted 19 January 1996
Abstract
Three drugs that affect the neuroendocrinesystem(amphetamine,methylphenidate,and codeine) causeddecreasesin body weights and in the incidence of spontaneouslyoccurring mammarygland neoplasmsin the female F344/N rat in 2-year carcinogenicity studies. Using a mathematical model that relates body weight changes to the incidence of mammary gland neoplasms, we find that the decrease in mammary gland tumors seen in female rats cannot be fully explained by body weight
decreasesrelative to control animals. Further, the observeddecreasesin body weight in treated female rats were not a function of differences in feed consumption between treated and control groups. These pharmaceuticalsare thought to affect the biologic system through interaction with membrane receptors. This interaction and/or subsequent cell signaling events may
play a role in the observed decreasein spontaneously occurring mammary gland neoplasmsin the female rat treated with amphetamine,methylphenidate, or codeine. Keywords:
Mammary gland neoplasm;Amphetamine; Methylphenidate; Codeine; F344 rats
1. Introduction Experimental studies in animals are used in conjunction with epidemiology studies to assess the potential long term toxicity, carcinogenicity, and/or beneficial effects of drugs. We have investigated the long-term effects of amphetamine sulfate, methylphenidate hydrochloride, and codeine in the carcinogenesis bioassay because these drugs are important in our armamentarium of therapeutic agents. Am-
* Corresponding author.
Elsevier Science Ireland Ltd. PII: SO304-3835(96)04168-7
phetamine is used in the treatment of narcolepsy, attention deficit hyperactivity disorders, and in weight control; methylphenidate in the treatment of narcolepsy and attention deficit hyperactivity disorders; and codeine in pharmaceuticals used as analgesics, sedatives, hypnotics, antiperistaltics, and antitussive agents [ 11. Amphetamine sulfate, methylphenidate hydrochloride, or codeine administered in the feed to F344 rats and B6C3Fl mice for 2 years did not show sitespecific carcinogenic activity except some carcinogenic activity in the liver of mice receiving methylphenidate [2-61. In our bioassay we also found that these chemicals
78
J.K. Dunnick et al. I Cancer Letters 102 (1996) 77-83
Table 1
J’w
Sex and species
Dose, mgkg per daya (ppm in feed)
Level of carcinogenic activity
Decreasesin spontaneous neoplasms
Amphetamine sulfate
Male rats
0, 0.9, 5
No evidence
Adrenal pheochromocytoma (23/49, 15/44,6/50); pituitary gland adenoma(15/49, 15/48,
0, 1.1.5.2 (0,20, 100)
No evidence
9/49)
0, 3.8,31.6 to, m1w
No evidence
to, 20, 100) Female rats
Male mice
Mammary gland fibroadenoma/ adenoma/ adenocarcinoma(25/50, 12/50, 3/50); pituitary gland adenoma (31/50,24/48, 19/50); uterus, endometrial stromal polyp (10/50,6/50, 3/50)
Female mice
0, 3, 18.5 to, 20, 100)
No evidence
Harderian gland adenoma (4/50,2/50,0/50); lung adenomakarcinoma (S/50, 3/30, 4/50); liver adenoma/ carcinoma (14/50, 12/50, 2/50) Pituitary gland adenoma (12/49,6/49, l/46); Harderian gland adenoma(5/50,2/50, O/47);
Lung adenomakarcinoma (S/50,6/50, l/47); liveradenomakarcinoma (5/50, l/50, l/47) Male rats
Methyl phenidate hydrochloride
Female rats
0.4, 20,42(0, 100, 500,lOOO) 0,4, 22,47 (0, 100,
No evidence No evidence
500,1000)
Adrenal pheochromocytoma (18/49,7/48, 5/49, 10/50) Mammary gland tibroadenoma/ adenoma/ adenocarcinoma(16/49, 17/50. 8/48,5/50)
Malemice
Female mice Male rats
Codeine
Female rats
Male mice
0,5,28,56(0,50, 250,500) 0, 7, 34, 67 (0, 50, 250,500)
Someevidence (liver tumors) Someevidence (liver tumors)
0, 12,27,64 (0,400,800, No evidence 1600) 0, 15, 31,63 (0,400, 800, No evidence 1600)
0.72, 159, 342(0, 750,1500,3000)
No evidence
0,77, 159, 384(0, 750,1500,3000)
None
Adrenal pheochromocytoma (16/49,6/50,6/50,
3/50)
Mammary gland tibroadenoma/adenoma/adenocarcinema (30/.50,23/50, 29/51, 8/51) Liver adenomakarcinoma (29/50, 29/50, 23150, 16/50);
‘w Female mice
None
No evidence
malignant lymphoma (7/50, 1150,l/50,0/50) Liver adenomakarcinoma (16/50, 15/51, 15/51, 8/50)
J.K. Dunnick et al. I Cancer Letters IO2 (1996) 77-83
79
decreased the incidences of some spontaneously occurring tumors. In this paper we analyze the decrease in spontaneous tumors using a mathematical model that relates body weight changes to the incidence of mammary gland neoplasms, and discuss the significance of our findings.
the results of 13-week studies and were chosen to provide a minimally toxic high dose and lower doses within 10x the human dose based on mg drug/m2 body surface area per day [7]. The doses for each of the chemicals and tumor incidences are listed in Table 1.
2. Methods and materials
2.2. Mathematical model for analyzing decreases in tumor incidence
2.1. Curcinogenesis bioassay The relationship between site-specific tumor incidence and body weight in control animals was examined using a mathematical model derived from individual animal data from approximately 100 NTP bioassays in the F344 rat and the B6C3Fl mouse [8]. The derived formula uses a logistic model that provides an excellent prediction of tumor incidence based on an animal’s age at death and body weight at 52 weeks. We have applied this model to the analysis for decreases in mammary gland and liver neoplasms: For mammary tumors: probability of tumor for an
Amphetamine sulfate, methylphenidate hydrochloride, or codeine were studied in separate carcinogenesis bioassays. Drugs were administered in the diet to groups of male and female F344LN rats and B6C3Fl mice for periods of up to 2 years, at which time the animals were killed and a complete gross and histopathologic tissue evaluation performed. Each group contained 50 animals per species/sex with an additional 10 animals per species/sex added for interim evaluation at 15 months [2-4]. The doses selected for the 2-year studies were based on Table 2 Survival and mean body weight at 2 years Drug
Final mean body wt. (g)
% Survival at 2 years L
M
H
C
L
M
H
Amphetamine sulfate 60 Male rats 66 Female rats
62 84
-
66 14
461 349
424** 309**
-
391** 243**
96 70
96 12
-
98 88*
43 42
31*+ 34**
-
31** 27**
66 64 90 70
68 12 88 74
68 78 82 88
391 317 46 44
400 302 44 43
372 276** 40** 41
353** 252** 42* 42
40 76+ 76 72
42 58 90 86
40 64 86 70
411 342 48 50
405 335 51 51
390 315** 44* 49
362** 303** 42** 40**
Ca
Male mice Female mice Methylphenidate Male rats Female rats Male mice Female mice Codeine Male rats Femaie rats Male mice Female mace
hydrochloride 56 62 90 74
58 56 82 72
*P-c 0.05 versus control. **P < 0.01 versus control. aDoses as listed in Table 1; C, control; L, low dose; M, mid dose; H, high dose.
80
J. K. Dunnick et al. I Cancer Letters 102 (1996) 77-83
individual animal = I/( 1 + e-z), z = a + b(52 week body weight) + ~(52 week body weight)2 + d (survival in days). For female rat mammary gland: a = 8.68, b = 0.0162, c = 0, d = 0.0055. For liver tumors: probability of tumor for an individual animal = l/( 1 + e-z), z = a + b(52 week body weight) + ~(52 week body weight)2 + d(surviva1 in days). Individually housed mice: for male mouse liver: a = -13.15, b = 0, c = 0.00152, d = 0; for female mouse liver: a = -7.09, b = 0, c = 0.00115, d = 0.00552. Group housed female mouse liver: a = 7.34; b = 0; c =-0.00078; d = 0.00576. The parameters a, b, c, and d are estimated by lo-
gistic regression techniques for each site-specific tumor, using individual animal data from the NTP historical control database. 3. Results and discussion There was no evidence for treatment-related mortality in rats or mice after 2 years of administration of the drugs (Table 2). None of these drugs were mutagenic in the Salmonella assay, and the only one that exhibited carcinogenic activity was methylphenidate, which caused an increase in the incidence of liver tumors in mice [6].
Table 3 Estimated feed and drug consumption in the female F344 rat Ca
L
M
H
Amphetamine .sulfate Dose @pm) Feed consumption (51 weeks)b Compound consumption (5 1 weeks)c Feed consumption (102 weeks) Compound consumption (102 weeks) Estimated dose (mg/m2)d Estimated human dose (mg/kg per day) (mgh2)
0 12 0 13 0 0 0.07-1.4 2.6-52
20 11 1 14 0.9 5
Methylphenidute hydrochloride Dose @pm) Feed consumption (54 weeks) Compound consumption (54 weeks) Feed consumption (102 weeks) Compounds consumption (102 weeks) Estimated dose (mg/m*) Estimated human dose (mg/kg per day) (mg/m2)
0 12.7 0 12.2 0 0 0.3-1.0 1 l-37
100 11.8 5 11.1 4 21
500 10.9 23 11.2 20 104
1000 10.8 SO 10.8 44 229
400 12.2 18 11.2 13 68
800 13.1 39 10.4 26 135
1600 11.5 70 11.4 60 312
Codeine Dose @pm) Feed consumption (53 weeks) Compound consumption (53 weeks) Feed consumption (101 weeks) Compound consumption (101 weeks) Estimated dose (mg/m2) Estimated human dose (mg/kg per day) (mg/m*)
0 12.1 0 11.6 0 0 2.9 107
-
loo 9 4 12 5 26
aDose group: C, control; L, low dose; M, mid dose; H, high dose. bFeed consumption, g feed/animal per day. CCompound consumption, mg drug/kg animal body weight per day. dDose in mg/m2 body surface area based on Freireich et al. (1966) where mg/m2 = Km x (dose in mg/kg) where Km is 37 for humans and 5.2 for rats.
J. K. Dunnick et ul. I Cancer Letters 102 (I 996) 77-83
A finding common to each of these chemicals was a decreased incidence of spontaneous neoplasms in rats and mice (Table 1). All chemicals also showed a decreased body weight relative to controls, a response which has previously been associated with decreases in the incidences of some naturally occurring spontaneous tumors (e.g. mammary gland tumors in rats and liver tumors in mice) 193.Feed consumption in these present studies was similar between treated and respective control groups; decreases in body weight were not due to a decreased feed intake (data for female rats shown; Table 3). Analysis of the decrease in mammary gland tumors in female F344 rats (Table 4) shows that the decrease in body weight alone could not explain the decreased incidence observed in the highest codeine group; a similar trend appears to hold true for the other two chemicals. Table 4 Mathematical modeling of mammary gland neoplasms in female F344 rats C’
L
M
H
701 268 50 39
718 236 24 29
-
703 218 6 23
708 265 33 37
709 252 34 34
707 233 16 28
707 219 10 23
670 310 61 51
708 275 46 42
702 270 58 40
699 263 16 37
Amphetamine sulfate
Mean survival (days) b 52 week body weight (g) Observed tumor rate (%) b Predicted tumor rate (%)’ Methylphenidate
81
Table 5 Mathematical modeling of mouse liver tumors Male mouse
Female mouse
Ca
H
C
H
722 42.6 29 41
727 29.4 4 14
704 35.6 10 10
728 26.1 2 7
729 43.6 48 45
720 42.0 68 40
706 40.1 19 22
733 41.6 60 27
707 50.0 59 66
712 44.4 33 47
704 49.9 33 43
709 42.1 17 27
Amphetamine sulfate
Mean survival (days) 52 week body weight(g) Observedtumor rate (%)b Predictedtumor rate (%)’
Methylphenidate hydrochloride
Mean survival (days) 52 week body weight (g) Observed tumor rate (%) Predictedtumor rate (%) Codeine
Mean survival (days) 52 week body weight (g) Observedtumor rate (%) Predictedtumor rate (%)
aDosegroup - seeTable 1; C, control; H, high dose. bExcluding all animals dying prior to 52 weeks. CPredictedtumor rate based on logistic model [S]) relating tumor incidence to age at death and body weight at 52 weeks of age where probability of tumor for an individual animal = l/(1 + eVZ). z = a + b(52 week body weight) + ~(52 week body weight)2 + d(survival in days). individually housed mice: for male mouse liver: o = -3.15; b = 0; c = 0.00152; d = 0; for female mouseliver: a = -7.09; b = 0; c = 0.00115; d = 0.00552. Group housed female mouseliver: a = -7.34; b = 0; c = -0.00078; d = 0.00576.
hydrochloride
Mean survival (days) 52 week body weight (g.) Observed tumor rate (%) Predicted tumor rate (%) Codeine
Mean survival (days) 52 week body weight (g.) Observed tumor rate (%) Predicted tumor rate (%)
aDosegroup: C, control; L, low dose; M, mid dose; H, high dose. bExcluding all animals dying prior to 52 weeks. ‘Predicted tumor rate based on logistic model (6) relating tumor incidence to age at death and body weight at 52 weeks of age where tumor probabilities were calculated for each individual animal, and then averaged to obtain predicted tumor rate: probability of tumor for an individual animal = l/(1 + eeL).z.= a + b(52 week body weight) + ~(52 week body weight)2 + d(survival in days); for female rat mammary gland: a = -8.68; b = 0.0162; c = 0; d = 0.0055.
In contrast, the decrease in liver tumors in amphetamine and codeine mice, appeared to primarily be attributable to body weight changes (Table 5). The magnitude of the observed difference in the incidence of liver tumors between the high dose and control liver tumor rates is very similar to that predicted by the model, although some tumor incidences are slightly over-predicted. This slight over-prediction could just reflect lab-to-lab variability in tumor rates. These data suggest that the apparent chemically related decreases in liver tumor incidence in mice from the amphetamine and codeine studies are primarily a reflection of the reduced body weights in the dosed groups. Seilkop [8] found no strong correlation between body weight and tumor incidence for neoplasms of the adrenal gland and uterus of rats; or harderian
82
J.K. Dunnick et al. / Cancer Letters 102 (1996) 77-83
gland, lung, pituitary gland, and lymphoma of mice. It is uncertain what factors contributed to most of the decreases in tumor incidence at these sites in the amphetamine, methylphenidate, or codeine studies. However, the decrease in the incidence of adrenal pheochromocytomas in the male rat in the codeine study appears to be related to chemical treatment, because this decrease occurred in the two lower dose groups where decreases in body weight did not occur. Amphetamine, methylphenidate, and codeine share some common biological properties. These drugs all affect the neuroendocrine system [l]. This interaction has been characterized for codeine where interaction of codeine with specific opioid receptors on the cell membrane is thought to be the first step by which codeine (and other morphine analogs) affect biological processes [lo]. Membrane receptors for amphetamine and methylphenidate have not been cloned, but studies suggest that these chemicals may compete with morphine for binding to membrane receptors [ 1I]. The pathogenesis of mammary gland neoplasia in the rat is similar to that in humans [12-141, but the mechanism for the decrease in the spontaneous mammary gland neoplasm after treatment with amphetamine, methylphenidste, or codeine is not understood. Studies using narcotic antagonists [ 151 would help to determine if this effect is due to interaction of the chemicals with specific membrane receptors. Other studies have shown that opioids inhibit growth of breast cancer cells [ 16,171 and lung cancer cells in vitro [ 181, suggesting that the biologic properties by which opioids may control growth of mammary cancer cells, as suggested by our studies, may also be applicable to other types of cancer. Mammary tumor growth appears to be hormone dependent in the early stages [19]. Current approaches in the treatment of this cancer use antiestrogens [20,21] which interact with steroid receptors. The receptor-hormone complex is localized predominantly in the nucleus where it can interact with DNA and regulate subsequent cellular processes (e.g. cell proliferation) [ 10,22-241. The primary antiestrogen therapy under investigation is tamoxifen, but efficacy with this drug is not seen in all breast cancer cases, particularly estrogen-receptor negative breast cancer [25]. In addition, use of tamoxifen has been associated with potential side effects [26,27], indicat-
ing the need for new therapeutic strategies for breast cancer. An alternative treatment strategy may be offered by use of drugs that interact with membrane protein receptors such as the opiate ‘G-protein receptor family class’, a class of receptors widely distributed in organ systems throughout the body including brain, breast, and thyroid [28,29]. While there are few epidemiology studies which explore the use of alternative drugs to control cancer, one survey of patients in a medical care program found that long term use of methylphenidate had an association with a reduction in the incidence of cancer [30]. This is consistent with the present data showing that pharmaceuticals that affect the neuroendocrine system are associated with decreases in the incidence of spontaneously occurring mammary gland neoplasms. Acknowledgments We appreciate the review of the manuscript by Dr. Greg Dinse and Dr. Michael Dieter, National Institute of Environmental Health Sciences, Research Triangle Park, NC. References [II Gilman, A.G., Rail, T.W., Nies, AS. and Taylor, P. (1990) Ul
[31
141
VI @I 171
Goodman and Gilman’s The Pharmacological Basis of Therapeutics. National Toxicology Program (1995) NTP Technical Report on the toxicology and carcinogenesis studies of codeine in the F344/N rats and B6C3Fl mice (TR 455). Report, ResearchTriangle Park, NC. National Toxicology Program (1995) NTP Technical on the toxicology and carcinogenesisstudies of codeine in F344/N rats and B6C3Fl mice. TR 455 NIH Publication No. 95 3360. National Toxicology Program (1995) NTP Technical Report on the toxicology and carcinogenesis of methylphenidate hydrochloride. TR 439. NIH Publication No. 95-3355. Dunnick, J.K. and Eustis, S.L. (1991) Decreasesin spontaneous tumors in rats and mice after treatment with amphetamine. Toxicology, 67,325-332. Dunnick, J.K. and Hailey, J.R. (1995) Experimental studies on the long-term effects of methylphenidate hydrochloride. Toxicology, in press. Dunnick, J.K. and Elwell, M.R. (1989) Toxicity studies of amphetaminesulfate, ampicillin trihydrate, codeine, 8-methoxypsoralen, alpha-methyldopa, penicillin VK, and propantheline bromide in rats and mice. Toxicology, 56, 123-136.
J.K. Dunnick et al. I Cancer Letters 102 (1996) 7743
[S] Seilkop, SK. (1995) The effect of body weight on tumor incidence and carcinogenic@’ testing in B6C3Fl mice and F344 rats. Fund. Appl. Toxicol., 24, 241-259. [9] Rao, G.N., Piegorsch, W.W. and Haseman, J.K. (1987) Influence of body weight on the incidence of spontaneous tumors in rats and mice of long-term studies. Am. J. Clin. Nutr., 45, 2.52-260. [lo] Evans, C.J., Keith, Jr., D.E., Morrison, H., Magendzo, K. and Edwards, R.H. (1992) Cloning of a delta opioid by functional expression. Science, 258, 1952-1955. [ll] Jones, D.N.C. and Holtzman, S.G. (1992) Interaction between opioid antagonists and amphetamine: evidence for mediation by central delta opioid receptors. J. Pharmacol. Exp. Ther., 262,638-645. [12] Russo, J. and Russo, I.H. (1987) Biology of disease. Biological and molecular bases of mammary carcinogenesis. Lab. Invest., 62, 244-278. [13] Russo, J., Gusterson, B.A., Rogers, A.E., Russo, I.H., Wellings, S.R. and van Zwieten, J. (1990) Biology of disease. Comparative study of human and rat mammary tumorigenesis. Lab. Invest., 62, 244-278. [14] Dunnick, J.K., Elwell, M.R., Huff, J. and Barrett, J.C. (1995) Chemically induced mammary gland cancer in the National Toxicology Program’s carcinogenesis bioassay. Carcinogenesis, 16, 173-179. [15] Simpkins, J.W., Swager, D. and Millard, W.J. (1991) Evaluation of the sites of opioid influence on anterior pituitary hormone secretion using a quarternary opiate antagonist. Neuroendocrinology, 54, 384-390. [16] Maneckjee, R., Biswas, R. and Vonderhaar, B.K. (1990) Binding of opioids to human MCF-7 breast cancer cells and their effects on growth. Cancer Res., 50.2234-2238. [17] Hatzoglou, A., Ouafik, L., Bakogeorgou, E., Thermos, K. and Castanas, E. (1995) Morphine cross-reactswith somatostatin receptor SSTR2 in the T47D human breast cancer cell line and decreasescell growth. Cancer Res., 55, 56325636. [18] Maneckjee, R. and Minna, J.D. (1990) Opioid and nicotine receptors affect growth regulation of human lung cancer cell lines. Proc. Natl. Acad. Sci. USA, 87, 3294-3298.
83
[19] Nandi, S., Guzman, R.C. and Yang, J. (1995) Hormones and mammary carcinogenesisin mice, rats and humans: a unifying hypothesis. Proc. Natl. Acad. Sci. USA, 92,3650-3657. [20] Pfeffer, U., Fecarotta,E. and Vidali, G. (1995) Coexpression of multiple estrogen receptor variant messenger RNAs in normal and neoplastic breast tissues and MCF-7 cells. Cancer Res., 55,2158-2165. [21] Rutqvist, L.E., Johansson,H., Signomklao, T., Johansson, U., Fomander, T. and Wilking, N. (1995) Adjuvant tamoxifen therapy for early stage breast cancer and second primary malignancies.J. Natl. Cancer Inst., 87,645X51. [22] Carson-Jurica,M., Schrader,W.T. and O’Malley, B. (1990) Steroid receptor family: structure and functions. Endocr. Rev., 11,201-220. [23] Beato, M. (1989) Gene regulation by steroid hormones.Cell, 56,335-344. [24] Green, S., Walter, P., Kumar, B., Krust, A., Bomert, J., Argos, P. and Chambon, P. (1986) Human oestrogen receptor cDNA: sequence,expression and homology to v-erb-A. Nature, 320, 134-139. [25] Ferguson, A.T., Lapidus, R.G., Baylin, S.B. and Davidson, N.E. (1995) Demethylation of the estrogen receptor gene in estrogen receptor-negative breast cancer cells can reactivate estrogen receptor gene expression. Cancer Res., 55, 22792283. [26] Jordan, V.C. (1995) Tamoxifen and tumorigenicity: a predictable concern.J. Natl. Cancer Inst., 87.623-626. [27] Simon, R. (1995) Discovering the truth about tamoxifen: problems of multiplicity in statistical evaluation of biomedical data. J. Natl. Cancer Inst., 87.627-629. [28] Bockaert, J. (1991) G proteins and G-protein-coupled receptors: structure, function, and interactions. Curr. Biol., 1, 3242. [29] Bostwick, D.G., Null, W.E., Holmes, D., Weber, E., Barchas, J.D. and Bensch, K.G. (1987) Expression of opioid peptides in tumors. N. Engl. J. Med., 317, 1439-1443. [30] Selby, J.V., Friedman, G.D. and Fireman, B.H. (1989) Screening prescription drugs for possible carcinogenicity: eleven to fifteen years of follow-up. Cancer Res., 49, 57375747.
ELSEVIER
Cancer Letters 102 (1996) 77-83
Decreased incidence of spontaneous mammary gland neoplasms in female F344 rats treated with amphetamine, methylphenidate, or codeine June K. Dunnick*, Michael R. Elwell, Joseph K. Haseman Nutional Institute
of Environmental Health Sciences, PO. Box 12233, Research Triangle Park, NC 27709, USA
Received 26 December 1995; revision received 19 January 1996;accepted 19 January 1996
Abstract
Three drugs that affect the neuroendocrinesystem(amphetamine,methylphenidate,and codeine) causeddecreasesin body weights and in the incidence of spontaneouslyoccurring mammarygland neoplasmsin the female F344/N rat in 2-year carcinogenicity studies. Using a mathematical model that relates body weight changes to the incidence of mammary gland neoplasms, we find that the decrease in mammary gland tumors seen in female rats cannot be fully explained by body weight
decreasesrelative to control animals. Further, the observeddecreasesin body weight in treated female rats were not a function of differences in feed consumption between treated and control groups. These pharmaceuticalsare thought to affect the biologic system through interaction with membrane receptors. This interaction and/or subsequent cell signaling events may
play a role in the observed decreasein spontaneously occurring mammary gland neoplasmsin the female rat treated with amphetamine,methylphenidate, or codeine. Keywords:
Mammary gland neoplasm;Amphetamine; Methylphenidate; Codeine; F344 rats
1. Introduction Experimental studies in animals are used in conjunction with epidemiology studies to assess the potential long term toxicity, carcinogenicity, and/or beneficial effects of drugs. We have investigated the long-term effects of amphetamine sulfate, methylphenidate hydrochloride, and codeine in the carcinogenesis bioassay because these drugs are important in our armamentarium of therapeutic agents. Am-
* Corresponding author.
Elsevier Science Ireland Ltd. PII: SO304-3835(96)04168-7
phetamine is used in the treatment of narcolepsy, attention deficit hyperactivity disorders, and in weight control; methylphenidate in the treatment of narcolepsy and attention deficit hyperactivity disorders; and codeine in pharmaceuticals used as analgesics, sedatives, hypnotics, antiperistaltics, and antitussive agents [ 11. Amphetamine sulfate, methylphenidate hydrochloride, or codeine administered in the feed to F344 rats and B6C3Fl mice for 2 years did not show sitespecific carcinogenic activity except some carcinogenic activity in the liver of mice receiving methylphenidate [2-61. In our bioassay we also found that these chemicals
78
J.K. Dunnick et al. I Cancer Letters 102 (1996) 77-83
Table 1
J’w
Sex and species
Dose, mgkg per daya (ppm in feed)
Level of carcinogenic activity
Decreasesin spontaneous neoplasms
Amphetamine sulfate
Male rats
0, 0.9, 5
No evidence
Adrenal pheochromocytoma (23/49, 15/44,6/50); pituitary gland adenoma(15/49, 15/48,
0, 1.1.5.2 (0,20, 100)
No evidence
9/49)
0, 3.8,31.6 to, m1w
No evidence
to, 20, 100) Female rats
Male mice
Mammary gland fibroadenoma/ adenoma/ adenocarcinoma(25/50, 12/50, 3/50); pituitary gland adenoma (31/50,24/48, 19/50); uterus, endometrial stromal polyp (10/50,6/50, 3/50)
Female mice
0, 3, 18.5 to, 20, 100)
No evidence
Harderian gland adenoma (4/50,2/50,0/50); lung adenomakarcinoma (S/50, 3/30, 4/50); liver adenoma/ carcinoma (14/50, 12/50, 2/50) Pituitary gland adenoma (12/49,6/49, l/46); Harderian gland adenoma(5/50,2/50, O/47);
Lung adenomakarcinoma (S/50,6/50, l/47); liveradenomakarcinoma (5/50, l/50, l/47) Male rats
Methyl phenidate hydrochloride
Female rats
0.4, 20,42(0, 100, 500,lOOO) 0,4, 22,47 (0, 100,
No evidence No evidence
500,1000)
Adrenal pheochromocytoma (18/49,7/48, 5/49, 10/50) Mammary gland tibroadenoma/ adenoma/ adenocarcinoma(16/49, 17/50. 8/48,5/50)
Malemice
Female mice Male rats
Codeine
Female rats
Male mice
0,5,28,56(0,50, 250,500) 0, 7, 34, 67 (0, 50, 250,500)
Someevidence (liver tumors) Someevidence (liver tumors)
0, 12,27,64 (0,400,800, No evidence 1600) 0, 15, 31,63 (0,400, 800, No evidence 1600)
0.72, 159, 342(0, 750,1500,3000)
No evidence
0,77, 159, 384(0, 750,1500,3000)
None
Adrenal pheochromocytoma (16/49,6/50,6/50,
3/50)
Mammary gland tibroadenoma/adenoma/adenocarcinema (30/.50,23/50, 29/51, 8/51) Liver adenomakarcinoma (29/50, 29/50, 23150, 16/50);
‘w Female mice
None
No evidence
malignant lymphoma (7/50, 1150,l/50,0/50) Liver adenomakarcinoma (16/50, 15/51, 15/51, 8/50)
J.K. Dunnick et al. I Cancer Letters IO2 (1996) 77-83
79
decreased the incidences of some spontaneously occurring tumors. In this paper we analyze the decrease in spontaneous tumors using a mathematical model that relates body weight changes to the incidence of mammary gland neoplasms, and discuss the significance of our findings.
the results of 13-week studies and were chosen to provide a minimally toxic high dose and lower doses within 10x the human dose based on mg drug/m2 body surface area per day [7]. The doses for each of the chemicals and tumor incidences are listed in Table 1.
2. Methods and materials
2.2. Mathematical model for analyzing decreases in tumor incidence
2.1. Curcinogenesis bioassay The relationship between site-specific tumor incidence and body weight in control animals was examined using a mathematical model derived from individual animal data from approximately 100 NTP bioassays in the F344 rat and the B6C3Fl mouse [8]. The derived formula uses a logistic model that provides an excellent prediction of tumor incidence based on an animal’s age at death and body weight at 52 weeks. We have applied this model to the analysis for decreases in mammary gland and liver neoplasms: For mammary tumors: probability of tumor for an
Amphetamine sulfate, methylphenidate hydrochloride, or codeine were studied in separate carcinogenesis bioassays. Drugs were administered in the diet to groups of male and female F344LN rats and B6C3Fl mice for periods of up to 2 years, at which time the animals were killed and a complete gross and histopathologic tissue evaluation performed. Each group contained 50 animals per species/sex with an additional 10 animals per species/sex added for interim evaluation at 15 months [2-4]. The doses selected for the 2-year studies were based on Table 2 Survival and mean body weight at 2 years Drug
Final mean body wt. (g)
% Survival at 2 years L
M
H
C
L
M
H
Amphetamine sulfate 60 Male rats 66 Female rats
62 84
-
66 14
461 349
424** 309**
-
391** 243**
96 70
96 12
-
98 88*
43 42
31*+ 34**
-
31** 27**
66 64 90 70
68 12 88 74
68 78 82 88
391 317 46 44
400 302 44 43
372 276** 40** 41
353** 252** 42* 42
40 76+ 76 72
42 58 90 86
40 64 86 70
411 342 48 50
405 335 51 51
390 315** 44* 49
362** 303** 42** 40**
Ca
Male mice Female mice Methylphenidate Male rats Female rats Male mice Female mice Codeine Male rats Femaie rats Male mice Female mace
hydrochloride 56 62 90 74
58 56 82 72
*P-c 0.05 versus control. **P < 0.01 versus control. aDoses as listed in Table 1; C, control; L, low dose; M, mid dose; H, high dose.
80
J. K. Dunnick et al. I Cancer Letters 102 (1996) 77-83
individual animal = I/( 1 + e-z), z = a + b(52 week body weight) + ~(52 week body weight)2 + d (survival in days). For female rat mammary gland: a = 8.68, b = 0.0162, c = 0, d = 0.0055. For liver tumors: probability of tumor for an individual animal = l/( 1 + e-z), z = a + b(52 week body weight) + ~(52 week body weight)2 + d(surviva1 in days). Individually housed mice: for male mouse liver: a = -13.15, b = 0, c = 0.00152, d = 0; for female mouse liver: a = -7.09, b = 0, c = 0.00115, d = 0.00552. Group housed female mouse liver: a = 7.34; b = 0; c =-0.00078; d = 0.00576. The parameters a, b, c, and d are estimated by lo-
gistic regression techniques for each site-specific tumor, using individual animal data from the NTP historical control database. 3. Results and discussion There was no evidence for treatment-related mortality in rats or mice after 2 years of administration of the drugs (Table 2). None of these drugs were mutagenic in the Salmonella assay, and the only one that exhibited carcinogenic activity was methylphenidate, which caused an increase in the incidence of liver tumors in mice [6].
Table 3 Estimated feed and drug consumption in the female F344 rat Ca
L
M
H
Amphetamine .sulfate Dose @pm) Feed consumption (51 weeks)b Compound consumption (5 1 weeks)c Feed consumption (102 weeks) Compound consumption (102 weeks) Estimated dose (mg/m2)d Estimated human dose (mg/kg per day) (mgh2)
0 12 0 13 0 0 0.07-1.4 2.6-52
20 11 1 14 0.9 5
Methylphenidute hydrochloride Dose @pm) Feed consumption (54 weeks) Compound consumption (54 weeks) Feed consumption (102 weeks) Compounds consumption (102 weeks) Estimated dose (mg/m*) Estimated human dose (mg/kg per day) (mg/m2)
0 12.7 0 12.2 0 0 0.3-1.0 1 l-37
100 11.8 5 11.1 4 21
500 10.9 23 11.2 20 104
1000 10.8 SO 10.8 44 229
400 12.2 18 11.2 13 68
800 13.1 39 10.4 26 135
1600 11.5 70 11.4 60 312
Codeine Dose @pm) Feed consumption (53 weeks) Compound consumption (53 weeks) Feed consumption (101 weeks) Compound consumption (101 weeks) Estimated dose (mg/m2) Estimated human dose (mg/kg per day) (mg/m*)
0 12.1 0 11.6 0 0 2.9 107
-
loo 9 4 12 5 26
aDose group: C, control; L, low dose; M, mid dose; H, high dose. bFeed consumption, g feed/animal per day. CCompound consumption, mg drug/kg animal body weight per day. dDose in mg/m2 body surface area based on Freireich et al. (1966) where mg/m2 = Km x (dose in mg/kg) where Km is 37 for humans and 5.2 for rats.
J. K. Dunnick et ul. I Cancer Letters 102 (I 996) 77-83
A finding common to each of these chemicals was a decreased incidence of spontaneous neoplasms in rats and mice (Table 1). All chemicals also showed a decreased body weight relative to controls, a response which has previously been associated with decreases in the incidences of some naturally occurring spontaneous tumors (e.g. mammary gland tumors in rats and liver tumors in mice) 193.Feed consumption in these present studies was similar between treated and respective control groups; decreases in body weight were not due to a decreased feed intake (data for female rats shown; Table 3). Analysis of the decrease in mammary gland tumors in female F344 rats (Table 4) shows that the decrease in body weight alone could not explain the decreased incidence observed in the highest codeine group; a similar trend appears to hold true for the other two chemicals. Table 4 Mathematical modeling of mammary gland neoplasms in female F344 rats C’
L
M
H
701 268 50 39
718 236 24 29
-
703 218 6 23
708 265 33 37
709 252 34 34
707 233 16 28
707 219 10 23
670 310 61 51
708 275 46 42
702 270 58 40
699 263 16 37
Amphetamine sulfate
Mean survival (days) b 52 week body weight (g) Observed tumor rate (%) b Predicted tumor rate (%)’ Methylphenidate
81
Table 5 Mathematical modeling of mouse liver tumors Male mouse
Female mouse
Ca
H
C
H
722 42.6 29 41
727 29.4 4 14
704 35.6 10 10
728 26.1 2 7
729 43.6 48 45
720 42.0 68 40
706 40.1 19 22
733 41.6 60 27
707 50.0 59 66
712 44.4 33 47
704 49.9 33 43
709 42.1 17 27
Amphetamine sulfate
Mean survival (days) 52 week body weight(g) Observedtumor rate (%)b Predictedtumor rate (%)’
Methylphenidate hydrochloride
Mean survival (days) 52 week body weight (g) Observed tumor rate (%) Predictedtumor rate (%) Codeine
Mean survival (days) 52 week body weight (g) Observedtumor rate (%) Predictedtumor rate (%)
aDosegroup - seeTable 1; C, control; H, high dose. bExcluding all animals dying prior to 52 weeks. CPredictedtumor rate based on logistic model [S]) relating tumor incidence to age at death and body weight at 52 weeks of age where probability of tumor for an individual animal = l/(1 + eVZ). z = a + b(52 week body weight) + ~(52 week body weight)2 + d(survival in days). individually housed mice: for male mouse liver: o = -3.15; b = 0; c = 0.00152; d = 0; for female mouseliver: a = -7.09; b = 0; c = 0.00115; d = 0.00552. Group housed female mouseliver: a = -7.34; b = 0; c = -0.00078; d = 0.00576.
hydrochloride
Mean survival (days) 52 week body weight (g.) Observed tumor rate (%) Predicted tumor rate (%) Codeine
Mean survival (days) 52 week body weight (g.) Observed tumor rate (%) Predicted tumor rate (%)
aDosegroup: C, control; L, low dose; M, mid dose; H, high dose. bExcluding all animals dying prior to 52 weeks. ‘Predicted tumor rate based on logistic model (6) relating tumor incidence to age at death and body weight at 52 weeks of age where tumor probabilities were calculated for each individual animal, and then averaged to obtain predicted tumor rate: probability of tumor for an individual animal = l/(1 + eeL).z.= a + b(52 week body weight) + ~(52 week body weight)2 + d(survival in days); for female rat mammary gland: a = -8.68; b = 0.0162; c = 0; d = 0.0055.
In contrast, the decrease in liver tumors in amphetamine and codeine mice, appeared to primarily be attributable to body weight changes (Table 5). The magnitude of the observed difference in the incidence of liver tumors between the high dose and control liver tumor rates is very similar to that predicted by the model, although some tumor incidences are slightly over-predicted. This slight over-prediction could just reflect lab-to-lab variability in tumor rates. These data suggest that the apparent chemically related decreases in liver tumor incidence in mice from the amphetamine and codeine studies are primarily a reflection of the reduced body weights in the dosed groups. Seilkop [8] found no strong correlation between body weight and tumor incidence for neoplasms of the adrenal gland and uterus of rats; or harderian
82
J.K. Dunnick et al. / Cancer Letters 102 (1996) 77-83
gland, lung, pituitary gland, and lymphoma of mice. It is uncertain what factors contributed to most of the decreases in tumor incidence at these sites in the amphetamine, methylphenidate, or codeine studies. However, the decrease in the incidence of adrenal pheochromocytomas in the male rat in the codeine study appears to be related to chemical treatment, because this decrease occurred in the two lower dose groups where decreases in body weight did not occur. Amphetamine, methylphenidate, and codeine share some common biological properties. These drugs all affect the neuroendocrine system [l]. This interaction has been characterized for codeine where interaction of codeine with specific opioid receptors on the cell membrane is thought to be the first step by which codeine (and other morphine analogs) affect biological processes [lo]. Membrane receptors for amphetamine and methylphenidate have not been cloned, but studies suggest that these chemicals may compete with morphine for binding to membrane receptors [ 1I]. The pathogenesis of mammary gland neoplasia in the rat is similar to that in humans [12-141, but the mechanism for the decrease in the spontaneous mammary gland neoplasm after treatment with amphetamine, methylphenidste, or codeine is not understood. Studies using narcotic antagonists [ 151 would help to determine if this effect is due to interaction of the chemicals with specific membrane receptors. Other studies have shown that opioids inhibit growth of breast cancer cells [ 16,171 and lung cancer cells in vitro [ 181, suggesting that the biologic properties by which opioids may control growth of mammary cancer cells, as suggested by our studies, may also be applicable to other types of cancer. Mammary tumor growth appears to be hormone dependent in the early stages [19]. Current approaches in the treatment of this cancer use antiestrogens [20,21] which interact with steroid receptors. The receptor-hormone complex is localized predominantly in the nucleus where it can interact with DNA and regulate subsequent cellular processes (e.g. cell proliferation) [ 10,22-241. The primary antiestrogen therapy under investigation is tamoxifen, but efficacy with this drug is not seen in all breast cancer cases, particularly estrogen-receptor negative breast cancer [25]. In addition, use of tamoxifen has been associated with potential side effects [26,27], indicat-
ing the need for new therapeutic strategies for breast cancer. An alternative treatment strategy may be offered by use of drugs that interact with membrane protein receptors such as the opiate ‘G-protein receptor family class’, a class of receptors widely distributed in organ systems throughout the body including brain, breast, and thyroid [28,29]. While there are few epidemiology studies which explore the use of alternative drugs to control cancer, one survey of patients in a medical care program found that long term use of methylphenidate had an association with a reduction in the incidence of cancer [30]. This is consistent with the present data showing that pharmaceuticals that affect the neuroendocrine system are associated with decreases in the incidence of spontaneously occurring mammary gland neoplasms. Acknowledgments We appreciate the review of the manuscript by Dr. Greg Dinse and Dr. Michael Dieter, National Institute of Environmental Health Sciences, Research Triangle Park, NC. References [II Gilman, A.G., Rail, T.W., Nies, AS. and Taylor, P. (1990) Ul
[31
141
VI @I 171
Goodman and Gilman’s The Pharmacological Basis of Therapeutics. National Toxicology Program (1995) NTP Technical Report on the toxicology and carcinogenesis studies of codeine in the F344/N rats and B6C3Fl mice (TR 455). Report, ResearchTriangle Park, NC. National Toxicology Program (1995) NTP Technical on the toxicology and carcinogenesisstudies of codeine in F344/N rats and B6C3Fl mice. TR 455 NIH Publication No. 95 3360. National Toxicology Program (1995) NTP Technical Report on the toxicology and carcinogenesis of methylphenidate hydrochloride. TR 439. NIH Publication No. 95-3355. Dunnick, J.K. and Eustis, S.L. (1991) Decreasesin spontaneous tumors in rats and mice after treatment with amphetamine. Toxicology, 67,325-332. Dunnick, J.K. and Hailey, J.R. (1995) Experimental studies on the long-term effects of methylphenidate hydrochloride. Toxicology, in press. Dunnick, J.K. and Elwell, M.R. (1989) Toxicity studies of amphetaminesulfate, ampicillin trihydrate, codeine, 8-methoxypsoralen, alpha-methyldopa, penicillin VK, and propantheline bromide in rats and mice. Toxicology, 56, 123-136.
J.K. Dunnick et al. I Cancer Letters 102 (1996) 7743
[S] Seilkop, SK. (1995) The effect of body weight on tumor incidence and carcinogenic@’ testing in B6C3Fl mice and F344 rats. Fund. Appl. Toxicol., 24, 241-259. [9] Rao, G.N., Piegorsch, W.W. and Haseman, J.K. (1987) Influence of body weight on the incidence of spontaneous tumors in rats and mice of long-term studies. Am. J. Clin. Nutr., 45, 2.52-260. [lo] Evans, C.J., Keith, Jr., D.E., Morrison, H., Magendzo, K. and Edwards, R.H. (1992) Cloning of a delta opioid by functional expression. Science, 258, 1952-1955. [ll] Jones, D.N.C. and Holtzman, S.G. (1992) Interaction between opioid antagonists and amphetamine: evidence for mediation by central delta opioid receptors. J. Pharmacol. Exp. Ther., 262,638-645. [12] Russo, J. and Russo, I.H. (1987) Biology of disease. Biological and molecular bases of mammary carcinogenesis. Lab. Invest., 62, 244-278. [13] Russo, J., Gusterson, B.A., Rogers, A.E., Russo, I.H., Wellings, S.R. and van Zwieten, J. (1990) Biology of disease. Comparative study of human and rat mammary tumorigenesis. Lab. Invest., 62, 244-278. [14] Dunnick, J.K., Elwell, M.R., Huff, J. and Barrett, J.C. (1995) Chemically induced mammary gland cancer in the National Toxicology Program’s carcinogenesis bioassay. Carcinogenesis, 16, 173-179. [15] Simpkins, J.W., Swager, D. and Millard, W.J. (1991) Evaluation of the sites of opioid influence on anterior pituitary hormone secretion using a quarternary opiate antagonist. Neuroendocrinology, 54, 384-390. [16] Maneckjee, R., Biswas, R. and Vonderhaar, B.K. (1990) Binding of opioids to human MCF-7 breast cancer cells and their effects on growth. Cancer Res., 50.2234-2238. [17] Hatzoglou, A., Ouafik, L., Bakogeorgou, E., Thermos, K. and Castanas, E. (1995) Morphine cross-reactswith somatostatin receptor SSTR2 in the T47D human breast cancer cell line and decreasescell growth. Cancer Res., 55, 56325636. [18] Maneckjee, R. and Minna, J.D. (1990) Opioid and nicotine receptors affect growth regulation of human lung cancer cell lines. Proc. Natl. Acad. Sci. USA, 87, 3294-3298.
83
[19] Nandi, S., Guzman, R.C. and Yang, J. (1995) Hormones and mammary carcinogenesisin mice, rats and humans: a unifying hypothesis. Proc. Natl. Acad. Sci. USA, 92,3650-3657. [20] Pfeffer, U., Fecarotta,E. and Vidali, G. (1995) Coexpression of multiple estrogen receptor variant messenger RNAs in normal and neoplastic breast tissues and MCF-7 cells. Cancer Res., 55,2158-2165. [21] Rutqvist, L.E., Johansson,H., Signomklao, T., Johansson, U., Fomander, T. and Wilking, N. (1995) Adjuvant tamoxifen therapy for early stage breast cancer and second primary malignancies.J. Natl. Cancer Inst., 87,645X51. [22] Carson-Jurica,M., Schrader,W.T. and O’Malley, B. (1990) Steroid receptor family: structure and functions. Endocr. Rev., 11,201-220. [23] Beato, M. (1989) Gene regulation by steroid hormones.Cell, 56,335-344. [24] Green, S., Walter, P., Kumar, B., Krust, A., Bomert, J., Argos, P. and Chambon, P. (1986) Human oestrogen receptor cDNA: sequence,expression and homology to v-erb-A. Nature, 320, 134-139. [25] Ferguson, A.T., Lapidus, R.G., Baylin, S.B. and Davidson, N.E. (1995) Demethylation of the estrogen receptor gene in estrogen receptor-negative breast cancer cells can reactivate estrogen receptor gene expression. Cancer Res., 55, 22792283. [26] Jordan, V.C. (1995) Tamoxifen and tumorigenicity: a predictable concern.J. Natl. Cancer Inst., 87.623-626. [27] Simon, R. (1995) Discovering the truth about tamoxifen: problems of multiplicity in statistical evaluation of biomedical data. J. Natl. Cancer Inst., 87.627-629. [28] Bockaert, J. (1991) G proteins and G-protein-coupled receptors: structure, function, and interactions. Curr. Biol., 1, 3242. [29] Bostwick, D.G., Null, W.E., Holmes, D., Weber, E., Barchas, J.D. and Bensch, K.G. (1987) Expression of opioid peptides in tumors. N. Engl. J. Med., 317, 1439-1443. [30] Selby, J.V., Friedman, G.D. and Fireman, B.H. (1989) Screening prescription drugs for possible carcinogenicity: eleven to fifteen years of follow-up. Cancer Res., 49, 57375747.