Review Dietary supplements of dehydroepiandrosterone in relation to breast cancer risk

Transcription

1 European Journal of Clinical Nutrition (1999) 53, 771±775 ß 1999 Stockton Press. All rights reserved 0954±3007/99 $ Review Dietary supplements of dehydroepiandrosterone in relation to breast cancer risk 1 * 1 Oncology Department, St Thomas' Hospital, London, UK Objective: Dietary supplements of the adrernocertical hormone dehydroepiandrosterone (DHEA) are widely taken in the hope of staving off the aging process. Potential dangers have not been fully researched, particularly evidence of a correlation between increased serum concentrations of DHEA and higher breast cancer risk in postmenopausal women. Design: The review examines reports of clinical, epidemiological experimental studies for evidence that DHEA may be a factor in promoting the growth of mammary cancer. Biological mechanisms which might be involved are identi ed. Results: DHEA is reported to inhibit the growth of human mammary cancer cells in vitro and also the growth of chemically-induced mammary cancer in rats. However, growth inhibition occurs only in the presence of high oestrogen concentrations, and growth stimulation occurs in both models in the presence of a low-level oestrogen milieu. Epidemiological studies report a positive correlation between higher serum concentrations of DHEA and increased breast cancer risk in the case of postmenopausal but not premenopausal women. Postulated mechanisms include a direct effect on mammary cells by androgenic metabolites of DHEA or an indirect effect by an increase in bioavailable oestrogen levels. The increased serum concentration of free insulinlike growth factor 1 which follows prolonged DHEA intake may also have a role by stimulating oestrogen receptor activity in breast tissue. Conclusion: Late promotion of breast cancer in postmenopausal women may be stimulated by prolonged intake of DHEA, and the risk may be increased by the endocrine abnormality associated with pre-existing abdominal obesity. Caution is advised in the use of dietary supplements of DHEA particularly by obese postmenopausal women. Descriptors: breast cancer; dehydroepiandrosterone (DHEA); hyperinsulinaemia; IGF1; obesity Introduction Serum concentrations of the adrenocortical hormone dehydroepiandrosterone (DHEA) and its sulphate (DHEAS) increase progressively after the age of 30 y. DHEA tablets are widely on sale as a dietary supplement, and reports of its effect on experimental animals have encouraged its human use in the belief that it may counteract the manifestations of aging. However, its possible dangers and mechanisms of action are uncertain and the endocrinemetabolic response to DHEA administration differs between men and women (Larkin, 1988). Oral doses of 25 ± 100 mg DHEA daily have been given for periods of 3 ± 6 months to healthy age-advanced men and women in small randomised trials (Morales et al, 1994, 1998; Khorram et al, 1997; Casson et al, 1998). Serum concentrations of DHEA rose to levels typical of young adults (Casson et al, 1998; Morales et al, 1998). Those of testosterone, dihydrotestosterone and androstenediol were increased above young adult levels in women, although not in men. Serum concentrations of insulin-like growth factor 1 (IGF1) were elevated and sex hormone-binding globulin levels in the circulation were reduced in both sexes. Immune activity was activated both in men (Khorram et *Correspondence: Dr, Oncology Department, St Thomas' Hospital, London SE1 7EH, UK. Received 12 May 1999; revised 2 July 1999; accepted 15 July 1999 al, 1997) and women (Morales et al, 1998) while total body mass was increased in women but fat mass decreased in men. The increased IGF1 levels were similar for 50 mg DHEA daily for 3 months to those for 100 mg daily for 6 months (Morales et al, 1998) but no change was found in insulin concentrations. In experimental animals, pharmacological dosage of DHEA is reported to diminish obesity, reduce insulin resistance and prevent atherosclerosis (Svec, 1997). Relevant to this review is evidence that the growth of chemically-induced mammary cancer in rats is inhibited by DHEA in the presence of high oestrogen concentrations, but stimulated in a low-level oestrogen milieu (Boccuzzi et al, 1992). Opposite effects depending on oestrogen concentrations have also been shown by DHEA on human mammary cancer cell cultures in the laboratory (Boccuzzi et al, 1994). It is uncertain whether these oestrogen-dependent effects of DHEA on the growth of experimental tumours are relevant to epidemiological ndings on the association between breast cancer risk and serum concentrations of DHEA. In postmenopausal cases, six cohort or case-control studies (Gordon et al, 1990; Dorgan et al, 1996; Berrino et al, 1996; Zeleniuch-Jacquotte et al, 1997; Hankinson et al, 1998; Zumoff et al, 1981) have reported a positive association between increased breast cancer risk and higher serum concentrations of either DHEA or DHEAS but this was not

2 772 con rmed in two other studies (Barrett-Connor et al, 1990; Bernstein et al, 1990). In two studies on premenipausal cases, an increased breast cancer risk was associated with a lower serum concentration of DHEAS (Zumoff et al, 1981; Helzlsouer et al, 1992). Differences in DHEA and DHEAS levels between cases and controls were in most cases relatively small compared to large intersubject variations, whilst the effect of DHEA administration in postmenopausal women is to cause a 2 to 5-fold increase in concentrations (Casson et al, 1998; Morales et al, 1998). DHEA and DHEAS are considered together in this review and are referred to as DHEA(S). The serum concentration of DHEAS is up to 500 times higher than that of DHEA. While DHEAS is bound to albumin, DHEA is active at the tissue level where it is metabolised to steroids such as testosterone and oestrogen. The review examines: (a) Possible mechanisms for a promoting role by DHEA(S) in mammary carcinogenesis; (b) Correlation between DHEA and IGF1 bioactivity; and (c) Effect of obesity and nutritional factors on DHEA(S) concentrations. Possible role for DHEA in mammary carcinogenesis DHEA is a precursor of several steroid metabolites and is regarded as an androgen because it is metabolised to testosterone. Its effect on the rat mammary gland is predominantly androgenic (Sourla et al, 1998). Its intermediate metabolite 5-androstenediol can however compete with oestradiol for binding to the oestrogen receptor (Boccuzzi et al, 1994). It has been postulated that the association between increased breast cancer risk and higher DHEA(S) concentrations in postmenopausal women could result from stimulation of oestrogen receptor protein in mammary epithelium by 5-androstenediol when oestrogen levels are low (Adams, 1992; Ebeling & Koivisto, 1994). An alternative mechanism is suggested (Dorgan et al, 1999) by the nding in some of the DHEA epidemiological studies that a higher testosterone concentration is an independent factor in increasing the risk of postmenopausal breast cancer (Dorgan et al, 1996; Berrino et al, 1996). As noted above, testosterone levels are markedly raised in women taking DHEA as a dietary supplement (Morales et al, 1994, 1998; Khorram et al, 1997; Casson et al, 1998). International studies report that Chinese women show lower mean testosterone concentrations than do Western women and also have a very much lower incidence of breast cancer (Key et al, 1990). Mammary tissue can metabolise androgen to oestrogen which may then synergise with local growth factors in stimulating mammary epithelial growth (Lippman et al, 1998). There is also evidence that androgen competes for oestrogen receptors in human mammary cancer and elicits oestrogenic effects at pharmacological doses (Rochefort & Garcia, 1984). Androgen receptors have been identi ed in a high proportion of mammary cancer specimens and it has been postulated that at physiological levels, androgen might occupy the androgen receptor while at higher levels it might occupy the oestrogen receptor in addition (Zava & McGuire, 1978). It is uncertain whether the excess testosterone concentrations seen in women after prolonged intake of DHEA are high enough for such an effect but the perimenopausal switch in major sex steroid production from the ovary to the adrenal gland may lead to clonal selection in the mammary epithelium. Overgrowth of androgen-sensitive cells in precancerous breast lesions around the time of menopause may render them more sensitive to androgen stimulation. A higher testosterone=oestradiol rate is reported in postmenopausal women with breast cancer compared to controls (Secreto et al, 1983) and it is logical to consider a role for oestrogen=androgen balance rather than oestrogen alone in the promotion of mammary carcinogenesis in postmenopausal women. The following observations support the hypothesis that increased androgen concentrations lead to clonal selection of androgen-sensitive cells in the mammary epithelium at the time of the menopause: (a) The incidence of oestrogen receptor-positive precancerous lesions falls sharply at the onset of the menopause when oestrogen levels are falling (Nielsen et al, 1987); (b) Testosterone at physiological levels has been reported to stimulate the growth of mammary cancer cell lines derived from postmenopausal tumours but not those derived from premenopausal tumours (Simon et al, 1984); (c) Response by advanced breast cancer to high dose androgen therapy varies according to a woman's age in relation to the menopause. In a survey of 521 treated cases, objective response to testosterone propionate was seen in 28% of those more than 5 y postmenopausal, in 17% of those who were postmenopausal 1 ± 5 y and in 8.7% of these within a year after the menopause (Cooperative Breast Cancer Group, 1994). The mechanism of testosterone-induced regression of postmenopausal breast cancer is uncertain but it is unlikely to be mediated by inhibition of gonadotropin. A study of 22 premenopausal cases of advanced breast cancer treated by high-dose androgen found no evidence of objective response (Stoll, 1972). Like high-dose oestrogen therapy, the mechanism of high-dose androgen therapy in causing regression of postmenopausal breast cancer is likely to be mediated by sex steroid receptors in the tumour. Correlation between DHEA and IGF1 bioactivity Concentrations of DHEA and DHEAS usually decline with advancing age and are correlated with falling IGF1 concentrations (Paolisso et al, 1997). A rise in the free IGF1 level following DHEA intake (Morales et al, 1994, 1998; Khorram et al, 1997; Casson et al, 1998) may have a role in promoting mammary carcinogenesis. Considerable laboratory evidence shows that interaction between IGF1 and the oestrogen receptor is involved in regulating proliferative activity, apoptosis and carcinogenesis in mammary epithelium (Surmacz et al, 1998). In postmenopausal women, higher serum concentrations of oestrogen are clearly associated with higher breast cancer risk (Berrino et al, 1996; Toniolo et al, 1995). A synergistic role for IGF1 is suggested by observations that oestrogen-induced proliferation of human breast cancer cells in vitro is increased by overexpression of the IGF1 receptor (Daws et al, 1996) and that IGF1 stimulates oestrogen receptor activity (Lee et al, 1997). Breast cancer specimens show a positive correlation between IGF1 receptor and oestrogen receptor expression (Pekonen et al, 1988; Peyrat et al, 1988; Foekens et al, 1989; Papa et al, 1993), suggesting that their signals may interact in the control of cancer cell activity. The bioavailability of IGFs is modulated by binding proteins and of these, the IGFBP3 level is inversely correlated with oestrogen receptor expression in breast cancer specimens (McGuire et al, 1994; Shao et al, 1992; Pekonen et al, 1992). However, IGFBP3 can either inhibit the mitogenic

3 effect of IGFs by blocking their binding to the IGF1 receptor or else enhance their action by increasing the bioavailability of IGFs in mammary tissue. (Rocha et al, 1997). Epidemiological studies on the association between IGF1 activity and breast cancer risk have also been confounded by incomplete knowledge on in vivo proteolysis of IGFBP3 in relation to IGF1 bioactivity (Rocha et al, 1996). A positive association between free IGF1 concentration and breast cancer risk is shown in 6 out of 9 reported studies (Peyrat et al, 1993, 1998; Barni et al, 1994; Bruning et al, 1995; Favoni et al, 1995; Del Giudice et al, 1998; Hankinson et al, 1998; Ng et al, 1998; Vadgama et al, 1999) including a prospective study in 397 cases, and of the three studies failing to show a positive association, two showed instead a positive association with IGFBP3 concentrations (Del Guidice et al, 1998; Ng et al, 1998). Interpretation of these ndings awaits further research. The age distribution of the association with IGF1 level is not consistent in the studies. Of the six studies showing a positive association, four found it to apply both to pre- and postmenopausal cases (Peyrat et al, 1993, 1998; Barni et al, 1994; Vadgama et al, 1999) while two found it to apply only to premenopausal cases (Bruning et al, 1995; Hankinson et al, 1998). One of the latter was a prospective study where the menopausal status was de ned at the time of blood collection (Hankinson et al, 1998). The researchers point out that although risk was not increased in association with higher postmenopausal IGF1 concentrations, a raised IGF1 level before the menopause could well affect the risk of breast cancer after the menopause. Most of the experimental evidence that the oestrogen receptor interacts with the IFG1 receptor in regulating proliferative activity in breast tissue is based on observations on breast cancer cells (Daws et al, 1996; Lee et al, 1997). The interaction may however apply also to normal mammary epithelial cells (Pekonen et al, 1988). A study of normal human breast tissue implanted into the nude mouse has shown upregulation of IGF1 receptor activity following oestradiol treatment (Clarke et al, 1997). In the presence of oestradiol therefore, overexpression of IGF1 receptor protein might increase oestrogen receptor activity and favour promotion of carcinogenesis which has already been initiated. Effects of obesity and nutrition In postmenopausal women, obesity is predominantly abdominal in distribution and is reported to be positively associated with higher serum concentrations of DHEA (Barrett-Connor & Ferrara, 1996; Haffner et al, 1995). This contrasts with the nding in premenopausal women where abdominal obesity is associated with lower DHEA(S) concentrations (Paolisso et al, 1997; De Pergola et al, 1996; Mantzoros et al, 1996). The reason for the difference is unclear but multiple studies have shown abdominal obesity to be a marker of increased breast cancer risk in postmenopausal women (Stoll, 1996). Postmenopausal abdominal obesity is usually associated with hyperinsulinaemia, and various endocrine-metabolic changes may have a role in promoting mammary carcinogenesis. First, oestrogen levels may be increased by higher oestrogen production in the excess fat deposits, and oestrogen bioactivity increased by lower levels of sex hormonebinding globulin (Ballard-Barbash, 1996; Portischman et al, 1996). Second, free testosterone levels are increased as is also the production of testosterone and dihydrotestosterone (Kirschner et al, 1990). Third, serum concentrations of free IGF1 are increased and probably mediated by hyperinsulinaemia (Conover et al, 1992). Dietary supplements of DHEA are followed by marked increase in serum concentrations of both androgen and IGF1, and may further increase breast cancer risk in women showing evidence of abdominal obesity after the menopause. Evidence to date on the effect of nutrition and diet on DHEA(S) concentrations is limited and inconclusive. Asian women have signi cantly lower DHEA(S) levels than do Caucasian women, and also a relatively lower incidence of breast cancer especially after the menopause (Adams, 1992; Khaw, 1996). When they migrate to the USA or Hawaii, both their DHEA(S) levels and also their breast cancer risk are reported to rise. The high soya content of Asian diets is purported to protect against breast cancer and it may be relevant that supplementary soy feeding to premenopausal Western women is reported to reduce DHEA(S) and also oestradiol concentrations (Lu et al, 1996). A study on teenage vegetarians in the USA showed higher DHEA levels than in omnivores (Persky et al, 1995) although a high bre diet has been reported to reduce breast cancer risk in case-control studies from many countries. Again, a European study showed that a lactovegetarian diet increased DHEA concentrations in healthy young men and women (Remer et al, 1998). On the other hand, postmenopausal women show no difference in DHEAS concentrations between vegetarian and non-vegetarian women (Barbosa et al, 1990). Conclusion Administration of DHEA has been shown to inhibit the growth of human breast cancer cell xenografts in the nude mouse and on the basis of experimental studies, the National Institutes of Health in the USA are considering a trial of DHEA as a dietary supplement among its phase two chemopreventive trials in breast cancer (Couillard et al, 1998; Kelloff et al, 1997). The metabolic-endocrine changes observed in recent studies of prolonged DHEA administration to age-advanced individuals would suggest caution in such trials. DHEA tablets are classed as a dietary supplement and are widely on sale in many western countries without clear evidence of their safety. They are purported to have immunomodulating and antidepressant effects but considerably more clinical evidence is required. The review examines biological mechanisms by which the metabolic changes reported in elderly women might have a role in the late promotion of mammary carcinogenesis. The evidence would suggest caution both in the use of DHEA as an antiaging regimen and also in its proposed trial for breast cancer prevention. References Adams JB (1992): Human breast cancer; concerted role of diet, prolactin and adrenal C19 steroids in tumorigenesis. Int. J. Cancer 50, 854 ± 858. Ballard-Barbash R (1994): Anthropometry and breast cancer. Cancer 74, 1090 ± Barbosa JC, Shultz TD, Filley SJ & Nieman DC (1990): The relationship among adiposity, diet and hormone concentrations in vegetarian and non-vegetarian postmenopausal women. Am. J. Clin. Nutr. 51, 798 ±

4 774 Barni S, Lissoni P & Brivio F (1994): Serum levels of IGF1 in operable breast cancer in relation to their main prognostic variables. Tumor 80, 212 ± 215; (English Abst). Barrett-Connor E, Friedlander NJ & Khaw KT (1990): DHEAS and breast cancer risk. Cancer Res. 50, 6571 ± Barrett-Connor E & Ferrara A (1996): DHEA, DHEAS, obesity, waist-hip ratio and NIDDM in postmenopausal women. J. Clin. Endocrinol. Metab. 81, 59 ± 64. Bernstein L, Ross RK & Pike MC (1990): Hormone levels in older women; a study of postmenopausal breast cancer patients and healthy population controls. Br. J. Cancer 61, 298 ± 302. Berrino F, Muti P & Micheli A (1996): Serum sex hormone levels after menopause and subsequent breast cancer. J. Natl. Cancer Inst. 88, 291 ± 296. Boccuzzi G, Aragno M & Brignardello E (1992): Opposite effects of DHEA on the growth of DMBA-induced rat mammary carinomas. Anticancer Res. 12, 1479 ± Boccuzzi G, Brignardello & Di Monaco M. (1994): 5-androstenediol inhibits the growth of MCF7 breast cancer cells when oestrogen receptors are blocked by oestradiol. Br. J. Cancer. 70, 1035 ± Bruning PF, van Doorn J & Bonfrer JMG (1995): IGFBP3 is reduced in early stage operable premenopausal breast cancer. Int. J. Cancer 62, 266 ± 270. Casson PR, Santoro N & Elkind-Hirsch K (1998): Postmenopausal DHEA administration increases free IGF1 and decreases HDL; a six month trial. Fertil. Steril. 70, 107 ± 110. Clarke RB, Howell A & Anderson E (1997): Type 1 IGF receptor gene expression in normal human breast tissue treated with oestrogen and progesterone. Br. J. Cancer 75, 251 ± 257. Conover CA, Lee PDK & Kanaley JA (1992): Insulin regulation of IGFBP1 in obese and non obese humans. J. Clin. Endocrinol. Metab. 74, 1355 ± Cooperative Breast Cancer Group (1964): Testosterone propionate therapy in breast cancer. J. Am. Med. Assoc. 188, 106 ± 119. Couillard S, Labrie C & Belanger A (1998): Effect of DHEA and the antiestrogen EM-800 on growth of human ZR-75-1 breast cancer xenografts. J. Natl. Cancer Inst. 90, 772 ± 778. Daws MR, Westley BR & May FE (1996): Paradoxical effects of overexpression of type 1 IGF receptor on the responsiveness of human breast cancer cells to IGFs and estradiol. Endocrinology 137, 1177 ± De Pergola G, Zamboni M & Sciaraf a M (1996): Body fat accumulation is possibly responsible for lower DHEA circulating levels in premenopausal obese women. Int. J. Obesity 20, 1105 ± Del Giudice ME, Fantus IG & Ezzat S (1998): Insulin and related factors in premenopausal breast cancer risk. Breast Cancer Res. Treat. 47, 111 ± 120. Dorgan F, Longcope C & Stephenson HC Jr (1996): Relation of prediagnostic serum estrogen and androgen levels to breast cancer risk. Cancer Epidemiol. Biomarkers Prev. 5, 533 ± 539. Dorgan JF, Longcope C, Stanczyk FZ, Stephenson HE Jr & Hoover RN (1999): Plasma sex steriod hormone levels and risk of breast cancer in postmenopausal women. J. Natl. Cancer Inst. 91, 380 ± 381. Ebeling P & Koivisto VA (1994): Physiological importance of dehydroepiandrosterone. Lancet 343, 1479 ± Favoni RE, de Cupis A & Perrotta A (1995): IGF1 and IGFBP blood serum levels in women with early and late stage breast cancer; mutual relationship and possible correlations with patients' hormonal status. J. Cancer Res. Clin. Oncol. 121, 674 ± 682. Foekens JA, Portengen H, Jenssen M & Klijn JG (1989): IGF1 receptors and IGF1-like activity in human primary breast cancer. Cancer 63, 2139 ± Gordon GB, Bush TL & Helzlsouer KJ (1990): Relationship of serum levels of DHEA and DHEAS to the risk of developing postmenopausal breast cancer. Cancer 50, 3859 ± Haffner SM, Newcomb PA & Marcus PM (1995): Relation of sex hormones and DHEAS to cardiovascular risk factors in postmenopausal women. Am. J. Epidemiol. 142, 925 ± 934. Hankinson SE, Willett WC & Colditz GA (1998): Circulating concentrations of IGF1 and risk of breast cancer. Lancet 351, 1393 ± Hankinson SE, Willett WC & Manson JE (1998): Plasma sex steroid hormone levels and risk of breast cancer in postmenopausal women. J. Natl. Cancer Inst. 90, 1292 ± Helzlsouer KJ, Gordon GB & Alberg AJ (1992): Relationship of prediagnostic serum levels of DHEA and DHEAS to the risk of developing premenopausal breast cancer. Cancer Res. 52, 1± 4. Kelloff CJ, Hawk ET & Karp JE (1997): Progress in clinical chemoprevention. Semin. Oncol. 24, 241 ± 252, Key TJA, Chen J, Wang DY, Pike MC & Boreham J (1990): Sex hormones in women in rural China and in Britain. Br. J. Cancer 62, 631 ± 636. Khaw KT (1996): DHEA, DHEAS and cardiovascular disease. J. Endocrinol. 150 (Suppl), S149 ± S153. Khorram O, Vu L, Yen SS (1997): Activation of immune function by DHEA in age-advanced men. J. Geront: Series A: Biol. Sci. Med. Sci. 52, M1 ± M7. Kirschner MA, Samolijk E & Drejka M (1990): Androgen-estrogen metabolism in women with upper body versus lower body obesity. J. Clin. Endocrinol. Metab. 70, 473 ± 479. Larkin M (1998): Will science con rm the headlines? Lancet 352, 208. Lee AV, Weng CN, Jackson JG & Yee D (1997): Activation of estrogenreceptor mediated gene transcription by IGF1 in human breast cancer cells. J. Endocrinol. 152, 39 ± 47. Lipman, ME, Dickson RB, Gelmann EB, Rosen N, Knabbe C, Bates S, Bronzert D, Huff K & Kasid A (1988): Growth regulatory peptide production by human breast carcinoma cells. J. Steroid Biochem. 30, 53 ± 61. Lu LJ, Anderson KE, Grady JJ & Nagamani M (1996): Effects of soya consumption for one month on steroid hormones in premenopausal women; implications for breast cancer risk reduction. Cancer Epidemiol. Biomarkers Prev. 5, 63 ± 70. Mantzoros CS, Georgiadis EI, Evangelopoulo K & Katsilambros N (1996): DHEAS and testosterone are independently associated with body fat distribution in premenopausal women. Epidemiology 7, 513 ± 516. McGuire SE, Hilsenbeck SG & Figueroa JA (1994): Detection of IGF binding proteins by ligand blotting in breast cancer tissues. Cancer Lett. 77, 25 ± 32. Morales AJ, Nolan JJ, Nelson JC & Yen SS (1994): Effects of replacement dose of DHEA in men and women of advancing age. J. Clin. Endocrinol. Metab. 78, 1360 ± Morales HJ, Haubrich RE & Hwang JY (1998): The effect of six months treatment with DHEA on circulating sex steroids, body composition and muscle strength in age-advanced men and women. Clin. Endocrinol. 49, 421 ± 432. Ng EH, Ji CY & Tan PH (1998): Altered serum levels of IGF binding proteins in breast cancer patients. Ann. Surg. Oncol. 5, 194 ± 201. Nielsen M, Thomson JL & Primdahl S (1987): Breast cancer and atypia among young and middle aged women; a study of 83 women. Br. J. Cancer 56, 814 ± 819. Paolisso G, Ammendola S & Rotondi M (1997): Insulin resistance and advancing age; what role for DHEA? Metabolism 46, 1281 ± Papa V, Gliozzo B & Clark GM (1993): IGF1 receptors are overexpressed and predict a low risk in human breast cancer. Cancer Res. 53, 3736 ± Pekonen F, Partanen S, Makinen T & Rutanen EM (1988): Receptors for EGF and IGFI and their relation to steroid receptors in human breast cancer. Cancer Res. 48, 1343 ± Pekonen F, Nyman T, Ilvesmaki V & Partanen S (1992): IGF binding proteins in human breast cancer tissue. Cancer Res. 52, 5204 ± Persky V & van Horn L (1995): Epidemiology of soy and cancer; perspectives and directions. J. Nutr. 125 (Suppl 3), S709 ± S712. Peyrat JP, Bonneterre J & Beuscart R (1988): IGF1 receptors in human breast cancer and their relation to estrogen and progesterone receptors. Cancer Res. 48, 6429 ± Peyrat JP, Bonneterre J & Hecquet B (1993): Plasma IGF1 concentration in human breast cancer. Eur. J. Cancer 29A, 492 ± 497. Peyrat J-Ph, Revillion F & Bonneterre J (1998): Plasma IGF in primary breast cancer patients treated with adjuvant chemotherapy. Br. J. Cancer 77, 1669 ± Potischman N, Swanson CA, Siiteri P & Hoover RN (1996): Reversal of relation between body mass and endogeneous estrogen concentrations with menopausal status. J. Natl. Cancer Inst. 88, 756 ± 758. Remer T, Pietrzik E & Manz F (1998): Short-term impact of a lactovegetarian diet on adrenocortical activity and adrenocortical androgens. J. Clin. Endocrinol. Metab. 83, 2132 ± Rocha RL, Hilsenbeck SG & Jackson JG (1996): Correlation of IGFBP 3 messenger RNA with protein expression in primary breast cancer tissue. J. Natl. Cancer Inst. 88, 601 ± 606. Rocha RL, Hilsenbeck SG & Jackson JG (1997): IGF binding protein-3 and insulin receptor substrate-1 in breast cancer; correlation with clinical parameters and disease-free interval. Clin. Cancer Res. 3, 103 ± 109. Rochefort H & Garcia M (1984): Estrogenic and anti estrogenic activities of androgens in female target tissues. Pharmacol. Ther. 23, 193 ± 216. Secreto G, Recchione C, Cavalleri A, Miraglia & Dati V (1983): Circulating levels of testosterone, 17b oestradiol, luteinising hormone and prolactin in postmenopausal breast cancer. Br. J. Cancer 47, 269 ± 275.

5 Shao ZM, Sheikh MS & Ordonez JV (1992): IGFBP-3 expression and estrogen receptor status in human breast carcinoma. Cancer Res. 52, 5100 ± Simon WE, Albrecht M & Trams G (1984): In vitro growth promotion of human mammary cells by steroid hormones, tramoxifen and prolactin. J. Natl. Cancer Inst. 73, 313 ± 320. Sourla A, Martel C, Labrie C & Labrie F (1998): Almost exclusive androgenic effect of dehydroepiandrosterone in the rat mammary gland. Endocrinology 139, 753 ± 764. Stoll BA (1972): Androgen, corticosteroid and progestin therapy. In: Stoll BA (ed) Endocrine Therapy in Malignant Disease. London: WB Saunders; pp 165 ± 191. Stoll BA (1996): Nutrition and breast cancer risk; can an effect via insulin resistance be demonstrated?. Breast Cancer Res. Treat. 38, 239 ± 246. Surmacz E, Guvakova MA, Nolan MK (1998): Type 1 IGF receptor function in breast cancer. Breast Cancer Res. Treat. 47, 255 ± 267. Svec F (1997): Ageing and adrenal cortical function. Baillieres. Clin. Endocrinol. Metab. 11, 271 ± 287. Toniolo PG, Levitz M & Zeleniuch-Jacquotte A (1995): A prospective study of endogenous estrogens and breast cancer in postmenopausal women. J. Natl. Cancer Inst. 87, 190 ± 197. Vadgama JV, Wu Y, Datta G, Khan H & Chillar R (1999): Plasma IGF1 and IGFBP3 can be associated with progression of breast cancer and predict risk of recurrence and survival probability in African-American and Hispanic women. Oncology (in press). Zava DT & McGuire WL (1978): Androgen action through estrogen receptor in a human breast cancer cell line. Endocrinology 103, 624 ± 630. Zeleniuch-Jacquotte A, Bruning PF & Bonfrer JMG (1997): Relation of serum levels of testosterone and DHEAS to risk of breast cancer in postmenopausal women. Am. J. Epidemiol. 145, 1030 ± Zumoff B, Levin J & Rosenfeld RS (1981): Abnormal 24 hr mean plasma concentrations of DHEA and DHEAS in women with primary operable breast cancer. Cancer Res 41, 3360 ±