Annals of Oncology 3:435-438,1992. O 1992 Kluwer Academic Publishers. Printed in the Netherlands.
Arena New hormone-related markers of high risk to breast cancer B. A. Stoll1 & G. Secreto2 1
Oncology Department, St. Thomas' Hospital, London, UK; 2 Laboratory of Hormonal Research, Istituto Nazionale Tumori, Milan, Italy
Summary. New markers of increased risk to breast cancer are examined and related to established risk markers. The following new evidence is highlighted: (1) Increased testosterone secretion by the ovaries is currently the only major steroid abnormality shown to be associated with increased risk of both premenopausal and postmenopausal breast cancer. (2) Upper body-type obesity is a marker for both hyperandrogenaemia and hyperinsulinaemia and is associated with an increased risk of breast cancer. Upper body type obesity may already be recognised in early puberty in Caucasian girls and is associated with a characteristic androgen/oestrogen profile. (3) Relative tallness in women is associated with an increased risk of breast cancer. A hypothesis is offered on the
significance of these markers in the aetiology of mammary cancer in women, and also a means of testing the hypothesis. The hormonal promotion of mammary carcinogenesis is likely to be greatest between puberty and the first full term pregnancy. The presence of hyperinsulinaemia can increase the ovarian production of androgen, and the abnormal hormonal profile may stimulate proliferative activity in mammary epithelium. This may increase the risk of epithelial atypia and carcinogenesis.
We need markers of high risk to breast cancer if we are to target screening and protective measures to selected subgroups of women, and also provide individuals seeking advice with an estimate of their personal risk. Recent research suggests that the development of human mammary cancer involves oncogene activation leading to upregulation of stimulating growth factors or their receptors, in addition to a promoting action by steroid hormones on previously initiated cells [1]. Currently, the best documented markers of increased risk are a family history of the disease and a group of hormone-related factors. The underlying hormonal interactions are uncertain but both laboratory [2] and epidemiological [3| evidence suggest that promotion of carcinogenesis is greatest during adolescence. A first pregnancy within a few years of the menarche reduces the risk of subsequent breast cancer by over 50%. The temporary reduction of incidence observed after ovarian ablation in adult life is generally assumed to result from growth dormancy in established cancer, rather than from decreased tumour promotion.
6], upper body-type obesity [7-10] and relative tallness [11-16]. They apply both to pre- and postmenopausal breast cancer cases but are more marked in the former category. These new markers may combine with new knowledge of peptide growth factor mechanisms to provide an insight into the causal mechanisms involved. Numerous studies have attempted to identify a hormonal profile associated with an increased risk of breast cancer. Oestrogens are widely believed to act as promoters in previously initiated mammary cells, but the 'oestrogen excess' hypothesis which held the stage for many years (recently with stress on the role of free oestradiol) is gradually being abandoned [17, 18]. An abnormally high serum testosterone level is the only disturbance of androgen secretion so far shown by case-control studies to be associated with both pre- and postmenopausal breast cancer [5, 6]. The abnormality had earlier been observed by several European laboratories [19-21]. That such hyperandrogenaemia is likely to be a risk marker rather than a result of the disease is suggested by the finding of a similar abnormality in women with mammary epithelial hyperplasia [22]. Abnormally high serum testosterone levels have also been observed in Japanese women with breast cancer [23]. Japanese women have in general, a much lower breast cancer incidence than do Caucasian women and also lower serum testosterone levels.
The classical, documented markers of increased risk differ between pre- and postmenopausal women. For premenopausal Caucasian women, they are a family history of the disease, the onset of menarche before the age of 13 and a delayed first full-term pregnancy, while for postmenopausal women, the major risk markers are onset of menopause delayed beyond the age of 49 and the presence of abnormal obesity [4]. In the last year or two, additional risk markers which have been confirmed are increased ovarian testosterone secretion |5,
Key words: breast cancer, risk markers, hyperandrogenaemia, hyperinsulinaemia
The abnormally increased testosterone secretion found in breast cancer patients is abolished by oophorectomy [19, 24] and is assumed to originate in the ovary. High levels of ovarian testosterone may be as-
436
sociated with decreased progesterone secretion and prolonged menstrual cycles in these patients [5], and a subclinical ovarian abnormality is suspected. This is likely to be stromal luteinisation, a condition allied to the polycystic ovary syndrome. Other observers have noted low levels of adrenal androgen in some premenopausal breast cancer patients (reviewed in [18]). In the Channel Islands study where younger (but not older) premenopausal breast cancer patients showed androgen levels at the lower end of the normal range, the observers regard this as a reflection of rapidly growing tumour but not as an indicator of increased predisposition [25]. A clearly-related new line of evidence is the observation by several groups that upper body-type obesity is a marker of increased risk of breast cancer in women [7-10]. For many years it was assumed that obese women are hyperoestrogenic because adipose tissue is a major extragonadal source of oestrogen [26]. However, it is confirmed that women with upper body (male type) fat distribution show high free testosterone levels, sometimes associated with prolonged menstrual cycles [27-29]. They also show metabolic abnormalities such as hyperinsulinaemia and subclinical diabetes [30]. It is believed that hyperinsulinaemia synergises with luteinising hormone in the development of hyperandrogenaemia and that the upper body-type obesity is a marker [27]. The condition is probably geneticallylinked and associated with an abnormal androgen/ oestrogen secretion ratio in early life [30]. It is probably an endocrinopathy allied to the polycystic ovary syndrome. Although high relative body weight is a classical marker of increased breast cancer risk, this does not apply to premenopausal women. While heavier postmenopausal women show between 30% and 60% increase in risk, it has recently been confirmed that heavier premenopausal women have a decreasedriskof breast cancer [31]. Another anthropometric marker of breast cancer risk is tallness, as first pointed out by de Waard [11] and recently confirmed by five other prospective studies [12—16]. Three of the studies involve more than 500,000 women. The Swedish study [13] showed the risk to be increased by 10% for every 5 cm of additional height, the effect being greater in premenopausal women. The Norwegian study [14] showed a 40% increased risk for 15 cm additional height and this applied both to pre- and postmenopausal women. The USA study [15] showed that an 8 cm increase in height increased risk in premenopausal women by 10% and in postmenopausal women by 30%. Evidence of these new risk markers combine with classical risk markers to identify a hormonal profile which may increase proliferative activity in pubertaJ mammary epithelium. Factors which increase proliferation also increase promotion of malignant change and this is most likely to occur at the time of maximum proliferative activity - puberty and adolescence [2]. While epithelial proliferation in the breast can be stimulated
by ovarian steroids, there is increasing evidence that polypeptide growth factors can mediate and modulate the steroid action. Transforming growth factors (TGFa, TGFb), epidermal growth factor (EGF), fibroblast growth factor (FGF) and insulin-like growth factors (IGF1, IGF11) may all play a part. Of these, IGF1 and IGF 11 have the most clearly defined mitogenic effect on mammary cancer cells and it is possible that they promote carcinogenesis in human mammary tissue [32]. The ductal epithelium of normal breast tissue contains abundant IGF1 receptors [33]. The bioactivity of IGF1 in the cell is regulated by a group of specific binding proteins (IGF BP) which in turn are related to circulating levels of growth hormone and insulin. The following observations suggest a role for IGF1 in mammary development at the time of puberty; the administration of recombinant growth hormone has been shown to accelerate breast development in prepubertal girls with isolated growth hormone deficiency [34]; the first onset of breast development in girls coincides with the onset of the adolescent growth spurt and this occurs 2-3 years before the major secretion of ovarian steroids at the menarche [35]; circulating IGF levels peak at puberty and correlate better with breast size than with chronological age [36]. The presence of excess androgen levels at puberty could also stimulate epithelial proliferation in the breast. While the hyperandrogenaemia found in a subgroup of breast cancer patients may merely be a marker of abnormal steroid metabolism in the ovary, it could also stimulate proliferative activity either directly or after aromatisation to oestrogen [37]. As noted above, the mean age at onset of breast development in Western girls in 2-3 years before the menarche [35] and probably involves androgen derived from the adrenal cortex. The concomitant rise in IGF1 level at the time of the growth spurt may stimulate the aromatase in mammary tissue, leading to local oestrogen synthesis [37]. In normal girls, puberty is associated with elevated circulating levels of growth hormone, insulin, IGF1 and testosterone [38] and with so many changes occurring at the same time, causal relationships are uncertain. Although the traditional assumption is that the rise in sex steroid levels at puberty increases the IGF levels [36], hyperinsulinaemia can stimulate both IGF1 and sex steroid activity [38]. Normal ovarian tissue contains abundant IGF1 receptors and IGF1 is said to synergise with gonadotropin in stimulating steroidogenesis in the granulosa cells [39]. Certainly in the case of women with ovarian hyperandrogenism, both IGF1 and insulin have been shown to stimulate testosterone release from samples of ovarian stroma, and the presence of hyperinsulinaemia is therefore thought to contribute to the development of hyperandrogenaemia in this condition [40]. In the case of polycystic ovary syndrome (similarly characterised by hyperandrogenaemia and hyperinsulinaemia) increased activity of the growth hormone/IGF 1 axis is
437
thought to play an important role |41—43]. Frank polycystic disease of the ovary is uncommon but it is noteworthy that an increased breast cancer risk is reported [44, 45]. It has been suggested that in this condition, the hyperinsulinaemia synergises with the increased levels of luteinising hormone to increase ovarian androgen secretion, mainly by upregulating the IGF1 receptor [46, 47]. It may be possible to recognise girls at high risk to breast cancer at the time of puberty. The presence of upper body-type obesity in adult women is associated with increased breast cancer risk [7-10] and upper body-type obesity with a characteristic androgen/ oestrogen ratio in the blood can already be recognised in girls between 10 and 11 years [48]. In a case-control study [49], a large body-mass index at the age of 12 was shown to be an independent risk marker for breast cancer but no details are given as to body fat distribution. While body mass is strongly influenced by nutrition, genetically-linked body fat distribution is likely to be a marker of the endogenous hormonal pattern which influences breast cancer risk. What is the relationship of childhood nutrition to obesity and the risk of developing breast cancer? In women with genetically-linked ovarian hyperandrogenism (allied to polycystic ovary syndrome) obesity may trigger severe hyperinsulinaemia leading to hyperandrogenaemia [47]. Similarly, in women with upper body-type obesity (also allied to polycystic ovary syndrome) the hyperinsulinaemia and hyperandrogenaemia may result from genetic abnormality [43]. However, recent reviews suggest that any cause of severe hyperinsulinaemia may play a key role in the development of ovarian hyperandrogenaemia [43,47, 50]. With regard to the association of tallness with increased breast cancer risk, it is widely accepted that better nutrition accelerates growth hormone release and that a critical body mas triggers the hypothalamic mechanism responsible for the onset of sexual maturity [51]. This may lead to early elevation of growth hormone, insulin and IGF1 levels in addition to sex steroid levels. An earlier adolescent growth spurt favours tallness in adult women [52). The association of early menarche with increased breast cancer risk cannot necessarily be assumed to involve hyperinsulinaemia and hyperandrogenaemia. Menarche before the age of 13 increases breast cancer risk with a highly significant trend to decreasing risk with each succeeding year after that age |53|, but there is equivocal evidence for a distinctive steroid profile in later years in girls manifesting an early menarche [54—56). An earlier menarche may permit a longer period for hormonal stimulation of proliferative activity in the mammary epithelium before it is counteracted by the first full-term pregnancy [57). A hypothesis is offered on the role of these new hormonal markers in the aetiology of mammary cancer in women. The presence of hyperinsulinaemia may increase ovarian production of androgen, or else, in-
creased levels of insulin-like growth factor may synergise with sex steroids in a mitogenic effect on developing mammary tissue in adolescence. The risk of epithelial atypia and carcinogenesis may be increased as a result. If major promotion of mammary carcinogenesis occurs in adolescence, is is possible to test the hypothesis that the association of hyperinsulinaemia and an abnormal ovarian steroid profile may stimulate mitogenesis in adolescent breast tissue. Studies are already under way to monitor circulating levels of IGF1, oestrogen and androgen in girls approaching puberty, and to relate them to observed changes in breast development and statural growth [58]. It would be of considerable interest to relate them also to fat distribution patterns. References 1. Barrett-Lee PJ. Growth factor expression in breast tissue. In Stoll BA (ed): Approaches to Breast Cancer Prevention. Dordrecht: Kluwer Academic Publishers 1991; 53-60. 2. Russo JH, Calaf G, Russo J. Hormones and proliferative activity in breast tissue. In Stoll BA (ed): Approaches to Breast Cancer Prevention. Dordrecht: Kluwer Academic Publishers 1991;35-51. 3. Cole P, MacMahon B. Oestrogen fractions during early reproductive life in aetiology of breast cancer. Lancet 1969; 1: 604-6. 4. Stoll BA. Prospects for breast cancer prevention. In Stoll BA (ed): Women at High Risk to Breast Cancer. Dordrecht: Kluwer Academic Publishers 1989; 107-19. 5. Secreto G, Toniolo P, Pisani P et al. Androgens and breast cancer in premenopausal women. Cancer Res 1989; 49:471-6. 6. Secreto G, Toniolo P, Berrino F et al. Serum and urinary androgens and risk of breast cancer in postmenopausal women. Cancer Res 1991; 51: 2572-6. 7. Schapira DV, Kumar NB, Lyman GH, Cox CE. Abdominal obesity and breast cancer risk. Ann Int Med 1990; 112: 182-6. 8. Folsom AR, Kaye SA, Prineas PJ et al. Increased incidence of carcinoma of the breast associated with abdominal adiposity in postmenopausal women. Amer J Epidem 1990; 131:794-803. 9. Berstein LM. Increased risk of breast cancer in women with central obesity; additional considerations. J Nat Cancer Inst 1990; 82: 1943-4. 10. Ballard-Barbash R, Schatzken A, Carter CL et al. Body fat distribution and breast cancer in the Framingham study. J Nat Cancer Inst 1990; 82: 286-90. 11. De Waard F. Breast cancer incidence and nutritional status with particular reference to body weight and height. Cancer Res 1975; 35: 3351-6. 12. Swanson CA, Jones DY, Schatzkin A et al. Breast cancer risk assessed by anthropometry in the NHANES; epidemiologic follow up study. Cancer Res 1988; 48: 5363-7. 13. Tornberg SA, Holm LE, Carstensen JM. Breast cancer risk in relation to serum cholesterol, serum beta lipoprotein, height, weight and blood pressure. Acta Oncol 1988; 27: 31-7. 14. Tretli S. Height and weight in relation to breast cancer morbidity and mortality; a prospective survey of 570,000 women in Norway. Int J Cancer 1989; 44: 23-30. 15. London SJ, Colditz GA, Stampfer MJ. Prospective study of relative weight, height and risk of breast cancer. J Am Med Assoc 1989; 262: 2853-8. 16. Vatten LJ, Kvinnsland S. Body height and risk of breast cancer, a prospective study of 23, 831 Norwegian women. Br J Cancer 1990;61:881-5. 17. Bulbrook RD, Leake RE, George WD. Oestrogens in the initiation and promotion of breast cancer. Proc Roy Soc Edin 1989; 95B: 67-76.
438 18. Zumoff B. Hormonal profiles in women with breast cancer. Anticancer Res 1988; 8:627-36. 19. Grattarola R. Ovariectomy alone or in combination with dexamethasone in patients with advanced breast cancer and high levels of testosterone excretion. J Nat Cancer Inst 1976; 56: 11-5. 20. MacFadyen IJ, Prescott RJ, Groom GV et al. Circulating hormone concentrations in women with breast cancer. Lancet 1976; 1:100-2. 21. Adami H-O, Johansson EDB, Vegelins J et aJ. Serum concentrations of estrone, androstenedione, testosterone and SHBG in postmenopausal women with breast cancer and in agematched controls. Upps J Med Sci 1979; 84: 259-75. 22. Secreto G, Fariselli G, Bandieramonte G et al. Androgen excretion in women with a family history of breast cancer or with epithelial hyperplasia or cancer of the breast. Eur J Cancer Clin Oncol 1983; 19:5-10. 23. Hill P, Garbaczewski L, Kasumi F. Plasma testosterone and breast cancer. Eur J Cancer Clin Oncol 1985; 21: 1265-6. 24. Secreto G, Zumoff B. Paradoxical effects associated with supranormal urinary testosterone excretion in premenopausal women with breast cancer. Increased risk of postmastectomy recurrence and higher remission rate after ovariectomy. Cancer Res 1983; 43: 3408-11. 25. Bulbrook RD, Thomas BS. Hormones are ambiguous risk factors for breast cancer. Acta Oncol 1989; 28: 841-7. 26. Kirschner MA. Obesity, androgens, oestrogens and cancer risk. Cancer Res 1982; 42 (suppl): 3281-5. 27. Kirschner MA, Samolijk E, Drejka M et al. Androgen- estrogen metabolism in women with upper body vs lower body obesity. J Clin Endocrinol Metab 1990; 70:473-9. 28. Evans DJ, Hoffman RG, Kalkhoff RJ et al. Relationship of androgenic activity to body fat topography, fat cell morphology and metabolic aberrations in premenopausal women. J Clin Endoc Metab 1983; 57: 304-9. 29. Schapira DV, Kumar NB, Lyman GH. Obesity, body fat distribution and sex hormones in breast cancer patients. Cancer 1991; 67: 2215-8. 30. Rissebak AH, Peiris AN. Biology of regional body fat distribution; relationship to non-insulin-dependent diabetes mellitus. Diabetec Metab Rev 1989; 5: 83-109. 31. Byers TE, Williamson DR. Diet, alcohol, body size and prevention of breast cancer. In Stoll BA (ed): Approaches to Breast Cancer Prevention. Kluwer Academic Publishers 1991; 113-34. 32. Cullen KJ, Smith HS, Hill S et al. Growth factor messenger RNA expression by human breast fibroblasts from benign and malignant lesions. Cancer Res 1991; 51:4978-85. 33. Pekonen F, Partanen S, Makinen T et al. Receptors for epidermal growth factor and insulin growth factor 1 and their relation to steroid receptors in human breast cancer. Cancer Res 1988; 48: 1343-7. 34. Darendeliler F, Hindmarsh PC, Preece MR. Growth hormone increases rate of pubertal maturation. Acta Endoc 1990; 122: 414-6. 35. Ratcliffe S. The clinical consequences of an extra sex chromosome. Growth Matters 1991; 6: 2-4. 36. Rosenfield RJ, Furlanetto R, Bock D. Relationship of somatomedin C concentration to pubertal changes. J Pediat 1983; 103:723-8. 37. Waxman J. The endocrinology of malignancy. In Sikora K, Hainan KE (eds): Treatment of Cancer. London: Chapman and Hall 1990; 147-52. 38. Holly JMP, Smith CP, Dunger DP et al. Relationship between the pubertaJ fall in SHBG and IGF-binding protein I. Clin Endoc 1989; 31: 277-84.
39. Clemmons DR. Structural and function analysis of insulin-like growth factors. In Waterfield MD (ed): Growth Factors. London: Churchill Livingstone 1989; 465-80. 40. Barbieri RL, Makris A, Randall RW et al. Insulin stimulates androgen accumulation in incubations of ovarian stroma obtained from women with hyperandrogenism. J Clin Endoc Metab 1986; 62: 904-10. 41. Burghen GA, Givens JR, Kitabachi AE. Correlation of hyperandrogenism with hyperinsulinaemia in polycystic ovarian disease. J Clin Endoc Metab 1980; 50: 113-6. 42. Kazer RR, Unterman TG, Glick RP Abnormality of the GH/ IGF1 axis in women with polycystic ovary syndrome. J Clin Endoc Metab 1990; 71: 958-62. 43. Moller DE, Flier JS. Insulin resistance: mechanisms, syndromes and implications. New Engl J Med 1991; 325: 838-48. 44. Coulam CB, Annegers JF, Kranz JJ. Chronic anovulation syndrome and associated neoplasms. Obstet Gynec 1983; 61: 403-7. 45. Wallace RB, Sherman BM, Bean JA. Relative weight and breast cancer. In Bulbrook RD, Taylor DJ (eds): Commentaries on Research in Breast Disease, Vol 3. New York: Alan R. Liss 1983; 163-79. 46. Dunaif A, Graf M. Insulin administration alters gonadal steroid metabolism independent of changes in gonadotropin secretion in women with the polycystic ovary syndrome. J Clin Invest 1989; 83: 23-9. 47. Barbien RL. Polycystic ovarian disease. Ann Rev Med 1991; 42: 199-204. 48. De Ridder CM, Bruning PF, Zonderland ML et al. Body fat mass, body fat distribution and plasma hormones in early puberty in females. J Clin Endoc Metab 1990; 70: 888-93. 49. Pryor M, Slattery ML, Robison LM, Egger M. Adolescent diet and breast cancer in Utah. Cancer Res 1989; 49: 2161 -7. 50. Nagamani M, van Dinh T, Kelver ME. Hyperinsulinaemia in hyperthecosis of the ovary. Am J Obstet Gynec 1986; 154: 384-9. 51. Frisch RE, McArthur JW. Menstrual cycles; fatness as a determinant of minimum weight for height necessary for their maintenance or onset. Science 1974; 185:949-51. 52. Preece MA. The trend to greater height and earlier maturation. Growth Matters 1989; 1:3-4. 53. Ewertz M. Risk factors for breast cancer and their prognostic significance. Acta Oncol 1988; 27: 733-7. 54. Apter D, Reinila M, Vikho R. Some endocrine characteristics of early menarche, a risk factor for breast cancer, are preserved into adulthood. Int J Cancer 1989; 44: 783-7. 55. Key TJA, Wang DY, Allen DS et al. Endocrine characteristics of early menarche. Int J Cancer 1991; 47: 317-8. 56. Bernstein L, Pike MC, Ross RK, Henderson BE. Age of menarche and estrogen concentrations of adult women. Cancer Causes and Control 1991; 2: 221-5. 57. De Waard F, Trichopoulos DA. Unifying concept of the aetiology of breast cancer. Int J Cancer 1988; 41:666-9. 58. Rudd BT. Personal communication 1991. Received 19 November 1991; accepted 12 February 1992. Correspondence to: Basil A. Stoll, FRCR Oncology Department St. Thomas' Hospital London SE1 7EH United Kingdom
Arena New hormone-related markers of high risk to breast cancer B. A. Stoll1 & G. Secreto2 1
Oncology Department, St. Thomas' Hospital, London, UK; 2 Laboratory of Hormonal Research, Istituto Nazionale Tumori, Milan, Italy
Summary. New markers of increased risk to breast cancer are examined and related to established risk markers. The following new evidence is highlighted: (1) Increased testosterone secretion by the ovaries is currently the only major steroid abnormality shown to be associated with increased risk of both premenopausal and postmenopausal breast cancer. (2) Upper body-type obesity is a marker for both hyperandrogenaemia and hyperinsulinaemia and is associated with an increased risk of breast cancer. Upper body type obesity may already be recognised in early puberty in Caucasian girls and is associated with a characteristic androgen/oestrogen profile. (3) Relative tallness in women is associated with an increased risk of breast cancer. A hypothesis is offered on the
significance of these markers in the aetiology of mammary cancer in women, and also a means of testing the hypothesis. The hormonal promotion of mammary carcinogenesis is likely to be greatest between puberty and the first full term pregnancy. The presence of hyperinsulinaemia can increase the ovarian production of androgen, and the abnormal hormonal profile may stimulate proliferative activity in mammary epithelium. This may increase the risk of epithelial atypia and carcinogenesis.
We need markers of high risk to breast cancer if we are to target screening and protective measures to selected subgroups of women, and also provide individuals seeking advice with an estimate of their personal risk. Recent research suggests that the development of human mammary cancer involves oncogene activation leading to upregulation of stimulating growth factors or their receptors, in addition to a promoting action by steroid hormones on previously initiated cells [1]. Currently, the best documented markers of increased risk are a family history of the disease and a group of hormone-related factors. The underlying hormonal interactions are uncertain but both laboratory [2] and epidemiological [3| evidence suggest that promotion of carcinogenesis is greatest during adolescence. A first pregnancy within a few years of the menarche reduces the risk of subsequent breast cancer by over 50%. The temporary reduction of incidence observed after ovarian ablation in adult life is generally assumed to result from growth dormancy in established cancer, rather than from decreased tumour promotion.
6], upper body-type obesity [7-10] and relative tallness [11-16]. They apply both to pre- and postmenopausal breast cancer cases but are more marked in the former category. These new markers may combine with new knowledge of peptide growth factor mechanisms to provide an insight into the causal mechanisms involved. Numerous studies have attempted to identify a hormonal profile associated with an increased risk of breast cancer. Oestrogens are widely believed to act as promoters in previously initiated mammary cells, but the 'oestrogen excess' hypothesis which held the stage for many years (recently with stress on the role of free oestradiol) is gradually being abandoned [17, 18]. An abnormally high serum testosterone level is the only disturbance of androgen secretion so far shown by case-control studies to be associated with both pre- and postmenopausal breast cancer [5, 6]. The abnormality had earlier been observed by several European laboratories [19-21]. That such hyperandrogenaemia is likely to be a risk marker rather than a result of the disease is suggested by the finding of a similar abnormality in women with mammary epithelial hyperplasia [22]. Abnormally high serum testosterone levels have also been observed in Japanese women with breast cancer [23]. Japanese women have in general, a much lower breast cancer incidence than do Caucasian women and also lower serum testosterone levels.
The classical, documented markers of increased risk differ between pre- and postmenopausal women. For premenopausal Caucasian women, they are a family history of the disease, the onset of menarche before the age of 13 and a delayed first full-term pregnancy, while for postmenopausal women, the major risk markers are onset of menopause delayed beyond the age of 49 and the presence of abnormal obesity [4]. In the last year or two, additional risk markers which have been confirmed are increased ovarian testosterone secretion |5,
Key words: breast cancer, risk markers, hyperandrogenaemia, hyperinsulinaemia
The abnormally increased testosterone secretion found in breast cancer patients is abolished by oophorectomy [19, 24] and is assumed to originate in the ovary. High levels of ovarian testosterone may be as-
436
sociated with decreased progesterone secretion and prolonged menstrual cycles in these patients [5], and a subclinical ovarian abnormality is suspected. This is likely to be stromal luteinisation, a condition allied to the polycystic ovary syndrome. Other observers have noted low levels of adrenal androgen in some premenopausal breast cancer patients (reviewed in [18]). In the Channel Islands study where younger (but not older) premenopausal breast cancer patients showed androgen levels at the lower end of the normal range, the observers regard this as a reflection of rapidly growing tumour but not as an indicator of increased predisposition [25]. A clearly-related new line of evidence is the observation by several groups that upper body-type obesity is a marker of increased risk of breast cancer in women [7-10]. For many years it was assumed that obese women are hyperoestrogenic because adipose tissue is a major extragonadal source of oestrogen [26]. However, it is confirmed that women with upper body (male type) fat distribution show high free testosterone levels, sometimes associated with prolonged menstrual cycles [27-29]. They also show metabolic abnormalities such as hyperinsulinaemia and subclinical diabetes [30]. It is believed that hyperinsulinaemia synergises with luteinising hormone in the development of hyperandrogenaemia and that the upper body-type obesity is a marker [27]. The condition is probably geneticallylinked and associated with an abnormal androgen/ oestrogen secretion ratio in early life [30]. It is probably an endocrinopathy allied to the polycystic ovary syndrome. Although high relative body weight is a classical marker of increased breast cancer risk, this does not apply to premenopausal women. While heavier postmenopausal women show between 30% and 60% increase in risk, it has recently been confirmed that heavier premenopausal women have a decreasedriskof breast cancer [31]. Another anthropometric marker of breast cancer risk is tallness, as first pointed out by de Waard [11] and recently confirmed by five other prospective studies [12—16]. Three of the studies involve more than 500,000 women. The Swedish study [13] showed the risk to be increased by 10% for every 5 cm of additional height, the effect being greater in premenopausal women. The Norwegian study [14] showed a 40% increased risk for 15 cm additional height and this applied both to pre- and postmenopausal women. The USA study [15] showed that an 8 cm increase in height increased risk in premenopausal women by 10% and in postmenopausal women by 30%. Evidence of these new risk markers combine with classical risk markers to identify a hormonal profile which may increase proliferative activity in pubertaJ mammary epithelium. Factors which increase proliferation also increase promotion of malignant change and this is most likely to occur at the time of maximum proliferative activity - puberty and adolescence [2]. While epithelial proliferation in the breast can be stimulated
by ovarian steroids, there is increasing evidence that polypeptide growth factors can mediate and modulate the steroid action. Transforming growth factors (TGFa, TGFb), epidermal growth factor (EGF), fibroblast growth factor (FGF) and insulin-like growth factors (IGF1, IGF11) may all play a part. Of these, IGF1 and IGF 11 have the most clearly defined mitogenic effect on mammary cancer cells and it is possible that they promote carcinogenesis in human mammary tissue [32]. The ductal epithelium of normal breast tissue contains abundant IGF1 receptors [33]. The bioactivity of IGF1 in the cell is regulated by a group of specific binding proteins (IGF BP) which in turn are related to circulating levels of growth hormone and insulin. The following observations suggest a role for IGF1 in mammary development at the time of puberty; the administration of recombinant growth hormone has been shown to accelerate breast development in prepubertal girls with isolated growth hormone deficiency [34]; the first onset of breast development in girls coincides with the onset of the adolescent growth spurt and this occurs 2-3 years before the major secretion of ovarian steroids at the menarche [35]; circulating IGF levels peak at puberty and correlate better with breast size than with chronological age [36]. The presence of excess androgen levels at puberty could also stimulate epithelial proliferation in the breast. While the hyperandrogenaemia found in a subgroup of breast cancer patients may merely be a marker of abnormal steroid metabolism in the ovary, it could also stimulate proliferative activity either directly or after aromatisation to oestrogen [37]. As noted above, the mean age at onset of breast development in Western girls in 2-3 years before the menarche [35] and probably involves androgen derived from the adrenal cortex. The concomitant rise in IGF1 level at the time of the growth spurt may stimulate the aromatase in mammary tissue, leading to local oestrogen synthesis [37]. In normal girls, puberty is associated with elevated circulating levels of growth hormone, insulin, IGF1 and testosterone [38] and with so many changes occurring at the same time, causal relationships are uncertain. Although the traditional assumption is that the rise in sex steroid levels at puberty increases the IGF levels [36], hyperinsulinaemia can stimulate both IGF1 and sex steroid activity [38]. Normal ovarian tissue contains abundant IGF1 receptors and IGF1 is said to synergise with gonadotropin in stimulating steroidogenesis in the granulosa cells [39]. Certainly in the case of women with ovarian hyperandrogenism, both IGF1 and insulin have been shown to stimulate testosterone release from samples of ovarian stroma, and the presence of hyperinsulinaemia is therefore thought to contribute to the development of hyperandrogenaemia in this condition [40]. In the case of polycystic ovary syndrome (similarly characterised by hyperandrogenaemia and hyperinsulinaemia) increased activity of the growth hormone/IGF 1 axis is
437
thought to play an important role |41—43]. Frank polycystic disease of the ovary is uncommon but it is noteworthy that an increased breast cancer risk is reported [44, 45]. It has been suggested that in this condition, the hyperinsulinaemia synergises with the increased levels of luteinising hormone to increase ovarian androgen secretion, mainly by upregulating the IGF1 receptor [46, 47]. It may be possible to recognise girls at high risk to breast cancer at the time of puberty. The presence of upper body-type obesity in adult women is associated with increased breast cancer risk [7-10] and upper body-type obesity with a characteristic androgen/ oestrogen ratio in the blood can already be recognised in girls between 10 and 11 years [48]. In a case-control study [49], a large body-mass index at the age of 12 was shown to be an independent risk marker for breast cancer but no details are given as to body fat distribution. While body mass is strongly influenced by nutrition, genetically-linked body fat distribution is likely to be a marker of the endogenous hormonal pattern which influences breast cancer risk. What is the relationship of childhood nutrition to obesity and the risk of developing breast cancer? In women with genetically-linked ovarian hyperandrogenism (allied to polycystic ovary syndrome) obesity may trigger severe hyperinsulinaemia leading to hyperandrogenaemia [47]. Similarly, in women with upper body-type obesity (also allied to polycystic ovary syndrome) the hyperinsulinaemia and hyperandrogenaemia may result from genetic abnormality [43]. However, recent reviews suggest that any cause of severe hyperinsulinaemia may play a key role in the development of ovarian hyperandrogenaemia [43,47, 50]. With regard to the association of tallness with increased breast cancer risk, it is widely accepted that better nutrition accelerates growth hormone release and that a critical body mas triggers the hypothalamic mechanism responsible for the onset of sexual maturity [51]. This may lead to early elevation of growth hormone, insulin and IGF1 levels in addition to sex steroid levels. An earlier adolescent growth spurt favours tallness in adult women [52). The association of early menarche with increased breast cancer risk cannot necessarily be assumed to involve hyperinsulinaemia and hyperandrogenaemia. Menarche before the age of 13 increases breast cancer risk with a highly significant trend to decreasing risk with each succeeding year after that age |53|, but there is equivocal evidence for a distinctive steroid profile in later years in girls manifesting an early menarche [54—56). An earlier menarche may permit a longer period for hormonal stimulation of proliferative activity in the mammary epithelium before it is counteracted by the first full-term pregnancy [57). A hypothesis is offered on the role of these new hormonal markers in the aetiology of mammary cancer in women. The presence of hyperinsulinaemia may increase ovarian production of androgen, or else, in-
creased levels of insulin-like growth factor may synergise with sex steroids in a mitogenic effect on developing mammary tissue in adolescence. The risk of epithelial atypia and carcinogenesis may be increased as a result. If major promotion of mammary carcinogenesis occurs in adolescence, is is possible to test the hypothesis that the association of hyperinsulinaemia and an abnormal ovarian steroid profile may stimulate mitogenesis in adolescent breast tissue. Studies are already under way to monitor circulating levels of IGF1, oestrogen and androgen in girls approaching puberty, and to relate them to observed changes in breast development and statural growth [58]. It would be of considerable interest to relate them also to fat distribution patterns. References 1. Barrett-Lee PJ. Growth factor expression in breast tissue. In Stoll BA (ed): Approaches to Breast Cancer Prevention. Dordrecht: Kluwer Academic Publishers 1991; 53-60. 2. Russo JH, Calaf G, Russo J. Hormones and proliferative activity in breast tissue. In Stoll BA (ed): Approaches to Breast Cancer Prevention. Dordrecht: Kluwer Academic Publishers 1991;35-51. 3. Cole P, MacMahon B. Oestrogen fractions during early reproductive life in aetiology of breast cancer. Lancet 1969; 1: 604-6. 4. Stoll BA. Prospects for breast cancer prevention. In Stoll BA (ed): Women at High Risk to Breast Cancer. Dordrecht: Kluwer Academic Publishers 1989; 107-19. 5. Secreto G, Toniolo P, Pisani P et al. Androgens and breast cancer in premenopausal women. Cancer Res 1989; 49:471-6. 6. Secreto G, Toniolo P, Berrino F et al. Serum and urinary androgens and risk of breast cancer in postmenopausal women. Cancer Res 1991; 51: 2572-6. 7. Schapira DV, Kumar NB, Lyman GH, Cox CE. Abdominal obesity and breast cancer risk. Ann Int Med 1990; 112: 182-6. 8. Folsom AR, Kaye SA, Prineas PJ et al. Increased incidence of carcinoma of the breast associated with abdominal adiposity in postmenopausal women. Amer J Epidem 1990; 131:794-803. 9. Berstein LM. Increased risk of breast cancer in women with central obesity; additional considerations. J Nat Cancer Inst 1990; 82: 1943-4. 10. Ballard-Barbash R, Schatzken A, Carter CL et al. Body fat distribution and breast cancer in the Framingham study. J Nat Cancer Inst 1990; 82: 286-90. 11. De Waard F. Breast cancer incidence and nutritional status with particular reference to body weight and height. Cancer Res 1975; 35: 3351-6. 12. Swanson CA, Jones DY, Schatzkin A et al. Breast cancer risk assessed by anthropometry in the NHANES; epidemiologic follow up study. Cancer Res 1988; 48: 5363-7. 13. Tornberg SA, Holm LE, Carstensen JM. Breast cancer risk in relation to serum cholesterol, serum beta lipoprotein, height, weight and blood pressure. Acta Oncol 1988; 27: 31-7. 14. Tretli S. Height and weight in relation to breast cancer morbidity and mortality; a prospective survey of 570,000 women in Norway. Int J Cancer 1989; 44: 23-30. 15. London SJ, Colditz GA, Stampfer MJ. Prospective study of relative weight, height and risk of breast cancer. J Am Med Assoc 1989; 262: 2853-8. 16. Vatten LJ, Kvinnsland S. Body height and risk of breast cancer, a prospective study of 23, 831 Norwegian women. Br J Cancer 1990;61:881-5. 17. Bulbrook RD, Leake RE, George WD. Oestrogens in the initiation and promotion of breast cancer. Proc Roy Soc Edin 1989; 95B: 67-76.
438 18. Zumoff B. Hormonal profiles in women with breast cancer. Anticancer Res 1988; 8:627-36. 19. Grattarola R. Ovariectomy alone or in combination with dexamethasone in patients with advanced breast cancer and high levels of testosterone excretion. J Nat Cancer Inst 1976; 56: 11-5. 20. MacFadyen IJ, Prescott RJ, Groom GV et al. Circulating hormone concentrations in women with breast cancer. Lancet 1976; 1:100-2. 21. Adami H-O, Johansson EDB, Vegelins J et aJ. Serum concentrations of estrone, androstenedione, testosterone and SHBG in postmenopausal women with breast cancer and in agematched controls. Upps J Med Sci 1979; 84: 259-75. 22. Secreto G, Fariselli G, Bandieramonte G et al. Androgen excretion in women with a family history of breast cancer or with epithelial hyperplasia or cancer of the breast. Eur J Cancer Clin Oncol 1983; 19:5-10. 23. Hill P, Garbaczewski L, Kasumi F. Plasma testosterone and breast cancer. Eur J Cancer Clin Oncol 1985; 21: 1265-6. 24. Secreto G, Zumoff B. Paradoxical effects associated with supranormal urinary testosterone excretion in premenopausal women with breast cancer. Increased risk of postmastectomy recurrence and higher remission rate after ovariectomy. Cancer Res 1983; 43: 3408-11. 25. Bulbrook RD, Thomas BS. Hormones are ambiguous risk factors for breast cancer. Acta Oncol 1989; 28: 841-7. 26. Kirschner MA. Obesity, androgens, oestrogens and cancer risk. Cancer Res 1982; 42 (suppl): 3281-5. 27. Kirschner MA, Samolijk E, Drejka M et al. Androgen- estrogen metabolism in women with upper body vs lower body obesity. J Clin Endocrinol Metab 1990; 70:473-9. 28. Evans DJ, Hoffman RG, Kalkhoff RJ et al. Relationship of androgenic activity to body fat topography, fat cell morphology and metabolic aberrations in premenopausal women. J Clin Endoc Metab 1983; 57: 304-9. 29. Schapira DV, Kumar NB, Lyman GH. Obesity, body fat distribution and sex hormones in breast cancer patients. Cancer 1991; 67: 2215-8. 30. Rissebak AH, Peiris AN. Biology of regional body fat distribution; relationship to non-insulin-dependent diabetes mellitus. Diabetec Metab Rev 1989; 5: 83-109. 31. Byers TE, Williamson DR. Diet, alcohol, body size and prevention of breast cancer. In Stoll BA (ed): Approaches to Breast Cancer Prevention. Kluwer Academic Publishers 1991; 113-34. 32. Cullen KJ, Smith HS, Hill S et al. Growth factor messenger RNA expression by human breast fibroblasts from benign and malignant lesions. Cancer Res 1991; 51:4978-85. 33. Pekonen F, Partanen S, Makinen T et al. Receptors for epidermal growth factor and insulin growth factor 1 and their relation to steroid receptors in human breast cancer. Cancer Res 1988; 48: 1343-7. 34. Darendeliler F, Hindmarsh PC, Preece MR. Growth hormone increases rate of pubertal maturation. Acta Endoc 1990; 122: 414-6. 35. Ratcliffe S. The clinical consequences of an extra sex chromosome. Growth Matters 1991; 6: 2-4. 36. Rosenfield RJ, Furlanetto R, Bock D. Relationship of somatomedin C concentration to pubertal changes. J Pediat 1983; 103:723-8. 37. Waxman J. The endocrinology of malignancy. In Sikora K, Hainan KE (eds): Treatment of Cancer. London: Chapman and Hall 1990; 147-52. 38. Holly JMP, Smith CP, Dunger DP et al. Relationship between the pubertaJ fall in SHBG and IGF-binding protein I. Clin Endoc 1989; 31: 277-84.
39. Clemmons DR. Structural and function analysis of insulin-like growth factors. In Waterfield MD (ed): Growth Factors. London: Churchill Livingstone 1989; 465-80. 40. Barbieri RL, Makris A, Randall RW et al. Insulin stimulates androgen accumulation in incubations of ovarian stroma obtained from women with hyperandrogenism. J Clin Endoc Metab 1986; 62: 904-10. 41. Burghen GA, Givens JR, Kitabachi AE. Correlation of hyperandrogenism with hyperinsulinaemia in polycystic ovarian disease. J Clin Endoc Metab 1980; 50: 113-6. 42. Kazer RR, Unterman TG, Glick RP Abnormality of the GH/ IGF1 axis in women with polycystic ovary syndrome. J Clin Endoc Metab 1990; 71: 958-62. 43. Moller DE, Flier JS. Insulin resistance: mechanisms, syndromes and implications. New Engl J Med 1991; 325: 838-48. 44. Coulam CB, Annegers JF, Kranz JJ. Chronic anovulation syndrome and associated neoplasms. Obstet Gynec 1983; 61: 403-7. 45. Wallace RB, Sherman BM, Bean JA. Relative weight and breast cancer. In Bulbrook RD, Taylor DJ (eds): Commentaries on Research in Breast Disease, Vol 3. New York: Alan R. Liss 1983; 163-79. 46. Dunaif A, Graf M. Insulin administration alters gonadal steroid metabolism independent of changes in gonadotropin secretion in women with the polycystic ovary syndrome. J Clin Invest 1989; 83: 23-9. 47. Barbien RL. Polycystic ovarian disease. Ann Rev Med 1991; 42: 199-204. 48. De Ridder CM, Bruning PF, Zonderland ML et al. Body fat mass, body fat distribution and plasma hormones in early puberty in females. J Clin Endoc Metab 1990; 70: 888-93. 49. Pryor M, Slattery ML, Robison LM, Egger M. Adolescent diet and breast cancer in Utah. Cancer Res 1989; 49: 2161 -7. 50. Nagamani M, van Dinh T, Kelver ME. Hyperinsulinaemia in hyperthecosis of the ovary. Am J Obstet Gynec 1986; 154: 384-9. 51. Frisch RE, McArthur JW. Menstrual cycles; fatness as a determinant of minimum weight for height necessary for their maintenance or onset. Science 1974; 185:949-51. 52. Preece MA. The trend to greater height and earlier maturation. Growth Matters 1989; 1:3-4. 53. Ewertz M. Risk factors for breast cancer and their prognostic significance. Acta Oncol 1988; 27: 733-7. 54. Apter D, Reinila M, Vikho R. Some endocrine characteristics of early menarche, a risk factor for breast cancer, are preserved into adulthood. Int J Cancer 1989; 44: 783-7. 55. Key TJA, Wang DY, Allen DS et al. Endocrine characteristics of early menarche. Int J Cancer 1991; 47: 317-8. 56. Bernstein L, Pike MC, Ross RK, Henderson BE. Age of menarche and estrogen concentrations of adult women. Cancer Causes and Control 1991; 2: 221-5. 57. De Waard F, Trichopoulos DA. Unifying concept of the aetiology of breast cancer. Int J Cancer 1988; 41:666-9. 58. Rudd BT. Personal communication 1991. Received 19 November 1991; accepted 12 February 1992. Correspondence to: Basil A. Stoll, FRCR Oncology Department St. Thomas' Hospital London SE1 7EH United Kingdom
