PERIPHERAL HORMONE LEVELS AND THE ENDOMETRIAL CONDITION IN POSTMENOPAUSAL WOMEN
S. Brody, K. Carlström, A .- K. von Uexküll, A. Lagrelius, N .- O. Lunell and L. Rosenborg
From the Department of Obstetrics and Gynecology, Karolinska Institutet, Huddinge University Hospital, Huddinge, Sweden
Abstract. Thirteen postmenopausal women with benign en- dometrial changes including proliferative, secretory and polypous endometrium, endometrial hyperplasia and atypia (group I) and 13 randomly selected age-matched controls with normal atrophic endometrium (group II) were studied with respect to serum levels of dehydroepiandrosterone (DHA) and its sulfate (DHAS), testosterone, total estrone, estradiol-178, progesterone, FSH and prolactin. Serum lev- els of DHA, DHAS, testosterone and total estrone were sig- nificantly higher in group I than in group II; otherwise no significant differences were found. Mean values for body weight and for Broca’s index, respectively, were almost identical in the two groups.
It is speculated that the adrenal androgens may affect the endometrium in two ways, viz. via peripheral conversion to estrogens and/or via direct interaction with endometrial steroid receptors. The results give further support to the hypothesis of an association between adrenocortical hyper- activity and endometrial abnormalities including endometri- al carcinoma.
The concept of endometrial carcinoma as an endo- crine-related disease in the postmenopausal woman is generally accepted (24). The genesis of endometrial carcinoma has been suggested to include a progres- sion of changes in the endometrium from benign pro- liferation to cystic hyperplasia, adenomatous hyper- plasia and varying degrees of anaplasia, including in- vasive adenocarcinoma (18). Unopposed estrogen stimulation of the endometrium is usually considered to be the underlying mechanism in this chain of events; however, androgens (11) and prolactin (24) have also been discussed in this respect. Most authors describe elevated levels of estrogens (2, 4, 5, 11, 19, 28, 30) and also androgens (5, 10, 11, 16, 19) in postmenopausal patients with endometrial carcinoma compared with healthy controls; however, absence of differences (21, 22) or depressed androgen levels (3, and references cited therein) have also been reported. Adrenocortical hyperplasia and the polycystic ovary syndrome are two hyperandrogenic conditions associated with endometrial carcinoma (24).
As is the case in endometrial carcinoma, elevated estrogen levels have also been reported in postmeno- pausal women with endometrial lesions preceding this disease (2, 5, 19, 28, 30). Based upon estrogen determinations before and after oophorectomy (30) and of steroid measurements in ovarian vein blood (5) it was postulated that elevated estrogen levels in women with precancerous lesions were due to ovarian hyperactivity. Considerable amounts of testosterone, amounting to 50% of the levels in the peripheral cir- culation, arise from steroidogenic activity of the ovarian stroma even after the menopause (12). Ele- vated testosterone levels have also been associated with precursors of endometrial carcinoma (5, 16, 19), which may be considered as a support for the hypo- thesis of ovarian hyperactivity. However, the circu- lating levels of estrone and estradiol-170 are reported to have almost exclusive (indirect) adrenal origin in the postmenopausal woman (34). Increased estrogen levels may therefore indicate adrenal rather than ovarian hyperactivity in postmenopausal women.
In previous communications (6,7) we described a significant positive correlation between basal serum levels of dehydro-epiandrosterone sulfate (DHAS) and the response in dehydroepiandrosterone (DHA) to ACTH, respectively, and trabecular bone mineral content of the distal forearm in postmenopausal women. Furthermore, a significant correlation was also found between bone mineral content and a pre- cancerous/cancerous state of the uterine epithelium. These findings indicate an association between the adrenal androgens DHA and DHAS and endometrial pathology. The present communication describes the measurements of serum levels of DHA and DHAS in a group of postmenopausal women having endomet- rial lesions considered to be precursors to endometrial carcinoma and in an age matched control group. Together with DHA and DHAS, testosterone, estro- gens, progesterone, prolactin and FSH were also measured.
MATERIAL AND METHODS
Clinical material
The clinical material comprised postmenopausal women aged 49 - 59 (mean 53.7 years). They had no clinical signs of hepatic, biliary or renal malfunction and were not taking any medication known to interfere with the hormone valu- es. None of the subjects had undergone estrogen replace- ment therapy. All had normal postmenopausal serum levels of FSH. Venous blood samples were collected between 8.30 a.m. and 10 a.m., before anesthesia was started. After cent- rifugation, serum was withdrawn and stored at - 20℃ until analysed.
Group I consisted of 13 women, of whom 11 underwent di- agnostic curettage for postmenopausal bleeding and 2 for excessive ovarian mass. One had a proliferative, 1 a secre- tory and 6 a polypous endometrium; 2 had a hyperplastic and 3 an atypic endometrium.
Group II consisted of 13 randomly selected age-matched women with a normal, atrophic endometrium. Ten under- went diagnostic curettage for postmenopausal bleeding, 1 due to atypical cervical smear, 1 due to enlarged uterus and 1 due to uterine prolaps.
Hormonal analysis
Serum levels of DHAS, estradiol-176, total estrone (mainly estrone sulfate), progesterone, FSH and prolactin were de- termined by radioimmunological methods described previ- ously (11, 20, 25). The anti-estradiol-173-6-(0-carboxymeth- yl oxime)-bovine serum albumine used in the estradiol-178 assay cross-reacts with estrone to 50% at an estrone concen- tration of 120 pmol/1, to 34% at 370 pmol/l and to 10% at 2 000 pmol/1 (cf 25). The values are expressed as pmol im- munoreactive estradiol-176 equivalents/1 and are referred to as estradiol-176 in the text.
DHA was measured after extraction with diethyl ether, us- ing the same radioimmunological system as for DHAS (11). Testosterone was determined radioimmunologically after ether extraction, using a commercial kit from Diagnostic Products Corp., Los Angeles, Calif. Bound and free [125]]
testosterone were separated by a double antibody-polyethyl- ene glycol precipitation technique. Cross-reacting steroids account for approximately 10-15% of the values obtained. Progesterone was determined after extraction with n-hexane using a commercial kit from Nordiclab Oy, Oulu, Finland, with minor modifications. Bound and free [125]] progester- one were separated by precipitation with polyethylene gly- col.
Statistical methods
Progesterone, body weight and Broca’s index showed a nor- mal distribution, while other variables were lognormally distributed. Student’s t-test for unpaired observations was applied, with or without logarithmic transformation respec- tively.
RESULTS
Peripheral serum hormone levels, body weight and Broca’s index for the two groups of women are given in Table I. Mean values for DHA, DHAS, testostero- ne and total estrone were significantly higher in group I. Higher values were also noted for estradiol-178 in this group, although this difference was not statisti- cally significant. There were no differences in serum progesterone, prolactin and FSH, nor in body weight or Broca’s index.
DISCUSSION
In a recent review on adrenal androgens and endo- metrial disorders, Andriesse et al. (3) suggested an as- sociation between endometrial disorders, including endometrial carcinoma, and a decreased adrenal an- drogens synthesis. However, some ciritcism can be di- rected against several of the studies cited as well as
| Group I benign endometrial changes | Group II atrophic endometrium | Difference p-value | |
|---|---|---|---|
| DHA, nmol/l | 18.9 (8.2-51.8) | 9.6 (4.5-17.6) | p<0.01 |
| DHAS, nmol/l | 3 162 (1 418 - 7 660) | 1 714 (1 103 - 2 903) | p<0.01 |
| Testosterone, nmol/l | 1.14 (0.50-1.95) | 0.63 (0.30-1.80) | p<0.05 |
| Estradiol-173, pmol/l | 129 (50-315) | 87 (65-167) | n.s. |
| Total estrone, nmol/l | 1.67 (0.84-4.10) | 1.18 (0.53-1.80) | p<0.05 |
| Progesterone, nmol/l | 2.40±0.60 | 1.94±0.50 | n.s. |
| FSH, U/1a) | 45.0 (26.0-80.0) | 47.0 (32.0-74.0) | n.s. |
| Prolactin, µg/1b) | 5.2 (2.4-13.9) | 5.3 (2.2-8.2) | n.s. |
| Body weight, kg | 69.9±10.2 | 69.0±11.2 | n.s. |
| Broca's index | 1.09±0.16 | 1.12±0.18 | n.s. |
a) Human Pituitary FSH 68/39, b) Prolactin N.I.H. V.L.S.1.
against certain parts of the review itself. This includes lack of age data and/or not entirely satisfactory age- matching of controls and pathological subjects (8, 14, 31 and their own results presented in the review). The well known dramatic decrease in adrenal androgens between age 50 and 60 years requires strictly age- matched materials. Furthermore, in two of the inves- tigations (8, and their own results given in the review) samples were collected 3- 9 days after diagnostic cu- rettage or surgery. It has recently been shown that se- rum DHA and DHAS levels are depressed 3-5 days after surgery in postmenopausal women (1). Finally, it cannot be recommended to draw conclusions upon the basis of comparisons of mean values for patho- logical and healthy subjects, calculated from diffe- rent investigations on different subjects carried out in different laboratories using different methods, as has been done for androstenedione. The suggestion of Andriesse et al. (3) of a decreased adrenal androgen synthesis in endometrial disorders must therefore be considered as doubtful.
The results of the present investigation indicate an increased adrenal androgen production in postmeno- pausal women with benign endometrial changes. The elevated testosterone levels may also suggest an incre- ased activity of the ovarian stroma. Our results agree with previous reports on elevated testosterone levels in women constituting a risk group for endometrial carcinoma (5,16,19) and also with previous reports on elevated androgen levels in endometrial carcinoma (5,10,11,19). A tendency to higher testosterone levels in endometrial carcinoma was also reported by Cala- nog and co-workers (9), though, this difference was not statistically significant. No significant differences in testosterone and androstenedione between endo- metrial cancer patients and healthy controls were fo- und by Judd et al (22).
The present results give additional support to the hypothesis of an important role for the adrenal cortex in the etiology of endometrial carcinoma (6, 8, 24, 32) but do not exclude a contribution of the ovarian stro- ma in this respect (5, 12). The factor(s) responsible for the increased adrenal androgen levels in the group with endometrial disorders remain(s) enigmatic. Ex- cept for the endometrial pathology, our two groups are very similar from a clinical point of view. In- creased DHAS levels have been associated with hy- perprolactinemia (for references see 17); however, no differences in serum prolactin were found in the pres- ent study. In a previous work (7) we found no corre- lation between prolactin and DHA/DHAS in post-
menopausal women. Differences in ACTH stimula- tion or in the synthetic capacity of the adrenal zona reticularis are conceivable causes (13). The regulation of the adrenal androgen synthesis is not fully under- stood and may include hitherto unidentified pituitary factors such as an “adrenal androgen-stimulating hormone, AASH” (17).
The elevated total estrone levels found in a group with endometrial disorders is in accordance with pre- vious reports on estrogens in this category of patients (2, 5,19, 28, 30) as well as in endometrial carcinoma (2, 4, 5, 11, 19, 28, 30), an exception being the absence of differences in estrogen levels reported by Judd and co-workers (21, 22). The estrogens in the postmenopausal woman arise almost exclusively by peripheral aromatization of adrenal androgens (33, 34). Thus the estrogen production depends upon the supply of substrate, i.e. adrenal androgens, and the rate of aromatization. It is at present not known which of these two factors is the decisive one in this respect. The rate of aromatization has been related to the degree of obesity, to liver disease and to hyperthy- roidism (33). However, there were no differences in body weight or Broca’s index between our two groups of patients and they were free from liver and thyroid disease. Therefore it seems likely that the aromatizing capacity is similar in both groups, and that the adre- nal androgen levels may in fact be the decisive factor. It should be recalled that significant correlations be- tween estrogen and androgen levels have been report- ed in postmenopausal women (7, 13, 35).
The increased estrogen values, together with un- changed low progesterone levels may be one cause of the endometrial disorders in the present patient seri- es. If so, the role of the androgens is restricted to ser- ve as a substrate for the peripheral estrogen synthesis. However, a direct action of androgens on the endo- metrium cannot be excluded. There are reports on the existence of androgen-binding macromolecules in normal and pathologic uterine tissues (15,26). Fur- thermore - and this may be of even greater biolog- ical significance - certain androgens interact with the uterine estrogen receptor in an estrogen-like man- ner (23, 27, 29). The mode of action of androgens upon the human endometrium may therefore be mul- tifactorial and complex, and its details remain to be elucidated.
ACKNOWLEDGEMENT
This work was supported by grants from Riksföreningen mot cancer (Project 1288-B 80-02XB), from Karolinska In- stitutets Fonder and from the Folksam insurance company.
REFERENCES
1. Adami, H.O., Axelsson, O., Carlström, K., Vegelius, J & Åkerström, G .: Serum levels of cortisol, dehydroepi- androsterone, dehydroepiandrosterone sulphate, estro- ne and prolactin after surgical trauma in postmeno- pausal women. Upsala J Med Sci 87:201, 1982.
2. Aleem, F. A., Moukthar, M. A., Hin Cheung Hung & Romney, S. L .: Plasma estrogens in patients with endometrial hyperplasia and carcinoma. Cancer 38:2 101, 1976.
3. Andriesse, R., Thijssen, J. H. H., Donker, G. H. & Poortman, J .: Adrenal androgens and endometrial car- cinoma.In Adrenal Androgens (ed A.R. Genazzani, J. H. H. Thijssen & P. K. Siiteri), p. 355. Raven Press, New York, 1980.
4. Benjamin, F & Deutsch, S .: Plasma levels of fractiona- ted estrogens and pituitary hormones in endometrial carcinoma. Am J Obstet Gynecol 126:638, 1976.
5. Botella-Lluisa, J., Oriol-Bosch, A., Sanchez-Garrido, F & Treguerres, J. A. F .: Testosterone and 178-oestra- diol secretion of the human ovary. II. Normal post- menopausal women, postmenopausal women with endometrial hyperplasia and postmenopausal women with adenocarcinoma of the endometrium. Maturitas 2:7, 1979.
6. Brody, S., Carlström, K., Lagrelius, A., Lunell, N .- O. & Rosenborg, L .: Adrenal steroids, bone mineral con- tent and endometrial pathology in postmenopausal wo- men. Acta Obstet Gynecol Scand 60:325, 1981.
7. Brody, S., Carlström, K., Lagrelius, A., Lunell, N .- O. & Rosenborg, L .: Adrenocortical steroids, bone mineral content and endometrial condition in postmenopausal women. Maturitas 4:113, 1982.
8. Brush, M.G., Milton, P.J.D., Swain, M. & Moore, J .: Endocrine factors in the aetiology of cystic hyperplasia of endometrium. J Endocrinol 64:55, 1975.
9. Calanog, A., Sall, S., Gordon, C.G., Olivo, J. & Southren, A.L .: Testosterone metabolism in endomet- rial cancer. Am J Obstet Gynecol 124:60, 1976.
10. Calanog, A., Sall, S., Gordon, C.G. & Southren, A.L .: Androstenedione metabolism in endometrial cancer. Am J Obstet Gynecol 129:553, 1977.
11. Carlström, K., Damber, M .- G., Furuhjelm, M., Joels- son, I., Lunell, N .- O. & von Schoultz, B .: Serum levels of total dehydroepiandrosterone and total estrone in postmenopausal women with special regard to carcino- ma of the uterine corpus. Acta Obstet Gynecol Scand 58:179, 1979.
12. Chang, R.J. & Judd, H.J .: The ovary after the meno- pause. Clin Obstet Gynecol 24:181, 1981.
13. Crilly, R. G., Francis, R. M. & Nordin, B.E.C .: Steroid hormones, ageing and bone. Clin Endocrinol Metab 10:115, 1981.
14. De Waard, F., Thyssen, J.H.H., Veeman, W. & San- der, P.C .: Steroid hormone excretion pattern in women with endometrial carcinoma. Cancer 22:988, 1968.
15. Friberg, L. G., Kullander, S., Persijn, J. P. & Korsten, C. B .: On receptors for estrogens (E2) and androgens (DHT) in human endometrial carcinoma and ovarian tumors. Acta Obstet Gynecol Scand 57:261, 1978.
16. Grattarola, R., Secreta, G., Reccione, C. & Castellini, W .: Androgens in breast cancer. II. Endometrial adenocarcinoma and breast cancer in married post- menopausal women. Am J Obstet Gynecol 118:173, 1974.
17. Grumbach, M. M., Richards, G. E., Conte, F. A. & Kaplan, S. L .: Clinical disorders of adrenal function and puberty: An assessment of the role of the adrenal cortex in normal and abnormal puberty in man and evi- dence for an ACTH-like pitiutary adrenal androgen stimulating hormone. In The Endocrine Function of the Human Adrenal Cortex. (ed V. H. T. James, M. Se- rio, G. Giusti, & L. Martini), Serono Symposium, Vol. 18, p. 583, Academic Press, New York, 1978.
18. Gusberg, S .: Hormone dependence of endometrial car- cinoma. Obstet Gynecol 30:287, 1967.
19. Haldemann, R. & Gigon, U .: Serum FSH, LH, estradi- ol (E2), progesterone (P) and testosterone (T) levels in postmenopausal women with and without endometrial cancer. Cancer Treatment Repts 63:1 207, 1979.
20. Helgasson, S., Wilking, N., Carlström, K., Damber, M .- G. & von Schoultz, B .: A comparative study on the estrogenic effects of tamoxifen and estradiol-173 in postmenopausal women. J Clin Endocrinol Metab 54:404, 1982.
21. Judd, H. L., Lucas, W. E. & Yen, S. S. C .: Serum es- tradiol-17 and estrone levels in postmenopausal women with and without endometrial cancer. J Endocrinol Me- tab 43:272, 1976.
22. Judd, H. L., Davidson, B. J., Frumar, A. M., Sha- monki, I. M., Lagasse, L. D. & Ballon, S. C .: Serum androgens and estrogens in postmenopausal women with and without endometrial cancer. Am J Obstet Gynecol 136:859, 1980.
23. Kreitmann, B. & Bayard, F .: Androgen interaction with the oestrogen receptor in human tissues. J Steroid Biochem 11:1 589, 1979.
24. Lucas, W. E .: Causal relationship between endocrine- metabolic variables in patients with endometrial carci- noma. Obstet Gynecol Survey 29:507, 1974.
25. Lyrenäs, S., Carlström, K., Bäckström, T. & von Schoultz. B .: A comparisom of serum oestrogen levels after percutaneous and oral administration of oestra- diol-173. Br J Obstet Gynecol 88:181, 1981.
26. Menon, M., Tananis, C., Hicks, L. L., Hawkins, E. F., McLoughlin, M. G. & Walsh, P. C .: Characterization of the binding of a potent synthetic androgen, methyl- trienolone, to human tissues. J Clin Invest 61:150, 1978.
27. Poortman, J., Prenen, J. A. C., Schwarz, F. & Thij- ssen, J. H. H .: Interaction of 45-androstene-36, 173- diol with estradiol and dihydrotestosterone receptors in human myometrial and mammary cancer tissue. J Clin Endocrinol Metab 40:137, 1975.
28. Reti, L. L., Rome, R. M., Brown, J. B. & Fortune, D. W .: Urinary oestrogen and pregnanediol excretion, en- dometrial and ovarian pathology and body weight in women with postmenopausal bleeding. Br J Obstet Gynecol 85:857. 1978.
29. Rochefort, H., Capony, F. & Garcia,M .: Mechanism of action of antiestrogens and androgens on the estrogen receptor. J Steroid Biochem 11:1635, 1979.
30. Rome, M., Brown, J. B., Mason, T., Smith, M. A., Laverty, C. & Fortune, D .: Oestrogen excretion and ovarian pathology in postmenopausal women with atypical hyperplasia, adenocarcinoma and mixed adenosquamous carcinoma of the endometrium. Br J Obstet Gynecol 84, 88, 1977.
31. Sall, S. & Calanog, A .: Steroid excretion patterns in postmenopausal women with benign and neoplastic endometrium. Am J Obstet Gynecol 114:153, 1972.
32. Siiteri, P. K .: Steroid hormones and endometrial can- cer. Cancer Res 36:4360, 1978.
33. Siiteri, P. K. & Seron-Ferré, M .: Secretion and metab- olism of adrenal androgens to estrogens. In The Endo- crine Function of the Human Adrenal Cortex (ed. V. H. T. James, M. Serio, G. Giusti & L. Martini), Serono Symposium, Vol., p. 251, Academic Press, New York, 1978.
34. Vermeulen, A .: The hormonal activity of the post- menopausal ovary. J Clin Endocrinol Metab 42:247, 1976.
35. Vermeulen, A. & Verdonck, L .: Factors affecting sex hormone levels in postmenopausal women. J Steroid Biochem 11:899, 1979.
Submitted for publication July 21, 1982 Accepted December 7, 1982
S. Brody, M.D. Department of Obstetrics and Gynecology Huddinge University Hospital S-141 86 Huddinge Sweden