Feminization as a result of both peripheral conversion of androgens and direct estrogen production from an adrenocortical carcinoma
A. Zayed*, J.L. Stock*, M.K. Liepman*, M. Wollin ** , and C. Longcope ***
*Division of Endocrinology and Departments of Medicine, and ** Urology Medical Center of Central Massachusetts and *** Departments of Obstetrics and Gynecology and Medicine, University of Massachusetts Medical School, Worcester, Massachusetts, USA
ABSTRACT. A 45-year-old man presented with gy- necomastia, hypertension and a large left adrenal mass. Further evaluation revealed elevated serum concentrations of estrogen, estrone sulfate, an- drostenedione, dehydroepiandrosterone, dehy- droepiandrosterone sulfate, deoxycorticosterone, and aldosterone and increased 24-hour urinary 17-ke- tosteroid and free cortisol excretion. Removal of a 10 kg adrenocortical carcinoma led to normalization of the hormone concentrations and partial resolution of
the gynecomastia. There was no clinical evidence of metastases. Incubation of tumor slices demonstrated that the tumor had an active aromatase and sulfo- transferase. We estimated that about half the serum estrone arose from peripheral conversion of an- drostenedione. Feminizing adrenocortical carcino- mas are rare and this case is unusual given the lack of clinical metastases and the probable dual source of estrogen from tumor as well as from the peripher- al conversion of tumor-derived androgens.
INTRODUCTION
Feminizing adrenal tumors are rare. Serum and/or urinary concentrations of estrogen are usually ele- vated (1-4) and the tumors have been shown to syn- thesize and secrete estrogens in vitro (1-3, 5-7). We present a patient with a feminizing adrenocor- tical carcinoma with clearly demonstrated aro- matase and sulfotransferase activity in vitro, and high production rates of androstenedione (A), dehydroepiandrosterone (DHEA), and dehydro- epiandrosterone sulfate (DHEA-S). The elevated circulating estrogen most likely arose both from di- rect tumor production as well as peripheral con- version of tumor-derived androgens.
CASE REPORT
A 45-year-old male noted the onset of non-tender bilateral breast enlargement shortly after he began lifting weights. He shaved daily and reported no
change in libido or sexual dysfunction but had lost 15 pounds on a diet. There was no history of jaun- dice, liver disease or exposure to hepatotoxins, es- trogens or androgens and he was on no medica- tions. He denied abdominal pain or discomfort. The medical history was significant for hypertension for five years followed without medication. On physi- cal examination the blood pressure was 170/100 without orthostatic changes. He did not appear Cushingoid and his voice and hair distribution were normal. There was no thyromegaly. Tanner Stage Ill gynecomastia was present bilaterally. Abdominal examination revealed a large nondiscrete mass in the left periumbilical area and left lower quadrant. There was no hepatomegaly or lymphadenopathy. External genitalia were normal and the testes were of normal size and consistency.
An abdominal CT scan revealed a large left-sided retroperitoneal solid mass displacing the left kidney and encroaching on the spine and the major ves- sels. The chest x-ray was normal, as were a chest CT scan, inferior venogram, and renal angiogram. Laboratory studies are shown in Table 1 and showed a marked increase in all androgens and estrogens. In addition, the gonadotropins were suppressed, and there was evidence for low-renin hyperaldos-
Key-words: Feminization, adrenocortical carcinoma, adrogen, estrogen. Correspondence: John L. Stock, M.D., Medical Center of Central Massachusetts, 119 Belmont St., Worcester, MA 01605, USA.
Received June 8, 1993; accepted December 20, 1993.
| Preoperative | Postoperative | Normal Range | |
|---|---|---|---|
| Estrone | 268 | 21 | 10-56 pg/ml |
| Estradiol | 69 | 21 | 10-50 pg/ml |
| Estrone sulfate | 3,015 | 182 | 300-500 pg/ml |
| Androstenedione | 8.44 | 0.88 | 0.3-2.0 ng/ml |
| Testosterone | 5.71 | 8.45 | 3.0-11.0 ng/ml |
| Sex-hormone binding globulin | 85.4 | 31.9 | 11-71 mm/L |
| DHEA | 24.2 | 2.02 | 1-4.5 ng/ml |
| DHEA-S | 18.3 | 1.20 | 1-4.0 µg/ml |
| LH | 1.2 | 7.5 | 1.5-9.2 mlu/ml |
| FSH | 0.5 | 9.5 | 1-14 mlu/ml |
| Renin1 | 0.2 | 1.0 | 0.2-2.3 ng/dl |
| Aldosterone 1 | 28.3 | 11.6 | <16 ng/dl |
| Deoxycorticosterone | 75 | 6 | 3.5-11.9 ng/dl |
1Drawn in the morning, sitting position, on an ad lib sodium diet
teronism. The sex hormone-binding globulin was al- so elevated. Additional tests included serum potas- sium 4.2 meq/L, 08:00 h cortisol 21.6 µg/dl, ACTH 21 pg/ml (normal, 9-52), urinary free cortisol 190 ug/24 h (normal, 5-47), 17-ketosteroids 235 mg/24 h (normal, 5-15). The urinary free cortisol was mea- sured by reverse phase high pressure liquid chro- matography after extraction with dichloromethane and a wash with sodium hydroxide, a process that excludes most cortisol-like C-21 steroids. The serum concentrations of TSH and prolactin were normal and hCG was undetectable in the serum.
He underwent a left adrenalectomy and nephrecto- my (Fig. 1). Pathological examination revealed a 10 kg adrenocortical carcinoma with extensive necro- sis, broad fibrous bands, vascular invasion, diffuse growth pattern, pleomorphism and a mitotic rate greater than one per high power field. The tumor abutted directly on the renal capsule but there was no invasion. As shown in Table 1, the elevated levels of steroids and SHBG returned to the normal range postoperatively. The gonadotropoins and renin in- creased into the normal range. There was a poor cortisol response to ACTH stimulation and perioper-
ative steroids were continued and gradually tapered. Continued hypertension was treated adequately with oral nifedipine. A whole body CT scan four months postoperatively was negative for metastases.
IN VITRO STUDIES
In order to determine whether there was aromatase or sulfotransferase activity, tumor slices were incubat- ed in 5.0 ml KR bicarbonate containing 1 mg/ml glu- cose and 2 uCi [7-3H]androstenedione and 10,000 cpm of [4-14C]-estrone. Incubations were carried out for 2 h at 37C in 95% O2/5% CO2 in a Dubnoff incu- bator (8). As a control, 5 ml KR bicarbonate, 1 mg/ml glucose and the radiolabeled steroids were incubat- ed in each of 5 flasks without tissue. After 2 h the me- dia were removed and 200 dpm of [4-14C]-A and [4- 14C] testosterone (T) and 200 µg estrone (E1) and estradiol (E2) were added to all flasks and 300 µg es- trone sulfate (E1SO4) was added to flasks 1, 2 and 5. All samples were analyzed for radiolabeled A, T, E1 and E2 following organic extraction, phenolic parti- tion and chromatographic and derivatization steps (8). The [7-3H] and [4-14C] estrone sulfate were mea- sured as follows: the spent media, after extraction of the unconjugated steroids from flasks 1, 2 and 5, was solvolyzed and the resultant estrone processed as above (9). Following final purification all samples were measured for radioactivity and the dpm corrected for recoveries.
As shown in Table 2, the tumor converted signifi- cant amounts of A to T, E1, E2, and E1S04 as well as E1 to E2 and E1SO4, indicating the presence of aromatase and sulfotransferase activity.
DISCUSSION
Fewer than 100 cases of feminizing adrenocortical carcinoma in men have been reported (1, 2, 4, 10).
| % conversion of added androstenedione | |||||
|---|---|---|---|---|---|
| A | T | E1 | E2 | E1SO4 | |
| Tumor | 10.8 | 0.95 | 1.93 | 0.07 | 1.07 |
| Control | 91.4 | 0.05 | 0.01 | 0.001 | 0.05 |
[p] A, E1= 14.6%
| % conversion of estrone | |||
|---|---|---|---|
| E1 | E2 | E1SO4 | |
| Tumor | 30.8 | 1.25 | 18.0 |
| Control | 100% | 0.02 | 1.6 |
These tumors are usually large and produce multi- ple steroids. Our patient is unusual given the ex- tremely large tumor size, the lack of clinical metas- tases, and the diversity of hormone production (es- trogens, androgens, mineralocorticoids and glu- cocorticoids). Although there was biochemical ev- idence for excessive glucocorticoid production (el- evated urinary free cortisol), and clinical adrenal in- sufficiency postoperatively (abnormal ACTH stim- ulation), the lack of Cushingoid features was prob- ably related to the rapid growth of the tumor, marked androgen production and possibly short exposure to elevated glucocorticoid production al- though the age of the tumor was unknown. Similarly, although elevated serum concentrations of mineralocorticoids were documented, the lack of clinical change in blood pressure is consistent with a fixed component and/or a coexistent diag- nosis of essential hypertension.
Most patients with feminizing adrenal tumors have elevated serum and/or urinary levels of E1 and E2 (1-4). In vitro studies have demonstrated the tu- mor’s ability to synthesize and secrete estrogens (1, 3, 5-7) and the in vivo conversion of androstene- dione and testosterone to estrogen has been shown (2). In a young female patient with an estro- gen-producing adrenal tumor, the increased es- trogen levels appeared to arise directly from the tu- mor (7). However, the specific source of estrogen (either directly from tumor or from peripheral con- version) has not been documented in other cases. Incubation of tumor slices from our patient demon- strated that the tumor had active aromatase and sulfotransferase enzyme systems since the tumor tissue was able to convert A to E1, E2, and E1SO4; and to convert E1 to E2 and E1SO4. Unfortunately, we were not able to obtain normal adrenal tissue from the patient as a comparison. However, others have shown that adrenals secrete estrogens and sulfates (11, 12) and hence possess enzymes nec- essary for aromatization and sulfotransferase.
If we assume the metabolic clearance rate (MCR) of A is 2200 L/day, of E1 is 2200 L/day (13-15), then one can estimate the blood production rate of A (PB-A) and E1 (PB-E1) using the formula PB=MCR x concentration (15). Thus, PB-A=18.6 mg/day (nor- mal=0.7-4.4 (15)) and PB-E1=590 µg/day (nor- mal=22-110 (15)). These values are thus clearly in- creased over normal. If we assume that the percent of A entering the blood that is aromatized to E1 {p)A,E1=1.6% (13-15), then the amount of E1 that would arise from peripheral conversion can be es- timated as PB-A *{p)A,E1 (15), and thus we calculate that about 297 µg/day of E1 (50%) would arise from the peripheral conversion of A. Since the tumor has
aromatase activity some of this peripheral conver- sion might occur in the tumor. However, it is prob- able that the major part of the peripheral conver- sion would take place in the usual sites, including fat and muscle (14, 16). These calculations are based on assumptions using data obtained from subjects without such elevated circulating levels of androgens and estrogens. However, we have nev- er found a significant association between the MCR of androstenedione and its circulating level, or be- tween estradiol levels and aromatization (15).
In addition both DHEA and DHEA-S are aromatized in peripheral tissue but at very low rates (17, 18). However, because the estimated production rates would be large, it is likely that some of the E1 aris- es from these steroids by peripheral, including tu- mor, conversion. Thus, the presence of an active aromatase in the tumor and the estimated high pro- duction rates of androstenedione, DHEA and DHEA-S suggest that the source of estrogens in our patient was both directly from the tumor and from peripheral conversion of tumor derived androgens. We have reported a case of a feminizing adreno- cortical carcinoma which is unusual, given the large size of the tumor and lack of clinical metastases, the diversity of hormone production and the prob- able dual source of estrogen directly from the tu- mor as well as peripheral conversion of tumor-de- rived androgens.
ACKNOWLEDGMENTS
The authors wish to thank Charlene Franz, Charles Flood, and James A. Coderre for their technical assistance and Carol Ann Frazier for her expert secretarial support.
REFERENCES
1. Gabrilove J.L., Sharma D.C., Wotiz H.H., Dorfman R.I. Feminizing adrenal cortical tumors in the male: a re- view of 52 cases including a case report. Medicine 44: 37, 1965.
2. Gabrilove J.L., Nicolis G.L., Hausknecht R.V., Wotiz H.H.
Feminizing adrenocortical carcinoma in a man. Cancer 25: 153, 1970.
3. Golder M., Millington D., Cowley T., Griffiths K., Roberts H., London D.R., Butt W.R. Hormonal control of steroid synthesis by a feminizing adrenal tumor. J. Endocrinol. 67: 17p, 1975.
4. Veldhuis J., Sowers J., Rogol A., Klein F.A., Miller N., Dufau M.L.
Pathophysiology of male hypogonadism associated with endogenous hyperestrogenism.
N. Engl. J. Med. 312: 1371, 1985.
5. Fang V.S. Establishment and characterization of a strain of hu- man adrenal tumor cells that secrete estrogen. Proc. Natl. Acad. Sci. USA 74: 1067, 1977.
6. Mathur R.S., Williamson H.O., Moody L.O., Diczfalusy E.
In-vitro sterol and steroid biogenesis by a feminizing adrenocortical carcinoma. Acta Endocrinol. (Copenh.) 73: 518, 1973.
7. McKenna T.J., O’Connell Y., Cunningham S., McCabe M., Culliton M. Steroidogenesis in an estrogen-producing adrenal tumor in a young woman: comparison with steroid profiles associated with cortisol- and androgen-pro- ducing tumors.
J. Clin. Endocrinol. Metab. 70: 28, 1990.
8. Orczyk G.P., Mordes J.R., Longcope C. Aromatase activity in a rat Leydig cell tumor. Endocrinology 120: 1482, 1987.
9. Longcope C. The metabolism of estrone sulfate in normal males. J. Clin. Endocrinol. Metab. 34: 113, 1972.
10. Freeman D. Steroid hormone-producing tumors in man. Endocr. Rev. 7: 204, 1986.
11. Baird D.T., Uno A., Melby J.C. Adrenal secretion of androgens and oestrogens. J. Endocrinol. 45: 135, 1969.
12. Longcope C. Adrenal and gonadal androgen secretion in normal females. Clin. Endocrinol. Metabol. 15: 213, 1986.
13. Longcope C., Kato T., Horton R. Conversion of blood androgens to estrogens in nor- mal adult men and women. J. Clin. Invest. 48: 2191, 1969.
14. Longcope C. Methods and results of aromatization studies in-vivo. Cancer Res. 42: 3307S, 1982.
15. Baird D., Horton R., Longcope C., Tait J.F. Steroid prehormones. Pros. Biol. Med. 11: 383, 1968.
16. Longcope C. Peripheal aromatization: studies on controlling fac- tors. Steroids 50: 253, 1987.
17. Longcope C., Bourget C., Flood C. The production and aromatization of dehy- droepiandrosterone in postmenopausal women. Maturitas 4: 325, 1982.
18. Haning R. Jr., Carleson I., Flood C., Hackett R., Longcope C. Metabolism of dehydroepiandrosterone sulfate (DS) in normal women and women with high DS concen- trations.
J. Clin. Endocrinol. Metab. 73: 1210, 1991.