PERIPHERAL BLOOD STEROID LEVELS IN CUSHING SYNDROME DUE TO ADRENOCORTICAL CARCINOMA OR ADENOMA*
J. L. GABRILOVE, M.D. t
E. K. FREIBERG, D.O.
G. L. NICOLIS, M.D., PH.D.
From the Bronfman Department of Medicine, The Mount Sinai School of Medicine, City University of New York, The Mount Sinai Hospital and the Bronx Veterans Administration Hospital, New York, New York
ABSTRACT-Measurement of the peripheral vein levels of the steroids in the adrenal biosynthetic pathway affords a further possible definitive method of differentiating adrenocortical carcinoma from adrenal adenoma in the presence of Cushing syndrome. In patients with adrenal carcinoma the blood levels of 17-hydroxyprogesterone, testosterone, androstenedione, and dehydroepiandro- sterone are far higher than those seen in association with adenoma. In addition, the adrenal cortical carcinoma appears to have a higher ratio of precursors to product in the early part of the biosyn- thetic path and a lower one in the distal portion than adrenal adenomas. In Cushing syndrome due to adrenal tumor the measurement of serum cortisol is preferably made after chromatographic sep- aration from 11-deoxycortisol in view of the marked cross reactivity in the protein binding proce- dure.
This study is part of our continuing investiga- tion of the biosynthetic abnormalities in adrenocortical disease.1 It is concerned with the utilization of the peripheral vein concentrations of the adrenal steroids in the adrenal biosyn- thetic pathway to differentiate adrenal adenoma from carcinoma in the presence of Cushing syndrome.
The differentiation of an underlying adreno- cortical tumor, be it carcinoma or adenoma, from the adrenal nontumorous hyperfunction- ing syndrome is currently made by the level of serum ACTH and the demonstration of the
presence or absence of an adrenal tumor by means of computerized tomography (CT) scan- ning, sonography, scintillography or, in small tumors, by selective adrenal venography.2.3
The current most reliable preoperative crite- ria employed to differentiate a benign from a malignant adrenocortical tumor in Cushing syndrome are (1) the size of the tumor; (2) the ratio of conversion of 11-deoxycortisol into cor- tisol; (3) an increased urinary excretion of tetra- hydro-11-deoxycortisol (THS) in the presence of carcinoma; (4) the urinary titers of dehydro- epiandrosterone; and (5) the urinary excretion of neutral 17-ketosteroids. In regard to the for- mer, we have found that adrenocortical adeno- mas are usually no larger than 3 cm in one diameter and adrenocortical carcinomas are, usually, at least 5 cm in diameter in at least one dimension.4 The limitations in utilization of the
*Aided by Grants-in-aid from the NIH (HD 07685), Murray M. Rosenberg, the Jack Martin Fund, the Arthur R. Sohval Research Fund, Francis Levien, Meyer Willett, the Edmond de Rothschild Foundation, Milton Petrie and Marcelle Stein.
tBaumritter Professor of Medicine.
| Case | Age | Site and Size of Tumor | Metastases | Urinary Free Cortisol (µg/24 hr) | Plasma Cortisol (µg/dl) (AM) | Dexamethasone Suppression (2 mg Dose) | Circadian Rhythm: Plasma Cortisol (µg/dl) | 17-KS (mg/24 hr) | |
|---|---|---|---|---|---|---|---|---|---|
| CARCINOMA | |||||||||
| 1 | 56 | Right; 9 cm | Liver, lung | 176-310 | 45 | . . | . . | 20.5 | |
| 2 | 71 | Left flank | Liver, lung | 256 | 44 | NS | 42 AM | 30 | |
| mass | (overnight | 46 PM | |||||||
| -1 mg) | |||||||||
| 3 | 25 | Left; | None | 17a-OH 23.5 | 37.5 | NS | 38 AM | 33 | |
| 12 × 5 | mg; 912 | 28 PM | |||||||
| × 10 cm, | |||||||||
| 380 Gm | |||||||||
| 4 | 73 | Right; 17.5 | None | >225 | 32.5 | . . | . . | 57.7 | |
| cm | |||||||||
| ADENOMA | |||||||||
| 5 | 26 | Left; 3 cm | . . | >618 | 25.5 | . . | 25.5 AM | 13.3 | |
| 21.5 PM | |||||||||
| 6 | 68 | Left; 3 cm | . . | 311 | 23.5 | NS | . . | 17.2 | |
| 7 | 32 | Left; 3 cm | . . | 1,021 | 24 | NS | Absent | 19.3 | |
| 8 | 21 | Right; | .. | 297 | 21 | NS | . . | 14.2 | |
| 4 × 3 cm, | |||||||||
| 12 Gm | |||||||||
*NS = nonsuppressible.
conversion of 11-deoxycortisol into cortisol has been discussed by us previously.5 The aid af- forded by the finding of a high urinary content of tetrahydro S stems from the work of Lipsett, Hertz, and Ross6 among others. Previous studies have demonstrated that the urinary content of dehydroepiandrosterone6 and/or the neutral 17- ketosteroids have been of aid but not infallible in differentiating carcinoma from adenoma.7 On the other hand, urinary measurements of es- trogen have not been too helpful.6
In this study we present evidence that measurement in the serum of several of the ste- roids involved in the biosynthetic pathway to cortisol may, and in this study clearly do, dif- ferentiate the patients with adrenocortical car- cinoma from those with adrenocortical adenoma.
Material and Methods
Four patients with adrenocortical carcinoma and Cushing syndrome and 4 patients with Cushing syndrome due to adrenocortical adenoma have been studied. All subjects were women, had overt Cushing syndrome, and fulfilled all or most of the aforementioned crite- ria. Some pertinent details including the size of tumor measured by the pathologist and/or ra- diologist and, when applicable, metastases are
given in Table I. In some patients exhibiting a high urinary titer of free cortisol, an elevated plasma cortisol, and a demonstrable adrenal tu- mor by CT scanning, sonography or veno- graphy, the dexamethasone suppression test and/or evaluation of circadian rhythm of plasma cortisol were not necessary and, there- fore, not done. The peripheral blood levels of progesterone, 17a-hydroxyprogesterone, andro- stenedione, dehydroepiandrosterone, dehy- droepiandrosterone sulfate, testosterone, estra- diol, deoxycorticosterone, and aldosterone were measured by specific methods using radioim- munoassay and of 11-deoxycortisol, cortisol, and corticosterone by competitive protein binding as previously detailed.8-10 In the endocrine rou- tine laboratory the protein binding procedure without chromatographic separation was em- ployed for measuring cortisol8 (Table I), where- as in this study (Table II) cortisol, cortico- sterone, and 11-deoxycortisol were measured after chromatographic separation.9,10 Blood samples were drawn in the morning between 8 and 10 AM.
Results
The results obtained are given in Table II. It will be noted that there is a clear-cut differen- tiation in the two groups in regard to the titers
| Case | PG | 17-HPG | DES | DESS | ADS | TS | ES | 11-DC | C | DCS | CS | AS |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CARCINOMA | ||||||||||||
| 1 | 99 | 1343 | 3942 | . . | 7500 | 873 | 44.4 | 12.9 | 9.6 | 167 | 10.8 | |
| 2 | 159 | 577 | 1341 | . . | 1604 | 1057 | ||||||
| 3 | 41 | 568 | 682 | . . | 2142 | 441 | 2.7 | 2.7 | 5.0 | |||
| 4 | 142 | 474 | 519 | 745 | 3438 | 430 | 3.2 | 15.0 | 9.2 | 0.25 | 16 | |
| Mean | 110.25 | 740.5 | 1621 | . . | 3671 | 700.3 | 16.8 | 10.2 | 7.9 | 13.4 | ||
| SD | 45.6 | 350.2 | 1375 | 2309 | 272.7 | 19.5 | 5.4 | 2.1 | 2.6 | |||
| ADENOMA | ||||||||||||
| 5 | 107 | 185 | 75 | . . | 175 | 16 | 0.2 | 16 | ||||
| 6 | 38 | 21 | 21 | 7 | 89 | 9 | 0.7 | 0.1 | 9.4 | 18 | 0.16 | 8.7 |
| 7 | 32 | 42 | 127 | . . | 133 | 14 | 4.5 | .. | ||||
| 8 | 13 | 31 | 37 | 142 | 33 | 3.7 | 0.1 | 16.6 | ||||
| Mean | 47.5 | 69.8 | 65 | . . | 134.8 | 18 | 3.0 | 0.13 | 14 | |||
| SD | 35.6 | 67 | 40.8 | . . | 30.7 | 9.0 | 1.6 | 0.05 | 3.3 | |||
| NORMAL | <100 | 11-80 | 140-850 | 82-338 | 85-290 | <100 | 3-15 | <2 | 3-20 | 8-40 | <2 | 3-15 |
KEY: PG = progesterone (ng/dl); 17-HPG = 17-hydroxyprogesterone (ng/dl); DES = dehydroepiandrosterone (ng/dl); DESS = dehydroepidandrosterone sulfate (ug/dl); ADS = androstenedione (ng/dl); TS = testosterone (ng/dl); ES = estradiol (ng/dl); 11-DC = 11-deoxycortisol (ug/dl); C = cortisol (pg/dl); DCS = deoxycorticosterone (ng/dl); CS = corticosterone (ug/dl); AS = aldosterone (ng/ dl).
of 17a-hydroxyprogesterone, testosterone, an- drostenedione, and dehydroepiandrosterone. There is only one measurement in each group for dehydroepiandrosterone sulfate, but these two values would indicate that similar dif- ferences might be present between the two groups in regard to this steroid. It is noteworthy that all the values obtained for dehydroepian- drosterone (four) and dehydroepiandrosterone sulfate (one) in patients with adenoma were be- low the normal range. The levels of 11-deoxy- cortisol and the ratio of 11-deoxycortisol to cor- tisol also differentiate the two groups as demon- strated previously.8.9 Measurements of deoxy- corticosterone are also present only in 1 subject in each group but here, too, it would seem pos- sible that the titer may be of value in differen- tiating the two groups.
It is of note, in this small series, that there is overlap of the urinary excretion of the neutral 17-ketosteroids between adenoma and car- cinoma (Table I), the difference in titers of 20.5 (Case 1) and 19.3 (Case 7) being too small to be considered significant. The urinary excretion of free cortisol did not differentiate the two groups whereas the plasma cortisols (nonchromato- graphic method) (Table I) were higher in the patients with carcinoma. As in our previous re- port, at the time of diagnosis adenomas were usually no larger than 3 cm in diameter where- as carcinomas were much larger.
Comment
The data presented would suggest that pre- operative differentiation of malignant adrenal tumor causing Cushing syndrome from an adrenal adenoma may be afforded by measure- ment of the blood content of various steroids in the adrenal biosynthetic pathway, particularly dehydroepiandrosterone, as well as by the size of the tumor and the other measures previously mentioned. This is not to imply that the ability to make the differentiation would negate the need for surgical removal in any case. It is possi- ble that the high concentrations present in the patients with adrenocortical carcinoma may possibly reflect the large amount of functioning tissue, since adrenocortical carcinomas are usually much larger than the adenomas and may be accompanied by metastases or enzymic alterations in the malignant cell or a combina- tion of both. It is of note that the differences in concentrations do not extend to all the adrenal steroids. In addition measurement of these ste- roids may afford markers for following the course during treatment in patients with adrenocortical carcinoma. In patient 4, normal serum levels were observed six and twelve months after surgical removal of a well- encapsulated tumor. To date no evidence of metastatic disease has been manifested.
In the study by Kelly et al.11 of 4 patients with Cushing syndrome and adrenocortical
carcinoma, increased concentrations were en- countered for cortisol (4/4), 11-deoxycortisol (3/ 4), testosterone (2/4), and estradiol (3/4), but no studies were done on serum progesterone, 17a- hydroxyprogesterone, androstenedione, dehy- droepiandrosterone, or dehydroepiandro- sterone sulfate. In 1 subject increased blood levels were encountered for deoxycorticosterone and aldosterone. In another study,12 the urinary excretion of free deoxycorticosterone and aldo- sterone, particularly the former, were found frequently to be increased in all types of Cush- ing syndrome. No blood studies were employed. In neither study were patients with Cushing syndrome due to adrenocortical carcinoma and adenoma compared.
The patients with carcinoma exhibit a lower ratio of deoxycorticosterone to corticosterone and of testosterone to estradiol. On the other hand, patients with carcinomatous adrenals would appear to have a higher ratio of precur- sor to product than patients with adenoma in the early steps in the biosynthetic path, namely, progesterone to 17a-OH progesterone, 17a-OH progesterone to 11-deoxycortisol, 17a-OH pro- gesterone to androstenedione, and progesterone to 11-deoxycorticosterone.
It is the large amount of 11-deoxycortisol present in the peripheral serum (Table II) which, in part elevates the apparent peripheral serum “cortisol” determined by protein binding without chromatographic separation in pa- tients with adrenal carcinoma since there is marked cross reactivity between cortisol and 11-deoxycortisol in the protein binding proce- dure. In all of these subjects the clinical appear- ance was that of overt Cushing syndrome. However, it might seem advisable in Cushing syndrome associated with adrenal tumor to re- peat the measurement of serum cortisol after chromatographic separation from 11-deoxy- cortisol or to do cortisol production rates to as- certain the actual extent of excessive true corti- sol production to facilitate the differentiation of
benign and malignant adrenal tumors. None- theless, the loss of circadian rhythm and lack of suppressibility of serum and urinary free corti- sol with small dose dexamethasone administra- tion (2 mg/day) seen in Cushing syndrome with adrenal tumor as well as nontumorous adreno- cortical hyperfunction exists even if a large part of the “cortisol fraction” actually is deoxycorti- sol.
Mount Sinai School of Medicine New York, New York 10029 (DR. GABRILOVE)
References
1. Sharma DC, and Gabrilove JL: A study of the adrenocorti- cal disorders related to the biosynthesis and regulation of steroid hormones and their computer simulation, Mt Sinai J Med 42:(Suppl I) 39 (1975).
2. Nicolis GL, Mitty H, Modlinger R, and Gabrilove JL: Per- cutaneous adrenal venography: a clinical study of fifty patients, Ann Intern Med 76: 899 (1972).
3. Yeh Hsu-Chong, et al: Ultrasonography of adrenal masses- usual manifestations, Radiology 127: 467 (1978).
4. Soffer LJ, Iannaccone A, and Gabrilove JL: Cushing’s syn- drome, Am J Med 30: 129 (1961).
5. Nicolis GL, and Gabrilove JL: Studies on the efficiency of adrenocortical 113-hydroxylation in the human subject, J Clin Endocrinol Metab 29: 831 (1969).
6. Lipsett MG, Hertz R, and Ross GT: Clinical and pathophy- siologie aspects of adrenocortical carcinoma, Am J Med 35: 374 (1963).
7. Forbes AP, and Albright F: A comparison of the 17 ketoste- roid excretion in Cushing’s syndrome associated with adrenal tu- mor and with adrenal hyperplasia, J Clin Endocrinol 11: 926 (1951).
8. Murphy BEP: Some studies of the protein binding of steroids and their application to the routine micro and ultra micro measurement of various steroids and body fluids by competitive protein binding radioassay, J Clin Endocrinol Metab 27: 973 (1967).
9. Newsome HH Jr, Clements AS, and Borum EH: The si- multaneous assay of cortisol, corticosterone, 11-deoxycortisol and cortisone in human blood, ibid. 34: 473 (1972).
10. Gabrilove JL, Freiberg EK, and Nicolis GL: Testicular function in Klinefelter’s syndrome, J Urol 124: 825 (1980).
11. Kelly WF, et al: Cushing’s syndrome due to adrenocortical carcinoma-a comprehensive clinical and biochemical study of patients treated by surgery and chemotherapy, Acta Endocrinol 91: 303 (1979).
12. Cassar J, et al: Deoxycorticosterone and aldosterone excre- tion in Cushing’s syndrome, Metabolism 29: 115 (1980).