Virilizing Adrenal Tumor in a Child Suppressed with Dexamethasone for Three Years. Effect of o,p’-DDD on Serum and Urinary Androgens
SIGRUN KORTH-SCHUTZ, LENORE S. LEVINE, JOEL A. ROTH, PAUL SAENGER, AND MARIA I. NEW
Department of Pediatrics, Division of Pediatric Endocrinology, New York Hospital-Cornell Medical Center, 525 East 68th Street, New York, N. Y. 10021, and Department of Pathology, Overlook Hospital, Summit, N. J. 07901, College of Physicians and Surgeons, Columbia University, N.Y.
ABSTRACT. Extensive hormonal evaluation was performed in a girl with adrenal carcinoma during the primary tumor stage, following adrenalectomy, during the period when metastases were evident and while on treatment with o,p’-DDD.
At the age of 14 months a diagnosis of congenital adrenal hyperplasia was made and treatment with dexamethasone (0.125 to 0.25 mg/day) resulted in a fall-off in growth rate, normal advancement in bone age, decrease in virilization and suppression of 17- ketosteroid excretion which continued until 4 3/12 years of age when virilization increased. At five years of age elevated serum and urinary androgen levels unsuppressible with dexamethasone were
noted. Following removal of a large right adrenal carcinoma, serum and urinary hormone levels re- turned to normal. Three months following surgery, liver metastases were documented associated with elevated levels of serum androgens. With o,p’- DDD treatment, serum dehydroepiandrosterone sul- fate (DS) and urinary 17-ketosteroid (17-KS) excre- tion fell rapidly while there was a delay in the fall of free androgens. The persistence of free steroid secretion with decreased formation of DS suggests that the o,p’-DDD may have altered sulfatase ac- tivity before causing tumor necrosis and total de- crease in steroidogenesis. (J Clin Endocrinol Metab 44: 433, 1977)
V IRILIZATION in a prepubertal girl is most often due to congenital or early acquired adrenal hyperplasia (1). Adrenal or ovarian tumors are less often the cause of progressive enlargement of the clitoris, deepening of the voice, growth spurt, in- creased muscularity and adult body odor.
In this report, we present hormonal data on a girl who was thought to have congenital adrenal hyperplasia but was found to have a virilizing adrenal tumor at age five years.
Urinary steroid excretions, and serum androgen concentrations were measured during the preoperative phase, during post- operative remission, after appearance of
metastases, and during treatment with o,p’- DDD (2,2-bis [2-chlorophenyl-4-chloro- phenyl] 1,1-dichloroethane). In our patient abnormal concentrations of unconjugated androgens were the best indicator for the presence of tumor during treatment with this drug, whereas serum dehydroepiandros- terone sulfate (DS) levels decreased early during treatment together with urinary 17- ketosteroids (17-KS). O,p’-DDD induced necrosis of the metastases but was diffi- cult to use because of side effects.
Case Report (Fig. 1)
A five-year-old girl was referred to The New York Hospital because of progressive virilization. At birth she had normal female genitalia, but after the age of eight months increasing amounts of pubic hair were noted. At the age of 14 months, height was at the 86th percentile, clitormegaly was present, and bone age was 2 years, 8 months. Urinary 17-ketosteroid (17-KS) and 17-hydroxy- steroid (17-OHS) excretions measured on two occasions, were increased to 4.3-5.8 mg/d and 3.2-3.6 mg/d, respectively. Pregnanetriol excre-
Received July 8, 1975.
Supported in part by grant awards: USPHS, NIH, HD 00072; USPHS, NIH, Division of Research Facili- ties and Resources, Pediatric Clinical Research Center, RR 47; National Foundation-March of Dimes CRBS 278; USPHS, NIH Research Fellowship AM 00329; and by Deutsche Forschungsgemeinschaft.
Reprint address: Maria I. New, M. D., The New York Hospital-Cornell Medical Center, 525 East 68th Street, New York, N.Y. 10021.
ADRENALECTOMY
22
20
WEIGHT
18
kg
16
120
14
12
110
10
100
HEIGHT
(5.5)
8
(4.1)
6
4
90
cm
(2.7)
80
(Bone age in years)
KS, mg/d
100
80
70
60
40
60
20
0
50
0.5
Dexamethasone
HC, 30 mg / d
op’ DDD
mg/d
g/d
0.0
0
AGE 0
1
2
3
4
5 years
tion was also increased to 1.3 mg/d (normal 0.05-0.2 mg/d). An IVP was interpreted as normal. 17-KS excretion was suppressed with dexamethasone administration (0.75 mg daily for 7 days) and a diagnosis of adrenal hyperplasia was made. Treatment with 0.125 to 0.25 mg/day of dexamethasone was continued with a decrease in signs of virilization, fall-off in growth velocity to less than 2 cm/yr, slow advancement of bone age and continued suppression of 17-KS excre- tion.
At the age of 4 3/12 years, following an at- tempt to reduce dexamethasone to 0.125 mg every other day, urinary 17-KS excretion and later, 17- OHS excretion, began to rise despite an increase of dexamethasone and then a change to 30 mg/d of hydrocortisone (HC) (Fig. 1). This was as- sociated with deepening of the voice, increased pubic and body hair, increased muscularity and a growth spurt.
Admission to the Pediatric Clinical Research Center of The New York Hospital-Cornell Medi- cal Center revealed a bright, sturdy, muscular five-year-old girl. Her height was 100 cm (<3rd percentile) and her weight was 19.2 kg (50th percentile). Blood pressure was elevated (120/80 mm Hg). She had coarse hair on her arms, legs and back in addition to Tanner stage III pubic hair. The clitoris was 2.5 cm long and 1.5 cm in diameter. There was no labial fusion or rugation. No masses were palpable. Bone age was 4 years, 2 months.
An IVP and adrenal arteriogram demonstrated a large vascular supra-renal mass. There was no radiologic evidence of metastases to liver, lungs or bones. At laparotomy a large anaplastic adreno- cortical tumor was removed. The left adrenal gland appeared normal or small.
The patient made an uneventful recovery fol- lowing surgery and glucocorticoid treatment was discontinued within a week. No electro- lyte disturbances occurred. Urinary 17-KS excre- tion and plasma androgens decreased to normal prepubertal levels within 13 days postopera- tively and remained normal for one month (Fig. 2). During the following months catchup growth occurred and virilization did not progress, but three months postoperatively, bilateral breast discs were noted and hormone concentrations were abnormal again. Multiple liver metastases were demonstrated on arteriography.
Treatment with o,p’-DDD was begun, daily doses being increased from 0.5 g to 5.0 g on the tenth day. Therapy with prednisolone, 2.5 mg b.i.d., and 9 alpha-fluorocortisol, 0.05 mg/day, was begun concommitantly. Following hospital discharge, toxic side effects of o,p’-DDD (vomit- ing, anorexia, ataxia) increased despite a decrease in doses (Fig. 2). Medication had to be with- held repeatedly and symptoms improved mark- edly on these days.
An EEG taken during the seventh week of treatment suggested cerebral dysfunction with paroxysmal features which had not been present before o,p’-DDD treatment. During the last weeks of life, o,p’-DDD had to be given hidden in food because of the patient’s refusal to take the tablets.
At the age of five years and six months, the girl expired suddenly after two days of som- nolence and vomiting despite withdrawal of o,p’- DDD. Autopsy showed a few foci of necrotic and organizing adrenal cortical carcinoma in the region of the resected right adrenal gland
and attached to the under surface of both lobes of the liver. The left adrenal gland weighed 3.2 g and showed a diffuse mononuclear inflammatory infiltrate of lymphocytes, plasma cells and im- munoblasts. The adrenal cortical cells were sparse in number, the medulla was unremark- able. Additional autopsy findings included a mild non-specific hepatitis and lymphoid hyperplasia of mesenteric lymph nodes and of ileal mucosa. There was cerebral edema, but no focal lesions were noted.
Materials and Methods
Serum androgens were determined by radio- immunoassays using celite columns for separa-
tion of 44-androstenedione (44) from dehy- droepiandrosterone (DHEA) and testosterone (T); and of dihydrotestosterone (DHT) from es- trone (E1) and estradiol (E2) as previously de- scribed (2). For the radioimmunoassay of E, and E2 an antibody against estriol-3, 16alpha, 17beta-trihemisuccinate-HSA (S 310 #5, supplied by Dr. G. Abraham, Los Angeles) was used that bound both estrogens equally well. Cross-reac- tion with T that is partially eluted in the E, and E2 fractions was <0.001% at 50% displacement. Sensitivity of the standard curves was 5 pg. Blanks were below sensitivity and not subtracted. Recovery of 200 to 1000 pg E, and E2 added to charcoal-stripped serum was within 20% of
ADRENALECTOMY
ACTH!
ACTH
Dex (mg/d)
HC (mg/d)
&
0
15
ng/dl
E1
op’DDD (g/d)
8
0
E2
2000
Δ4
ng / dl
1000
0
500
T
ng / dl
DHT
0
2000
DS, ug /dl
1000
DHEA, ng/dl
DHEA
0
DS
100
mg/d
50
17 KS
0
25
17 OHS
F
mg/d
ug %
0
170
10
20
30
10
7
9
30
10
20 30
10 20
4
10
Aug
Sept
Oct Dec
Jan
Feb
Mar
the amount added. Intraassay coefficient of varia- tion (CV) was 16% for E1, 12% for E2, the respec- tive coefficients of variations between assays were 23% and 9.5%. Specificity as judged from the linearity of a dose-response curve of a serum pool was good (r = 0.999 for E1, and 0.977 for E2). Values in eight normal girls from three to eight years were <1.5 ng/dl for E1, and <1 ng/dl for E2. In six late-pubertal girls the normal concentrations ranged from 5.9 to 13.5 ng/dl for E1 and from 6.6 to 37.8 ng/dl for E2.
Serum DS was measured in diluted serum without solvolysis or extraction according to Buster and Abraham (3). Plasma cortisol was de- termined using fluorimetry (4). Urinary 17-KS and 17-OHS excretions were determined using standard methods, pregnanetriol excretion ac- cording to Bongiovanni (5). Urinary 17-KS were fractionated by paper chromatography (6) and were measured with the Zimmerman reagent. Tracer amounts of H3-androsterone, H3-etiocho- lanolone and C14-DHEA were added before ex- traction to correct for procedural losses.
Results
Hormonal studies
Plasma, serum and urinary hormones were determined prior to and after surgery in the baseline state; during administration of ACTH, 40 units iv over 6 h; and during
dexamethasone administration, 2 mg for 2 days and 8 mg for 2 days. Hormone deter- minations were also obtained postopera- tively and during administration of o,p’- DDD together with prednisolone, 2.5 mg b.i.d. and 9 alpha fluorocortisol, 0.05 mg/day.
Results of steroid measurements are given in Fig. 2, Table 1 and Table 2.
1. Urinary metabolites. During the first ad- mission and before resection of the adrenal tumor, urinary excretion of 17-KS was ele- vated (53-102 mg/d). Values fluctuated widely from day to day with no clearcut response to ACTH or dexamethasone. Within 13 days after removal of the tumor, 17-KS excretions returned to the normal prepubertal range (<2 mg/d). They re- mained normal one month later, but three months postoperatively they had increased again to values above 33 mg/d. At this time ACTH produced no significant change in the daily fluctuations of urinary 17-ketosteroid excretion and no suppression was observed during dexamethasone administration. O,p’- DDD induced a prompt decrease in 17-KS excretion reaching the normal range after five weeks of treatment.
| Stage | Age (yr/mo) | Date | Treatment | Urinary excretion | ||||
|---|---|---|---|---|---|---|---|---|
| 17-KS | Fractionation | |||||||
| DHEA | etio- cholano- lone (mg/24 h) | andros- terone | etio/ andros (ratio) | |||||
| Primary tumor | 4/11 | 8/19/74 | off hydrocortisone 1 week | 92 | 16 | 7.6 | 5.1 | 1.5 |
| 5/0 | 8/27/74 | ACTH | 89 | 32 | 8.2 | 14 | 0.6 | |
| Remission | 5/1 | 9/15/74 | none | 1.8 | 0.1 | .05 | 0.1 | 0.5 |
| Liver metastases | 5/4 | 12/27/74 | none | 25 | 4.4 | 4.8 | 8.3 | 0.6 |
| 5/5 | 1/8/75 | none | 59 | 11 | 6.6 | 12 | 0.6 | |
| 5/5 | 1/15/75 | o,p'-DDD 7 days | 34 | 5.3 | 7.7 | 4.7 | 1.6 | |
| 5/5 | 1/19/75 | o,p'-DDD 10 days | 22 | 2.1 | 4.5 | 2.2 | 2.0 | |
| 5/6 | 1/26/75 | o,p'-DDD 17 days | 16 | 3.4 | 2.2 | 1.0 | 2.1 | |
| Before ACTH | After ACTH | Dexamethasone for four days, (2 mg for 2 days, 8 mg for 2 days) | 6 days after surg. | 10 days after surg. | 30 days after surg. | 3 months after surg. | Normal prepubertal females (± SD) (2) | |
|---|---|---|---|---|---|---|---|---|
| DHEA (ng/dl) | 845 | 758 | 943 | <4 | 27.8 | 38.2 | 263 | 42.7 ± 38 |
| 44 (ng/dl) | 1715 | 1932 | 1922 | 21.5 | 50.2 | 52.9 | 479 | 18.2 ± 12 |
| T (ng/dl) | 289 | 362 | 400 | 6.6 | 4.8 | 11.8 | 55.9 | 5.9 ± 2.3 |
| DHT (ng/dl) | 76 | 64 | 122 | - | 2.6 | - | - | 3.5 ± 3.7 |
Fractionation of urinary 17-KS showed DHEA as the predominant steroid before removal of the adrenal tumor. Its excretion increased after ACTH stimulation. In remis- sion all urinary metabolites measured fell to very low levels. As liver metastases de- veloped, urinary excretions of etiocho- lanolone and androsterone were similar to those of DHEA. O,p’-DDD led to a prompt decrease of all fractions. During the first two and a half weeks of treatment, the ratio of etiocholanolone to androsterone in- creased from 0.6 to 2.1.
2. Serum dehydroepiandrosterone sulfate (DS). In the presence of the adrenal carci- noma, DS was markedly elevated, 1300- 1790 µg/dl as compared to a mean of 22 ug/dl in four normal girls of the same age. The concentrations fluctuated independ- ently of ACTH and dexamethasone ad- ministration. Decrease after removal of the tumor was slow but complete by the time of discharge 13 days post-operatively (1.2 µg/ dl). The liver metastases caused the same elevation of DS concentrations as the primary tumor (up to 1800 µg/dl) which were again not suppressible with dexamethasone. O,p’-DDD caused a prompt fall of DS levels that were in the normal range (7 ug/dl) after five weeks of treatment and were un- measurable (<1 µg/dl) thereafter.
3. Unconjugated serum androgens. Before removal of the adrenal tumor the unconju- gated serum androgens, T, DHT, 44, and DHEA were markedly elevated above those of age-matched controls. 44 and T increased slightly with ACTH administration, but an
increase over baseline was seen during dexamethasone administration as well. De- crease to the normal range had occurred by six days after removal of the tumor. A gradual increase was then noted during the first post- operative month (Table 2) and by three months post-operatively (12/9/74-Fig. 2) all free serum androgens were again in the abnormal range. They increased rapidly during the following weeks and four months post-operatively they were not suppressed with dexamethasone. The highest values were observed just prior to institution of o,p’-DDD treatment. They were in the same range as before operation. During the first month of treatment with o,p’-DDD, there was no change of T, DHT, and 44. DHEA concentrations decreased rapidly parallel to those of DS, but still remained markedly elevated until all free androgens decreased to unmeasurable levels during the fifth and sixth weeks of treatment.
4. Serum estrogens. At the onset E, was more elevated than E2 (12.3-13.9 ng/dl versus 1.5-3.5 ng/dl). Post-operatively con- centrations never returned to levels below 1.5 ng/dl which would be expected for the patient’s age. They rose again when the metastases developed. After four weeks of o,p’-DDD treatments no significant de- crease had occurred.
5. Plasma cortisol. Plasma cortisol was in the low-normal range without diurnal varia- tion, or any change after ACTH in the pres- ence of the primary tumor and in the meta- static state. Suppression with dexamethasone was not achieved in either phase. During
o,p’-DDD treatment, with concomitant ste- roid substitution, cortisol concentrations were consistently subnormal.
Discussion
The differential diagnosis of early ap- pearance of pubic hair in girls includes precocious adrenarche, congenital adrenal hyperplasia (CAH) and adrenal or ovarian tumors. Though oversecretion of androgens is present in all three disorders, in CAH there is usually labial fusion and genital ambiguity in the newborn female indicating the prenatal exposure to excessive andro- gens. In precocious adrenarche, androgen effects are restricted to the development of sexual hair and serum androgen concentra- tions are only slightly elevated (2). This is in contrast to the marked elevations usually found in adrenal tumor (7,8). Since tumors may have the same enzymatic deficiencies as exist in various forms of CAH (8), one cannot use the pattern of steroid concentra- tions or excretion for differential diagnosis. The critical test to rule out an adrenal tumor is the capacity of dexamethasone to suppress adrenal steroid secretion. Rare exceptions to this have been reported (9,10).
In our patient, suppression of androgen excretion by dexamethasone in early life suggested adrenal hyperplasia rather than tumor despite the normal female genitalia. The complete normalization of circulating adrenal hormones and of their urinary metabolites with clinically normal adrenal function after removal of the tumor, how- ever, made adrenal hyperplasia unlikely. The chronic cell inflammatory infiltration of the remaining adrenal gland which was found at autopsy, however, compounds the case. To our knowledge, it has never been described secondary to o,p’-DDD adminis- tration. Together with the paracortical hyperplasia of intestinal lymph nodes, the round cell infiltration suggests an intercur- rent autoimmune disease. No immunologi- cal studies were performed in this patient, and we do not know whether this disease
process had affected the gland during the period of remission and thus prevented the clinical and hormonal manifestations of a possibly preexisting hyperplasia.
As in most responders reported (11-13), o,p’-DDD caused a rapid decrease of all urinary 17-ketosteroids in our patient. This response could however not be attributed to an effect of the drug on the tumor, because the unconjugated serum androgens 44, T, DHT and the estrogens did not change at all during this time. The discrepancy may be explained when the effect of o,p’- DDD on peripheral steroid metabolism is considered (14-16). The drug stimulates ac- tivity of oxidases and thus induces forma- tion of highly polar metabolites that are not determined by routine measurements. The increase of the ratio of 50 to 5a-metabolites during o,p’-DDD treatment (Table 1) could represent another effect of this drug on androgen metabolism which has been ob- served previously (15). In addition, we found an early decrease of serum DS con- centrations during o,p’-DDD administration which might be due to a drug-induced in- hibition of tumor sulfatase. We were not able to measure secretion rates or metabolic clearance rates of DS to confirm this possi- bility. The rapid decrease of DS concentra- tions very likely caused part of the fall of 17-KS excretions and of serum DHEA concentrations.
Later in the course of treatment, all androgens decreased, suggesting lysis of the metastases, which was indeed found at autopsy. Thus measurement of unconjugated serum androgens was a critical measure of presence of functional metastases in this patient.
O,p’-DDD produced complete necrosis of metastases in our patient and thus proved once more to be an effective drug in meta- static adrenal carcinoma. The marked toxicity of o,p’-DDD, however, made treat- ment very difficult in this young child. Furthermore, a serious question must be raised about the cause of death. It is possible that the child died of o,p’-DDD toxicity
either via central nervous system intoxica- tion or secondary to adrenal insufficiency. The latter seems to be less likely, because this patient received replacement therapy with adequate doses of prednisolone and 9x- fluorocortisol. These were given regularly and readministered when vomiting oc- curred, but during the last days of life repeated vomiting might have prevented complete absorption of the medication. Cen- tral nervous system toxicity of o,p’-DDD was manifested by ataxia, somnolence and ab- normalities in the EEG. Two other cases with irreversible damage to the brain secondary to o,p’-DDD medication were mentioned by Lubitz et al. (13), but detailed information on these patients is not avail- able.
Because of the toxicity of o,p’-DDD, every physician using the drug must be very cog- nizant of toxic symptoms. Indeed the dose necessary to produce lysis of metastases in some patients may be much less than generally used (11-13), as this patient re- ceived less o,p’-DDD than recommended and still demonstrated necrosis of metas- tases.
Acknowledgment
The authors wish to acknowledge Mrs. Diane M. Chow for her outstanding technical assistance.
References
1. Migeon, C. J., Functional adrenocortical tumors, In Metabolic, Endocrine and Genetic Disorders in Children, vol. 1, Harper and Row, Hagerstown, 1974, p. 281.
2. Korth-Schutz, S., L. S. Levine, and M. I. New, Serum androgens in normal prepubertal children and in children with precocious adrenarche, J Clin Endocrinol Metab 42: 117, 1976.
3. Buster, J. E., and G. E. Abraham, Radioimmuno-
assay of plasma dehydroepiandrosterone sulfate, Anal Lett 5: 543, 1972.
4. Mattingly, D., A simple fluorimetric method for the estimation of free 11-hydroxycorticoids in human plasma, J Clin Pathol 15: 374, 1962.
5. Bongiovanni, A. M., The adrenogenital syndrome with deficiency of 30-hydroxy-steroid dehydro- genase, J Clin Invest 41: 2086, 1962.
6. Peterson, R. E., Determination of urinary neutral 17-ketosteroids, In Standard Methods in Clinical Chemistry, Academic Press, New York, vol. 4, 1963, p. 151.
7. Saez, J. M., M. A. Rivarola, and C. J. Migeon, Studies of androgens in patients with adreno- cortical tumors, J Clin Endocrinol Metab 27: 615, 1967.
8. Lipsett, M. B., and H. Wilson, Adrenocortical cancer: Steroid biosynthesis and metabolism evalu- ated by urinary metabolites, J Clin Endocrinol Metab 22: 906, 1962.
9. Dahl, J., and P. Bastrup-Madsen, Adrenogenital syndrome caused by adrenal cortex carcinoma in an infant, Ugeskr Laeg 125: 326, 1963.
10. Gallagher, T. F., A. Kappas, H. Spencer, and D. Laszlo, Influence of invasiveness, hormones and amphenone on steroids in adrenal carcinoma, Science 124: 487, 1956.
11. Hutter, A. M., and D. E. Kayhoe, Adrenal cortical carcinoma. Results of treatment with o,p’-DDD in 138 patients, Am J Med 41: 581, 1966.
12. Hoffman, D. L., and V. A. Mattox, Treatment of adrenocortical carcinoma with o,p’-DDD, Med Clin North Am 5: 999, 1972.
13. Lubitz, J. A., L. Freeman, and R. Okin, Mito- tane use in inoperable adrenal cortical carcinoma, JAMA 223: 1109, 1973.
14. Bledsoe, T., D. P. Island, R. L. Ney, and G. W. Lid- dle, An effect of o,p’-DDD on the extraadrenal metabolism of cortisol in man, J Clin Endocrinol Metab 24: 1303, 1964.
15. Hellman, L., H. L. Bradlow, and B. Zumoff, Decreased conversion of androgens to normal 17- ketosteroid metabolites as a result of treatment with o,p’-DDD, J Clin Endocrinol Metab 36: 801, 1973.
16. Bradlow, H. L., B. Zumoff, D. K. Fukushima, and L. Hellman, Drug-induced alterations of steroid hormone metabolism in man, Ann NY Acad Sci 212: 148, 1973.