Adrenocortical Carcinoma with Feminization and Hypertension Associated with a Defect in 116-Hydroxylation
C. D. WEST, L. F. KUMAGAI, E. L. SIMONS, O. V. DOMINGUEZ,1 AND D. L. BERLINER1
Departments of Medicine, Biochemistry, and Anatomy, University of Utah College of Medicine; and Veterans Administration Hospital, Salt Lake City, Utah
ABSTRACT. An investigation of steroid metabolism in a 28-year-old male with meta- static adrenocortical carcinoma, who ex- hibited feminization, hypertension, edema and acne, has yielded information on the role of steroids in the production of these signs and symptoms. Initially, when feminization was the only finding, the patient excreted ab- normally large amounts of estrogens and nor- mal amounts of 17-ketosteroids and 17- hydroxycorticosteroids. Later, hypertension, edema and acne appeared in association with abnormally high and rapidly rising levels of estrogens and tetrahydro-11-deoxycortisol, but normal levels of tetrahydrocortisol, tetra- hydrocortisone, androsterone and etiocho- lanolone. ACTH administration caused a dis- proportionately marked increase in the excre- tion of estrogens and tetrahydro-11-deoxy- cortisol. Estrogen excretion also increased up- on the administration of methopyrapone (2- methyl-1,2-bis[3-pyridyl]-1-propanone), an 118-hydroxylase inhibitor.
These studies of steroid excretion suggested: 1) that the tumor possessed an overactive aromatizing enzyme system and a defect in 118-hydroxylation to account for the over- production of estrogens with feminization, and 2) that overproduction of the usual androgens and corticoids was not involved in
the initiation of the hypertensive phase.
In support of the first conclusion it was demonstrated directly by incubating neoplas- tic tissue with radioactive steroid precursors that the tumor possessed most of the biosyn- thetic enzymes known to occur in normal adrenocortical tissue, but that there was a relative deficiency in 118-hydroxylation and an overactive aromatization. The incubation results suggested that an overproduction of deoxycorticosterone might have been re- sponsible for the hypertensive phase of the disease, but this was not proved in vivo.
On the basis of both the in vivo and in vitro results, the most probable pathways utilized for steroid production in this patient were designated.
The excretion of 17-ketosteroids and 17- hydroxycorticosteroids reached abnormally high levels terminally without producing any significant changes in the clinical manifesta- tions. Treatment with o,p’-DDD(2,2-bis[4- chlorophenyl, 2-chlorophenyl]-1,1-dichloro- ethane) was partially effective in lowering steroid excretion, but had no demonstrable effect or tumor growth. Amphenone (3,3- bis[p-aminophenyl] butanone-2) was ineffec- tive in both regards. (J Clin Endocr 24: 567, 1964)
A N UNUSUAL combination of signs and symptoms in a 28-year-old male patient with metastatic adrenocortical carcinoma, namely, feminization, hyper- tension, edema and acne, led to this in-
Received January 15, 1964; accepted March 3,1964.
Supported in part by USPHS Grant No. C- 3588 and American Cancer Society Grant No. P-173 and P-174.
1 Key Investigators (American Cancer So- ciety).
vestigation of the role of steroids in the production of these clinical findings. The association of feminization with hyper- tension in adrenocortical carcinoma has been reported previously, but the etio- logical basis remains unknown (1, 2).
In the present study the excretion of urinary steroids was determined at fre- quent intervals during the course of the disease, and attempts were made to alter excretion by a variety of maneuvers de- signed to yield information on mecha-
nisms involved in steroid production. In addition, tumor tissue was incubated in vitro with radioactive steroid precursors, and the metabolic products were isolated and identified. In this manner the major steroids produced by this patient were characterized and the pathways utilized for their production were elucidated. A correlation of clinical course with steroid production provided some insight into the role of steroids in the production of this clinical syndrome.
Materials and Methods
1. Analytical procedures for urinary steroids. Urine was analyzed for estrogens, 17-keto- steroids and 17-hydroxycorticosteroids, both in groups and as individual compounds by the following techniques:
Urine collections. Twenty-four-hour urines were collected under refrigeration without preservatives. The completeness of the col- lection was judged by the constancy of the 24-hr urinary creatinine levels.
Group steroid analyses. Total urinary 17- ketosteroids were measured by a modifica- tion of the method of Callow et al. (3). The normal range for urinary 17-ketosteroids in males in our laboratory is 10-20 mg/24 hr. The method described by Peterson et al. was used to determine the level of excretion of urinary 17-hydroxycorticosteroids (4). The normal male range for urinary 17-hydroxy- corticosteroids in our laboratory is 6-12 mg/24 hr.
Urinary estrogens. The excretion of estrone, 170-estradiol and estriol in the urine was measured by the method of Brown (5). The upper limits for normal males in our labora- tory are 5 ug/24 hr for estrone, 178-estradiol and estriol. These 3 urinary estrogens were also identified by countercurrent distribu- tion and paper chromatography by tech- niques reported previously (6).
Procedure for the analyses of urinary pools. In some of the experiments, 24-hr urine collections were pooled and analyzed for individual C-19 and C-21 steroids. For controls, 3-day urine pools were obtained from 2 normal males comparable in age to the patient. The normal urine pools were analyzed in the same way as the patient’s urine samples. The urine pools were hydro-
lyzed first with 8-glucuronidase2 and then with acid as described by Kappas and Gal- lagher (7) and extracted 4 times with 2 volumes of ethyl ether. The combined ether extracts were washed successively with 1/10 vol of 0.1N NaOH, 0.1N HCI and distilled H2OX3. The washed ether extract was evaporated to dryness under air and labeled “neutral ether extract.”
Preparation of a- and B-ketonic fractions. The neutral ether extract was reacted with Girard’s reagent T to separate ketones from nonketones. The ketones were separated into a and ß fractions by means of the digitonin procedure described by Butt et al. (8).
Paper chromatographic analysis for neutral C-19 a-ketonic steroids. Aliquots of the neutral «-ketonic fractions were analyzed for androsterone and etiocholanolone by chromatography on Whatman No. 2 paper in heptane-propylene glycol for 18 hr as de- scribed by Zaffaroni (9). Areas in the un- known chromatogram corresponding to the reference standards were eluted, and the quantity of etiocholanolone and androster- one was determined by the Zimmermann reaction.
Paper chromatographic analysis for neutral C-21 a-ketonic steroids. Aliquots of the neu- tral a-ketonic fractions were chromato- graphed on Whatman No. 1 paper in chloro- form-formamide. The following fractions were obtained by this chromatography: 1) tetrahydrocortisol3 +allotetrahydrocorti-
2 Ketodase.
3 Trivial names for steroids: THF, tetrahydro- cortisol=3a,118,17,21-tetrahydroxy-58-preg- nan-20-one, allotetrahydrocortisol =3a,118, 17, 21-tetrahydroxy-5@-pregnan-20-one; THS, tet- rahydro-11-deoxycortisol =3a,17,21-trihydroxy- 56-pregnan-20-one, tetrahydrocortisol acetate =3a,21-diacetoxy-118,17-dihydroxy-58-preg- nan-20-one, allotetrahydrocortisol acetate =38, 21-diacetoxy-118,17-dihydroxy-5a-pregnan-20- one, tetrahydro-11-deoxycortisol acetate =3a, 21-diacetoxy-17-hydroxy-58- pregnan- 20-one; THE, tetrahydrocortisone =3a,17,21-trihy- droxy-58-pregnane-11,20-dione, tetrahydrocor- tisone acetate =3a,21-diacetoxy-17-hydroxy- 58-pregnane-11,20-dione, cortisone acetate =21- acetoxy-17-hydroxy-44-pregnene-3,11,20-trione, cortisol acetate =21-acetoxy-118,17-dihydroxy- 44-pregnene-3,20-dione, 11-deoxycortisol =17,21- dihydroxy-44-pregnene-3,20-dione, 17-hydroxy- progesterone = 17-hydroxy-44-pregnene-3,20- dione, equilenin =3-hydroxy-41,3,5(10),6,8-estra- pentaene-17-one, aromatic progesterone =3-hy-
sol, 2) tetrahydrocortisone +cortisol, and 3) tetrahydro-11-deoxycortisol +cortisone. Each of these fractions was eluted separately and acetylated and chromatographed on paper in benzene-formamide as described by Berliner and Salhanick (10). This procedure separated tetrahydrocortisone acetate from cortisol acetate and tetrahydro-11-deoxy- cortisol acetate from cortisone acetate, but could not separate tetrahydrocortisol acetate from allotetrahydrocortisol acetate. There- fore, the reported tetrahydrocortisol values include allotetrahydrocortisol. Where ap- propriate, the acetates were oxidized with chromic acid and rechromatographed in benzene-formamide. The Porter-Silber reac- tion was used for quantitation.
Urinary tetrahydro-11-deoxycortisol was also identified by its infrared absorption spectrum.
2. Incubation methods. Tumor nodules were removed from the lungs and liver within 1 hr after death and stored in ice cold buffer until incubation less than 2 hr later. Radioactive steroid substrates were incubated with tumor tissue by 2 different methods. In one, minces were incubated in phosphate buffer only by the method of Berliner et al. (11). In the other, tumor homogenates were in- cubated with added cofactors and energy sources as reported by Dominguez (12). The exact conditions of each incubation are given under Results.
In both instances, isolation, identification and quantitation of the incubation products were carried out by paper chromatography of the free steroids and their derivatives (11, 12) and by recrystallization to constant specific activity when specified.
The extracts from the incubations with phosphate buffer only were divided into neu- tral and phenolic fractions as described by Baggett et al. (13). An attempt was made to identify the phenolic metabolites by paper
droxy-19-nor-41,3,5(10)-pregnatrien-20-one, tetra- hydro-11-deoxycorticosterone =3a,21-dihy- droxy-58-pregnan-20-one, androstenedione = 44- androstene-3,17-dione, pregnenolone =36-hy- droxy-45-pregnen-20-one, 17-hydroxypregneno- lone =38,17-dihydroxy-45-pregnen-20-one, dehy- droepiandrosterone = 38- hydroxy - 45 - andros- ten 17-one, allopregnandione =5a-pregnane-3,17- dione, pregnandione =58-pregnane-3,17-dione, 68-hydroxy DOC =68,21-dihydroxy-44-preg- nene-3,20-dione.
chromatography of the free compounds and their acetate derivatives. The extracts from the incubations with added cofactors were not fractionated into neutral and phenolic compounds, but the presence of known estro- gens in the whole extract was examined by paper chromatography.
Results
The course of this patient with femin- izing adrenocortical carcinoma has been reported previously from the onset of dis- ease until the development of metastases (14). The present study was carried out in the period from the onset of metasta- ses until death. A brief summary of the premetastatic period is included for com- pleteness.
On his first hospital admission this 28- year-old male patient presented with loss of libido, gynecomastia, testicular atrophy and sparse facial and body hair. An adrenocortical carcinoma was re- moved surgically.
Eighteen months after surgery the gynecomastia had regressed completely. The testes had doubled in size, the sperm count had risen to normal, and libido had returned to normal. There was no evi- dence of metastatic disease.
Lung metastases were first detected by x-ray examination 22 months after sur- gery. At that time the patient was asymptomatic, and no abnormalities were found on physical examination. In Fig. 1 an attempt has been made to cor- relate the clinical course from the onset of metastases until death with the excre- tion of estrogens, 17-ketosteroids and 17- hydroxycorticosteroids. The time se- quence of various forms of therapy and tests is also indicated. The values for urinary steroids in Fig. 1 were all ob- tained during control periods while the patient was off therapy, except for the last three months. The excretion of es- trone, 173-estradiol and estriol was measured for the first time one month
AMPHENONE
TESTOSTERONE
1DOD
IS! ACTHI
2nd ACTHE
CORTISOL
Su 4885
PRED.
50
mgms 24 hrs.
HIGH b. p., ACNE,EDEMA
30
LUNG METS.
PLEURISY
GYNECO. HEPATO.
10
17 OHCS
17 KSod
ESTR.
20
30
40
50
MOS. POST - OP.
after lung metastases developed. The level of these three estrogens in the urine was two to six times higher than the upper limit for normal males in this laboratory. The excretion of 17-hy- droxycorticosteroids in the urine, meas- ured as Porter-Silber chromogens, was normal.
Twenty-seven months after surgery the patient developed pleuritic chest pain and raised some blood-tinged sputum. The chest roentgenogram and the excre- tion of estrogens and 17-hydroxycortico- steroids were essentially the same as on the previous examination.
During the thirty-third postoperative month the patient developed gynecomas- tia and a slightly enlarged liver. A sperm count was normal. His wife was delivered of a normal child during this month. The levels of urinary 17-ketosteroids and 17- hydroxycorticosteroids were within nor- mal limits, and the excretion of urinary estrogens had increased slightly.
Effect of cortisol and ACTH on estrogen excretion. The effect of cortisol adminis- tration on estrogen excretion was investi- gated during the thirty-fourth post- operative month to determine whether the inhibition of endogenous ACTH would influence estrogen excretion. Cor- tisol in doses increasing from 40 to 240 mg daily was administered for one month and had no effect upon estrogen excre- tion.
This finding suggested that endoge- nous ACTH did not play a significant role in estrogen production by the neoplastic adrenal tissue. This conclusion was strengthened by the observation that 40 units of ACTH gel im daily for seven days also had no effect upon estrogen ex- cretion.
Effect of amphenone on estrogen excre- tion. During the thirty-sixth postopera- tive month 45 g of amphenone were ad- ministered orally over a period of ten days. with a maximum daily dose of 6.0 g. The level of estrogen excretion was un- affected by the administration of amphe- none.
Effect of testosterone therapy on estrogen excretion. During the thirty-eighth post- operative month an experiment was undertaken to investigate the effect of
Mgm/24*
10아
TESTOSTERONE PROP. IM
0
ug/24°
500
0
Mg/24*
15001-
1000
500
ESTRONE
I
5
DAYS
10
| Steroid | Normal male controls | Patient | |||
|---|---|---|---|---|---|
| A | B | Pre- treatment control | Metho- pyrapone therapy | Post- treatment control | |
| EI | <. 005 | <. 005 | 0.85 | 3.2 | 1.5 |
| E2 | <. 005 | <. 005 | .09 | 0.16 | 0.17 |
| E3 | <. 005 | <. 005 | 2.0 | 1.9 | 3.9 |
| Etio | 1.7 | 1.1 | 3.6 | 9.9 | 6.8 |
| Andro | 1.9 | 1.8 | 0.88 | 3.0 | 1.7 |
| THE | 1.9 | 2.0 | 0.57 | 0.09 | 0.25 |
| THF | 1.4 | 1.0 | 0.28 | 0 | 7.3 |
| THS | 0.4 | 0.2 | 3.8 | 24.1 | 17.7 |
* All values are expressed in mg/24 hr. Urine was collected in pools for 3 days during the control periods and 4 days during the treatment period. Methopyrapone was administered orally in doses of 750 mg every 4 hr for 3 days, and 1200 mg of methopyrapone ditartrate was given iv on the fourth treatment day. EI =estrone; E2 =178-estradiol; E3 =estriol; Etio =etiocholanolone; Andro = androsterone; THE = tetrahydrocortisone; THF =tetrahydrocortisol; THS =tetrahydro-11- deoxycortisol.
testosterone administration on estrogen excretion. Testosterone propionate was administered im in doses of 100 mg daily for seven days and the daily excretion of estrone and 173-estradiol was measured before, during and after therapy. As shown in Fig. 2, the excretion of estrone increased to three to four times the con- trol level on testosterone, and 173-estra- diol excretion rose by a factor of 9. Upon discontinuing testosterone, estrogen ex- cretion promptly returned to control levels. These findings suggested that testosterone or its metabolites could be utilized by the tumor for the in vivo syn- thesis of estrogens, although the possi- bility that some other tissue or organ was responsible for this conversion can- not be ruled out.
From the thirty-fifth to thirty-eighth postoperative month the patient’s gen- eral condition deteriorated gradually but steadily, with increasing chest pain and hemoptysis. Chest roentgenograms re- vealed a slow progression in the lung metastases. Libido disappeared. The only physical findings were a slowly pro- gressive enlargement of the liver and a gradual loss in body weight.
During the fortieth postoperative
month the urinary 17-hydroxycortico- steroid excretion became definitely ele- vated for the first time, with no change in the patient’s clinical condition. It was decided to analyze the urine for individ- ual steroids of the C-21 series in an at- tempt to identify the urinary Porter- Silber chromogens. It was found that the levels of tetrahydrocortisone and tetra- hydrocortisol in the urine were de- creased and that elevated tetrahydro-11- deoxycortisol levels accounted for most of the Porter-Silber chromogenic ma- terial in the urine.
Effect of inhibition of 110-hydroxylation with methopyrapone on urinary steroid ex- cretion. The demonstration of increased levels of tetrahydro-11-deoxycortisol in the urine suggested that the neoplastic adrenal tissue had a defect in 118-hydrox- ylation, which could play a role in estro- gen production. If so, increasing the de- gree of the block with methopyrapone should increase the excretion of estro- gens. To explore this possibility, the effect of methopyrapone on the excretion of urinary steroids was investigated. Complete 24-hour urines were collected for three days and pooled for the pre- treatment control period. Methopyra-
pone was administered orally in doses of 750 mg every four hours for three days. On the fourth day of therapy 1200 mg of methopyrapone ditartrate was given iv over a 12-hour period. Twenty-four-hour urines were collected and pooled for the treatment period. A three-day urine col- lection was made after therapy for the “immediate post-treatment control.” Two weeks later a single 24-hour urine was collected for estrogen analysis.
The results of this experiment are shown in Table 1. The pretreatment con- trol levels of estrone, 178-estradiol and estriol were greatly increased, and the levels of tetrahydrocortisone and tetra- hydrocortisol were lower than those of the normal controls. The total excretion of etiocholanolone and androsterone was not grossly abnormal, but the ratio of etiocholanolone to androsterone was high (4:1). During methopyrapone therapy there was a marked increase in the excre- tion of tetrahydro-11-deoxycortisol, coupled with a striking decrease in tetra- hydrocortisone excretion and a disap- pearance of tetrahydrocortisol from the urine. Concurrently, there was an in- creased excretion of estrone and 173- estradiol, but estriol excretion remained at the control level. The levels of etio- cholanolone and androsterone in the urine also rose during the methopyra- pone period. All the elevated urinary steroids except 173-estradiol returned toward control levels during the im- mediate post-treatment control period. None had returned completely to their control levels within three days after therapy, presumably because of a pro- longed effect of the methopyrapone. In addition, estriol and tetrahydrocortisol reached their highest levels in the urine during the immediate post-treatment period for unexplained reasons. Two weeks later the excretion of all three es-
trogens had reached their pretreatment control levels.
It was concluded on the basis of the in- crease in tetrahydro-11-deoxycortisol and the decrease in tetrahydrocortisone and tetrahydrocortisol that an effective block in 116-hydroxylation had been achieved with methopyrapone and that this block had resulted in an increased production of estrogens. The increased excretion of C-19 steroids suggested that they or their precursors might have been utilized for estrogen synthesis.
Onset of hypertension, edema and acne. The clinical complexion of the case changed rather dramatically during the forty-fourth to forty-fifth postoperative months, with the patient developing hypertension, edema and acne. The hy- pertension persisted until death, with systolic pressures varying from 160 to 200 mm Hg and diastolic pressures vary- ing from 100 to 120 mm Hg. Repeated examinations failed to reveal a nonendo- crine basis for the hypertension. Edema was an intermittent problem throughout the remainder of the illness but re- sponded well to salt restriction and diuretics. The demonstration of normal plasma protein levels ruled out hypopro- teinemia as the cause of the edema. Acne with furunculosis was a recurrent problem until death.
The above findings suggested the pos- sibility that the clinical picture had shifted toward Cushing’s syndrome. However, several features usually found in Cushing’s syndrome were absent: (a) There was no disturbance in carbohy- drate metabolism. Periodic routine uri- nalyses were normal, and fasting blood sugar levels were normal on numerous occasions. (b) There was no evidence of osteoporosis by x-ray examination; uri- nary and serum calcium and phospho- rous levels were always normal. (c)
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName} ] on 20 November 2015. at 10:24 For personal use only. No other uses without permission. . All rights reserved.
| Steroid | Normal male controls | Patient | ||||
|---|---|---|---|---|---|---|
| A | B | Pre- treatment control | im ACTH | im +iv ACTH | Post- treatment control | |
| EI | <. 005 | <. 005 | 1.1 | 1.6 | 3.1 | 0.87 |
| E. | <. 005 | <. 005 | 0.15 | 0.22 | 0.36 | 0.15 |
| E3 | <. 005 | <. 005 | 3.5 | 4.2 | 5.6 | 4.0 |
| Etio | 1.7 | 1.1 | 4.7 | 8.4 | 16.3 | 4.9 |
| Andro | 1.9 | 1.8 | 1.3 | 1.7 | 4.0 | 1.7 |
| THE | 1.9 | 2.0 | 1.6 | 1.5 | 1.7 | |
| THF | 1.4 | 1.0 | 2.0 | 4.4 | 9.8 | 1.9 |
| THS | 0.4 | 0.2 | 19.6 | 30.7 | 29.7 | 18.9 |
* All values are expressed in mg/24 hr. Urine was collected for 3 days and pooled during im ACTH therapy, and for 1 day during im +iv ACTH therapy. Eighty units of crystalline ACTH and 40 units of ACTH gel were administered im alternately every 6 hr for 4 days. On the fourth day 100 units of crystalline ACTH was given iv in addition to the im ACTH. E 1 = estrone; E2 = 178- estradiol; E3 =estriol; Etio =etiocholanolone; Andro =androsterone; THE =tetrahydrocortisone; THF =tetrahydrocortisol; THS =tetrahydro-11-deoxycortisol.
Numerous hemograms were normal, in- cluding eosinophil counts.
In relation to the possibility of hy- peraldosteronism, serum sodium and potassium levels were measured fre- quently and were always found to be normal.
Effect of ACTH on urinary steroid ex- cretion. In the interval preceding the de- velopment of hypertension, edema and acne, the levels of urinary estrogens and 17-hydroxycorticosteroids had risen rap- idly, but the excretion of 17-keto- steroids had not changed significantly. With the change in clinical condition, it was decided to reinvestigate the pattern of urinary steroid excretion, both under control conditions and intense ACTH stimulation.
The experiment lasted ten days. From the fourth to the seventh experimental days the patient received 80 units of aqueous ACTH and 40 units of ACTH gel im every six hours alternately. On the seventh day 100 units of ACTH was ad- ministered iv over an eight-hour period in addition to the im ACTH. Three-day urine pools were analyzed during the pre- treatment control period, the period of
im ACTH, and the post-treatment con- trol period. Urine collected on the seventh day, when the patient received both im and iv ACTH, was analyzed separately as a 24-hour sample.
As shown in Table 2, the excretion of estrogens during the pretreatment con- trol period had risen considerably since the previous study, but the most impres- sive change was the rise in urinary tetra- hydro-11-deoxycortisol. The excretion of tetrahydrocortisone and tetrahydrocorti- sol had also risen in the interval, but not to abnormal levels. Tetrahydro-11-de- oxycortisol still accounted for most of the Porter-Silber chromogenic material in the urine.
Acne is often attributed to androgens, but in this patient the levels of andros- terone and etiocholanolone were not significantly elevated. The etiocholano- lone to androsterone ratio was high, as noted previously.
ACTH stimulated the excretion of all steroids measured except tetrahydrocor- tisone. Upon cessation of ACTH therapy all values returned toward control levels.
It was concluded from this study that the production of steroids by the adrenal
| Products (in % of extractable radioactivity) | Substrates | |||
|---|---|---|---|---|
| Prog-4-14C | 17a-OH-prog- 4-14℃ | cpd SR- 4-14C | Testo-4-14C | |
| prog | 0-1 | 0 | ||
| cpd SR | 13-17 | 51 | 68 | |
| cortisol | 0 | 0 | 12 | |
| cortisone | 0 | 0 | 10 | |
| DOC | 35-40 | 0 | 0 | |
| 4-andro | 0 | 0 | 0 | trace (<1%) |
| testo | 0 | 47 | ||
| phenols | 11 | 0 | 51 | |
| unidentified | 36-49 | 49 | 10 | 2 |
* One g of minced tumor tissue (from lung metastases) was incubated in 20 ml of phosphate buffer (pH 7.4, 0.1M) under air at 37 C for 3 hr. All were single incubations except for progesterone- 4-14C, which was incubated 3 separate times. Prog =progesterone; cpd SR =11-deoxycortisol; DOC =deoxycorticosterone; 4-andro = androstenedione; testo = testosterone; 17@-OH-prog=17a- hydroxyprogesterone.
tumor tissue was responsive to ACTH in large doses. No definite correlation could be made between the change in clinical findings and steroid excretion, although an abnormal excretion pattern was ob- served.
During the next six months (forty- sixth to fifty-second postoperative months) the patient’s condition deteri- orated rapidly. At 52 months after sur-
O,P’-DDD gm / 24*
12
0
3
EI
mg /24*
2
1
50
40
17 KS
30
mg /24*
20
10
0
30
17 OHCS
20
mg/24*
10
0
0
10
20
30
40
50
60
70
DAYS
80
gery he became bed-ridden and was ad- mitted to the hospital for the last time. The patient was started on prednisone, 5 mg P. O. bid, which was continued until death.
Effect of o,p’-DDD on urinary steroid excretion and tumor growth. During the last three months of his illness the pa- tient was treated with o,p’-DDD, as shown in Fig. 3. During the first course of o,p’-DDD, the excretion of 17-keto- steroids and estrogens, as represented by estrone, decreased promptly to values that were approximately 1/20 to 1/6 the control levels, respectively. The excre- tion of 17-hydroxycorticosteroids also decreased, but much less strikingly. After the o,p’-DDD was discontinued, the excretion of estrogens and 17-keto- steroids in the urine increased consider- ably but did not reach the pretreatment control levels. The level of urinary 17- hydroxycorticosteroids declined gradu- ally and progressively on stopping therapy, with no definite rebound in ex- cretion.
When the o,p’-DDD was restarted about one month later, the excretion of 17-ketosteroids again decreased promptly; but there was no definite
| Products (in % of extracted radioactivity) | Substrate | ||||||
|---|---|---|---|---|---|---|---|
| Prog-4-14C | Prog-4-14C +Arom Prog | Prog-21-14C | 17a-OH- prog-4-14C | 4-andro- 4-14C | Testo-4-14C | DOC- 21-14C | |
| prog | 0 | 11.3-44.9 | 0 | 0 | 0 | 0 | 0 |
| 17@-OH-prog | 1.6- 6.6+ | 0 | 0 | 0- 4.8 | 0 | 0 | 0 |
| 4-andro | 0.7- 1.3 | 0.6- 1.5 | 0 | 6.7-11.3+ | 50.9-81.1 | 10.9-58.1 | 0 |
| testo | 0.8- 1.6 | 2.2- 2.3 | 0 | 0 | 2.3-37.1 | 31.6-57.5 | 0 |
| cpd SR | 6.3-10.8+ | 2.8-12.9+ | 0 - 9.4+ | 63.8-78.0+ | 0 | 0 | 0 |
| DOC | 55.3-63.3 | 33.9-54.1 | 67.2-74.7 | 0 | 0 | 0 | 67.3-70.5 |
| P-dione | 0 | 0.1- 0.3 | 0 | 0 | 0 | 0 | 0 |
| allo-P-dione | 0.8- 2.7+ | 1.5- 4.3+ | 1.3- 2.1+ | 0 | 0 | 0 | 0 |
| 68-OH-andro | 0 | 0 | 0 | 0 | 2.1- 4.1+ | 1.7- 6.4+ | 0 |
| 68-OH-DOC | 0 | 0 | 0 | 0 | 0 | 0 | 8.9-10.5 |
| unidentified | 20.8-28.4 | 6.8-10.5 | 14.5-30.7 | 11.3-15.3 | 8.0-23.0 | 8.4-25.2 | 19.0-23.8 |
* Tumor tissue was homogenized in Krebs-bicarbonate buffer/bovine serum (1:1) (pH 7.4). Each flask contained: 1 ml of homoge- nate (equivalent to 200 mg of tumor tissue), 3 ml of SBN (bovine serum +Krebs-bicarbonate buffer [1:1] +4.88 g nicotinamide/l), 5 ml of SBN +Cof A (28.7 mg TPN, 108 mg DPN, 45.5 mg glucose 6-PO4, 40.6 mg sodium fumarate), 1 ml SBN +Cof B (40 mg glucose 6-PO4 dehydrogenase). The number of separate incubations and the amounts of substrates incubated were: 25, 50 and 100 mumoles progesterone-4-14C in 6 incubations; 25 mumoles only for progesterone-21-14C (2 incubations), deoxycorticosterone- 21-14C (2), 17a-hydroxyprogesterone (2), and testosterone (2); 16.7 and 334 mumoles of androstenedione-4-14C (4); 2000 mumoles of aromatic progesterone was incubated twice with 25 mumoles of progesterone-4-14C. Prog =progesterone; 17a-OH-prog =17a-hy- droxyprogesterone; 4-andro =androstenedione; testo =testosterone; cpd SR =11-deoxycortisol; DOC =deoxycorticosterone; P-dione =pregnanedione (pregnan-3,20-dione); allo-P-dione =allopregnanedione; 68-OH-andro =68-hydroxy-4-androstene-3,17-dione: 68-OH-DOC =68-hydroxydeoxycorticosterone; arom prog =aromatic progesterone. + Identified bv recrystallization to constant specific activity.
change in the excretion of either estrone or 17-hydroxycorticosteroids that could be attributed to the o,p’-DDD.
The patient tolerated 12 g of o,p’- DDD daily quite well, with no signs of toxicity except for some loss of appetite. No measurable tumor regression oc- curred on o,p’-DDD, and there was no symptomatic improvement. The patient died 55} months after surgery with far- advanced metastatic carcinoma.
Autopsy findings. At autopsy the lungs, liver, pleural spaces and abdomi- nal cavity were the principal sites of metastatic involvement. It was esti- mated that about 90% of the left lung and 35% of the right lung were replaced with tumor tissue. The liver weighed 6500 g and contained numerous large metastatic nodules.
The right adrenal gland was absent, and the left adrenal weighed 4 g. The cortex was severely atrophic. The testes, prostate and seminal vesicle were also atrophic. Microscopic examination of the
testes revealed no spermatogenesis. The tubules were lined with sertolic cells, and there was considerable peritubular fibro- sis. No definite Leydig cells were seen. There was a marked proliferation of the mammary ductal elements with epi- thelial hypertrophy, periductal fibrosis, and infiltration with mononuclear cells. The eosinophil cells in the anterior pitui- tary were markedly reduced in number and appeared to be replaced with chro- mophobe cells.
No other causes for the hypertension, besides the adrenal tumor, were found. Of special note, the kidneys, their blood supply and excretory ducts were entirely normal, with no invasion by tumor tis- sue.
Incubation results. The results ob- tained from incubating a number of 14C- labeled steroids with tumor tissue both with and without added cofactors are given in Tables 3 and 4. One of the most significant findings in both types of incu- bations was the observation that proges-
terone was converted mainly to deoxy- corticosterone and 11-deoxycortisol, and not to cortisol and cortisone as with normal adrenocortical tissue. 11-Deoxy- cortisol also accumulated when 17-hy- droxyprogesterone was incubated, and cortisol could not be demonstrated among the metabolites. The tumor tissue did not have the capacity to convert deoxycorticosterone to corticosterone, an activity possessed by normal adreno- cortical tissue. These observations sup- ported the postulate made on the basis of the steroid excretion studies that there was a defect in tumor 118-hydrox- ylase activity. That the block in 116- hydroxylation was not complete was demonstrated by the conversion of 11- deoxycortisol to cortisol and cortisone in small amounts.
Testosterone and, to a lesser degree, progesterone were readily converted to phenolic metabolities, suggesting the presence of an active 19-hydroxylase and an aromatizing enzyme system in the tumor tissue. Two radioactive com- pounds, accounting for most of the phe- nolic radioactivity, were isolated. One phenolic metabolite acted like estrone on paper chromatography, but the acetate derivative separated from authentic estrone acetate on paper chromatogra- phy. The possibility that this material might be equilenin was entertained but rejected when nearly all the radioac- tivity was lost to the mother liquor upon recrystallization with authentic equi- lenin. This phenolic metabolite was not identified.
The second phenolic metabolite ex- hibited a polarity intermediate between estrone and 173-estradiol on paper chromatography. The possibility that this unknown compound might be aro- matic progesterone was considered, but was ruled out by recrystallization with authentic aromatic progesterone. In
addition, when aromatic progesterone was used as a trapping agent in proges- terone-4-14C incubations, there was no radioactivity in the isolated aromatic progesterone, demonstrating that aro- matic progesterone was not in the steroid biosynthetic pathway.
In addition to the enzyme systems al- ready discussed, the results from the incubation experiments demonstrated that the tumor tissue possessed the fol- lowing enzyme systems: a 21-hydrox- ylase, a 17«-hydroxylase, a 66-hydrox- ylase, a desmolase, a 173-hydroxydehy- drogenase and a 44-hydrogenase.
Cofactor concentration appeared to make little difference insofar as the ma- jor enzymic activities in the tumor tissue were concerned. However, certain me- tabolites, such as the important inter- mediates, androstenedione and 17a-hy- droxyprogesterone, and the 66-hydrox- ylated compounds, could be demon- strated only with the addition of cofac- tors.
The site in the body from which the tumor tissue was removed, whether the lungs or liver, had no significant effect upon the steroid biosynthetic capacity of the tumor tissue.
Discussion
The clinical manifestations in this pa- tient appeared to be determined pri- marily by the balance of biologically active steroids produced by the adreno- cortical cancer. Feminization developed initially as a result of very high estrogen levels in contrast to low or normal levels of costicosteroids and androgens. Later, the patient developed persistent hyper- tension, edema and acne. These findings are reminiscent of Cushing’s syndrome; but the demonstration of normal levels of tetrahydrocortisone, tetrahydrocorti- sol, androsterone and etiocholanolone in the urine along with the lack of other
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName} ] on 20 November 2015. at 10:24 For personal use only. No other uses without permission. . All rights reserved.
features of Cushing’s syndrome-ruled against a common etiological factor.
From the results of this study it ap- pears highly probable that an excessive production of steroids was responsible for the induction of the hypertensive phase of this patient’s disease. No other etiological factors could be demonstrated either during life or upon post-mortem examination. Of the possible steroids which might have been involved, it ap- pears that the hypertension and edema were most likely caused by an overpro- duction of deoxycorticosterone. This ex- planation is supported by the evidence for a defect in 113-hydroxylation from both the in vivo and in vitro experiments and by the demonstration that deoxycor- ticosterone was the major metabolite formed when progesterone was incu- bated with tumor tissue. Unfortunately, the excretion of tetrahydro-11-deoxy- corticosterone in the urine, as an index to deoxycorticosterone formation, was not determined during the patient’s lifetime, and there is no proof that the patient was actually synthesizing excessive amounts of deoxycorticosterone at the time he was hypertensive.
There are observations in the medical literature which argue against the accept- ance of a defect in 113-hydroxylation as the only possible basis for the hyperten- sion in this patient. The association of an increased excretion of tetrahydro-11- deoxycortisol with adrenocortical car- cinoma, regardless of the clinical syn- drome, has been reported repeatedly. Al- though most of the reported patients were hypertensive, this has not always been the case. Furthermore, there does not seem to be any correlation in the de- gree of the block in 116-hydroxylation (judging by the levels of tetrahydro-11- deoxycortisol in the urine) and the hy- pertension.
The possibility that other steroids be-
ÇH3 Co
0
PROGESTERONE
CH20H co
CH3
Co
0’
-OH
DOC
0
ÇH2OH co
170 OH PROGESTERONE
HO
0
CH2 OH
01
OH
Co
CPD B
OH
0
0
0’
TESTOSTERONE
ANDROSTENEDIONE
1 CPD S
I
CH2 OH
HO
CO
-OH
UNKNOWN PHENOLIC INTERMEDIATES
0’
CORTISOL (+ CORTISONE )
I
0
+
CH2 OH
CH2OM
0
HO.
co
OH
co
-OH
HO
HO
ESTRONE [+ ESTRADIOL AND ESTRIOL)
HO
HO
ETIOCHOL ANOLONE (+ ANDROSTERONE )
THF (+ THE )
THS
sides deoxycorticosterone might have been involved in the induction of the hypertensive phase of the disease must be considered. Hyperestrogenism per se might have played a role, since it is well known that edema and hypertension may rarely develop in patients treated with estrogens. The production of 11- deoxycortisol was also very high at the critical time, but there is no good evi- dence that this steroid can cause hyper- tension. The possibility of hyperaldo- steronism remains, since aldosterone pro- duction was not measured in this study. The demonstration of normal serum sodium and potassium levels upon nu- merous occasions makes this possibility a little unlikely, but not impossible.
On the basis of the experimental re-
sults, a diagram of the metabolic path- ways that were most probably utilized for the production of estrogens and other steroids is given in Fig. 4. These path- ways are known to occur in normal adrenocortical tissue. The observed ab- normalities in steroid production in this patient can be explained quite satisfac- torily on the basis of normal pathways with an imbalance in activity of the se- quential biosynthetic enzymes, although the possibility that other unknown path- ways were also involved cannot be ex- cluded.
For an adrenocortical carcinoma to produce excessive amounts of estrogens and feminization, there appear to be two essential prerequisites: 1) an active aromatizing enzyme system and 2) an ample supply of precursor substrate. The rapid in vitro conversion of progesterone- 4-14C and testosterone-4-14C to phenolic metabolites and the in vivo conversion of testosterone to estrogens demonstrated that the tumor tissue in this patient pos- sessed a very active aromatizing enzyme system.
Unfortunately, the synthesis of known estrogens by tumor tissue could not be demonstrated. Unidentified phenolic compounds were isolated from the incu- bations with testosterone-4-14C and pro- gesterone-4-14C and partially character- ized. It was definitely proved that these unidentified phenolic metabolites were not estrone, 173-estradiol or estriol-all of which were identified in very large amounts in the urine. It is possible that these unidentified phenolic metabolites were intermediates in the synthesis of estrone and 170-estradiol. Estrogen bio- synthesis in the tumor might have been arrested at the stage of these intermedi- ates, which were then secreted and metabolized peripherally to estrone, 173- estradiol and estriol. It is equally possi- ble that the tumor normally possessed
the capacity to synthesize estrogens com- pletely, but that the final steps in estro- gen biosynthesis failed under the experi- mental conditions.
In reference to the supply of sub- strate for aromatization, androstenedi- one is probably a key intermediate. In normal adrenocortical tissue, andro- stenedione undergoes 19-hydroxylation followed by aromatization to estrogens. Any process that increases the level of androstenedione would be expected to increase estrogen synthesis in the pres- ence of an active aromatizing enzyme system. In normal biosynthesis andro- stenedione arises primarily from two separate pathways: 1) progesterone →17a-hydroxyprogesterone-androsten- edione, and 2) pregnenolone-17- hydroxypregnenolone → dehydroepian- drosterone-androstenedione. In our pa- tient only the first route was investi- gated. Obviously, the second route could have played a very important role in estrogen synthesis. Nevertheless, a mech- anism which involves the first route and which could produce high levels of androstenedione in the tumor tissue was found, namely, a relative insufficiency in tumor 116-hydroxylase activity. This enzymatic defect could cause an accumu- lation of intermediates, one of which is 17-hydroxyprogesterone, which could be readily converted to androstenedione and estrogens by the pathways shown in Fig. 4.
The in vivo demonstration that inten- sifying the partial block in 116-hydrox- ylation with methopyrapone increased the excretion of estrogens provided evidence that this enzymatic defect contributed toward the formation of excessive amounts of estrogens. However, this enzymatic defect could not be solely responsible. In the absence of an active aromatizing system, androstenedione, a potent androgen, would probably ac-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName} ] on 20 November 2015. at 10:24 For personal use only. No other uses without permission. . All rights reserved.
cumulate and be secreted. Presumably, this is the situation in patients with the virilizing adrenogenital syndrome and hypertension, who also have a relative insufficiency in 116-hydroxylation (15).
The role of ACTH in estrogen produc- tion in this case is uncertain. It was demonstrated that the tumor could re- spond to large unphysiological doses of ACTH, but the question whether endog- enous ACTH had any significant effect on tumor function was not completely resolved. The observation that low doses of ACTH and cortisol had no significant effect on estrogen excretion suggested that endogenous ACTH had little or no effect on tumor function under most con- ditions. On the other hand, it is difficult to explain the striking increase in estro- gen excretion on methopyrapone without involving endogenously produced ACTH. The possibility that an altera- tion in tumor responsiveness might have occurred in the interval between the ex- periments with low doses of ACTH and those with methopyrapone cannot be ruled out. Certainly, the total mass of tumor tissue available for ACTH stimu- lation increased tremendously in this in- terval. The responsiveness to large doses of ACTH was demonstrated shortly after the methopyrapone experiments; but, unfortunately, the effect of low doses of ACTH was not re-examined at this time.
Whether the findings in this patient are applicable to other patients with feminizing adrenocortical carcinoma is unknown at the present time. It appears highly probable that an active aromatiz- ing enzyme system is essential for the
overproduction of estrogens; but other mechanisms, besides a defect in 116-hy- droxylation, could increase the level of androstenedione and other precursors in the tumor and thereby affect estrogen synthesis. For example, a defect in 21- hydroxylase activity should be equally effective in this regard. A defect in 21- hydroxylation occurs in patients with the virilizing adrenogenital syndrome, and it seems reasonable to expect that this same enzymatic lesion coupled with an overactive aromatizing enzyme sys- tem might eventually be demonstrated in feminized patients.
References
1. Higgins, G. A., W. E. Brownlee, and F. A. Mantz, Jr., Amer Surg 22: 56, 1956.
2. Dohan, F. C., E. Rose, J. W. Eiman, E. M. Richardson, and H. Zintel, J Clin Endocr 13: 415, 1953.
3. Callow, M. H., R. K. Callow, and C. W. Emmens, Biochem J 32: 1312, 1938.
4. Peterson, R. E., J. B. Wyngaarden, S. L. Guerra, B. B. Brodie, and J. J. Bunim, J Clin Invest 34: 1779, 1955.
5. Brown, J. B., Biochem J 60: 185, 1955.
6. West, C. D., B. Danast, and O. H. Pearson, J Clin Invest 37: 341, 1958.
7. Kappas, A., and T. F. Gallagher, J Clin In- vest 34: 1566, 1955.
8. Butt, W. R., A. A. Henly, and C. J. O. R. Morris, Biochem J 42: 447, 1948.
9. Zaffaroni, A., R. B. Burton, and E. H. Keutmann, Science 111: 6, 1950.
10. Berliner, D. L., and H. A. Salhanick, Anal Chem 28: 1608, 1956.
11. Berliner, M. L., D. L. Berliner, and T. F. Dougherty, J Clin Endocr 18: 109, 1958.
12. Dominguez, O. V., J Clin Endocr 21: 663, 1961.
13. Baggett, B., L. L. Engel, K. Savard, and R. I. Dorfman, J Biol Chem 221: 931, 1956.
14. Wallach, S., H. Brown, E. Englert, and K. Eik-Nes, J Clin Endocr 17: 945, 1957.
15. Eberlein, W. R., and A. M. Bongiovanni, J Biol Chem 223: 85, 1956.