Hyperadrenocorticism associated with excessive sex hormone production by an adrenocortical tumor in two dogs
Harriet M. Syme, BSc, BVetMed, DACVIM; J. Catharine Scott-Moncrieff, VetMB, MS, DACVIM; Nancy G. Treadwell, DVM; Mary F. Thompson, BVSc, DACVIM; Paul W. Snyder, DVM, PhD, DACVP; M. Randall White, DVM, PhD, DACVP; Jack W. Oliver, DVM, PhD
> Adrenocortical tumors may produce glucocorti- coids, mineralocorticoids, or adrenal sex hor- mones.
> When adrenal sex hormones are produced by an adrenal tumor, clinical signs of hyperadrenocorti- cism may result, even though cortisol concentra- tions are less than the reference range.
A n 11-year-old spayed female Labrador Retriever weighing 32 kg (70.4 lb) was referred to Purdue University Veterinary Teaching Hospital (PUVTH) for evaluation of suspected hyperadrenocorticism (HAC). Clinical signs included polyphagia, polydipsia, and polyuria. Water consumption was variable but had been as high as 19 L (590 ml/kg of body weight)/d. Changes in coat color, weight gain, increased panting, and abdominal enlargement were also reported.
Abnormalities noticed during physical examina- tion were limited to a coarse coat and a pendulous abdomen. Rectal temperature was 101.8 F (38.8 C), heart rate was 86 beats/min, and the dog was panting.
Results of a CBC revealed mild lymphopenia (0.78 X 103 cells/ul; reference range, 1 to 5 X 103 cells/ul). Serum biochemical abnormalities included mild hypernatremia (149 mmol/L; reference range, 138 to 148 mmol/L), high total CO2 (28 mmol/L; reference range, 13 to 24 mmol/L), and moderately high alkaline phosphatase activity (1,001 U/L; reference range, 20 to 157 U/L). Thyroxine concen- tration was 2.0 µg/dl (reference range, 1.3 to 4.0 µg/dl). A urine sample collected by cystocentesis had a specific gravity of 1.004 but was otherwise unremarkable. The urine protein-to-creatinine ratio was within reference range, and results of bacterial culture of urine were nega- tive. Repeated noninvasive measurements of blood pres- sure, using a Doppler technique, were normal.
Abdominal radiography revealed a mineralized
From the Departments of Veterinary Clinical Sciences (Syme, Scott- Moncrieff, Treadwell, Thompson) and Veterinary Pathobiology (Snyder, White), School of Veterinary Medicine, Purdue University, West Lafayette IN 47907; and the Clinical Endocrinology Service, Department of Comparative Medicine, College of Veterinary Medicine, University of Tennessee, Knoxville, TN 37901 (Oliver). Dr. Syme’s present address is the Department of Veterinary Basic Sciences, Royal Veterinary College, London NW1 OTU, United Kingdom. Dr. Treadwell’s present address is Shreveport Veterinary Internal Medicine, 489 Albany Ave, Shreveport, LA 71105.
Address correspondence to Dr. Scott-Moncrieff.
mass cranial to the left kidney. Axial skeletal osteopenia and abundant intra- and extra-abdominal adipose tissue were also evident. A mild bronchoin- terstitial lung pattern was evident on thoracic radi- ography. No evidence of pulmonary metastases was seen. Ultrasound examination of the abdomen con- firmed that the mass originated from the region of the left adrenal gland. The diameter of the mass was 7 cm in its greatest dimension. The right adrenal gland was not identified. There was no sonographic evidence of metastasis within the abdomen. The liver was diffusely hyperechoic, compared with the falciform fat.
An ACTH stimulation test was performed, using 250 µg of synthetic ACTHª administered IM. Baseline cortisol concentration was within reference range (1.7 ug/dl; reference range, 1.0 to 6.0 µg/dl), but there was little change in cortisol concentration 1 hour after ACTH administration (2.2 µg/dl; reference range, 7 to 17 µg/dl). Results of a low dose dexamethasone sup- pression test (0.01 mg/kg [0.0045 mg/lb] of dexam- ethasone, IV) revealed suppression of cortisol at 8 hours but not at 4 hours after injection (basal corti- sol = 2.2 µg/dl, reference range, 1.0 to 6.0 µg/dl; 4 hour cortisol = 2.5 µg/dl; 8 hour cortisol = 1.7 µg/dl, refer- ence range < 2.0 µg/dl). The urine cortisol-to-creati- nine ratio was within reference range (23; reference range, 8 to 24). Endogenous ACTH concentration was low (1.7 pmol/L; reference range, 6.7 to 25.0 pmol/L).
Adrenalectomy was recommended but was declined by the owners because of the cost and the risk of periop- erative mortality. Aliquots of serum obtained before and after ACTH administration were submitted to the University of Tennessee for measurement of adrenal sex hormone concentrations. Validation of these assays and determination of sex-dependent reference ranges have been reported elsewhere.1 Increased concentrations of estradiol, progesterone, and 17-hydroxyprogesterone were identified (basal estradiol = 38.4 pg/ml, reference range, 28 to 53 pg/ml; poststimulation estradiol = 84.3 pg/ml, reference range, 26 to 51 pg/ml; basal proges- terone = 2.8 ng/ml, reference range, < 0.2 ng/ml; post- stimulation progesterone = 2.7 ng/ml, reference range, 0.4 to 1.1 ng/ml; basal 17-hydroxyprogesterone = 2.3 ng/ml; reference range, < 0.4 ng/ml; poststimulation 17- hydroxyprogesterone = 2.0 ng/ml, reference range, 0.5 to 1.5 ng/ml). Treatment was initiated with the adrenolytic drug mitotaneb (750 mg, q12 h, PO; 47 mg/kg [21.36 mg/lb] total daily dose). Prednisone (0.25 mg/kg [0.11 mg/lb] q 24 h, PO) was administered concomitantly.
Following 10 days of treatment, the dog was reex- amined. No improvement in clinical signs was evident to the owner, and physical examination findings were unchanged. An ACTH stimulation test was performed and revealed no change in the cortisol response to ACTH (basal cortisol = 1.8 µg/dl; poststimulation cor- tisol = 2.1 µg/dl). Mitotane was continued at the same dose for another 10 days, after which there was still no clinical improvement. Radiography of the abdomen revealed an increase in the size of the left adrenal mass. There was no evidence of metastatic disease or inva- sion of the vena cava. An ACTH stimulation test was again performed (basal cortisol = 2.0 µg/dl; poststimu- lation cortisol =1.5 µg/dl), and prestimulation adrenal sex hormone concentrations were measured. Concentrations of progesterone, testosterone, dihy- droepiadrostenedione sulfate (DHEAS) and 17- hydroxyprogesterone were all considerably higher than reference ranges and had increased from pretreatment values (basal progesterone = 17.9 ng/ml, reference range, < 0.2 ng/ml; basal testosterone = 3.0 ng/ml, ref- erence range, 0.1 to 0.4 ng/ml; basal DHEAS = 53.2 ng/ml, reference range, 0.3 to 12 ng/ml; basal 17- hydroxyprogesterone = 6.8 ng/ml, reference range, < 0.4 ng/ml). The owners declined any further treatment. The dog died unexpectedly 2 months later. At necrop- sy, a bilobulated mass measuring approximately 10 X 10 X 5.75 cm effaced the left adrenal gland. A thick and well-organized fibrous connective tissue capsule cov- ered the mass, and there were multiple fibrous adhe- sions involving the pancreas and the serosal surfaces of the small and large intestines. The parenchyma of the mass was friable and pale white. The right adrenal gland measured approximately 1.2 × 1.0 cm in its greatest dimensions. The pituitary gland was grossly normal. Microscopic examination of the left adrenal gland mass revealed adrenocortical carcinoma. Small nests of adrenocortical neoplastic cells were present within a delicate fibrovascular stroma. Neoplastic cells were polyhedral with abundant amounts of pale eosinophilic cytoplasm with large oval vesicular nuclei. Cellular atypia, anisocytosis, and anisokaryosis were commonly observed. Much of the mass was necrotic and contained large amounts of fibrin and extravasated erythrocytes. The right adrenal gland was histological- ly normal. Tissue section of an embolus from the valve of the caudal vena cava near the mass contained low numbers of adrenocortical neoplastic cells, similar to those observed in the adrenocortical carcinoma. The embolus also contained mineralized fibrin and cellular debris adherent to the neoplastic cells.
A 9-year-old castrated male miniature Poodle weighing 5.9 kg (12.98 lb) was referred to PUVTH for evaluation of suspected HAC. Clinical signs reported by the owner included polyuria, polydipsia, and polyphagia. Results of an ACTH stimulation test per- formed by the referring veterinarian revealed a subnor- mal cortisol response to ACTH (basal cortisol concen- tration = 3.0 µg/dl, reference range, 1.0 to 6.0 µg/dl; 1 hour poststimulation cortisol concentration = 3.6 ug/dl, reference range, 7.0 to 17.0 µg/dl).
On physical examination, the dog was obese, heart rate was 92 beats/min with an irregular rhythm, respi-
ratory rate was 32 breaths/min, and rectal temperature was 38.3 C (100.9 F). An ECG revealed an irregular rhythm, a wandering atrial pacemaker, and occasional second degree AV block.
Hematologic abnormalities included polycythemia (Hct, 68.2%; reference range, 37 to 55%), mature neu- trophilia (19.62 X 103 cells/ul; reference range, 3 to 12 X 103 cells/ul), lymphopenia (0.84 X 103 cells/ul; refer- ence range, 1.0 to 5.0 X 103 cells/ul), and eosinopenia (0; reference range, 0.1 to 1.25 X 103 cells/ul). Serum biochemical analysis was unremarkable with the exception of mild increases in alanine transaminase (99 U/L; reference range, 3 to 69 U/L) and alkaline phosphatase (238 U/L; reference range, 20 to 157 U/L) activities and cholesterol (713 mg/dl; reference range, 125 to 301 mg/dl) concentration. Urinalysis revealed a specific gravity of 1.018 and benign sediment.
Mild hepatomegaly was evident radiographically. Ultrasound examination of the abdomen revealed an enlarged and irregularly marginated right adrenal gland measuring 1.26 cm across its short axis. No inva- sion of the vena cava was evident. The dimensions of the left adrenal gland were normal. The liver was hyperechoic, compared with the spleen, and had a mottled echogenicity. Thoracic radiography was unre- markable. Cardiac ultrasound revealed mild left ven- tricular hypertrophy. Noninvasive systolic blood pres- sure measurements, using a Doppler technique, were made repeatedly and ranged from 140 to 180 mm Hg.
Endogenous ACTH concentration was low (1.4 pmol/L; reference range, 6.7 to 25.0 pmol/L). An ACTH stimulation test was repeated with measure- ment of adrenal sex hormones, as described. Concentrations of androstenedione, estradiol, proges- terone, and 17-hydroxyprogesterone were considerably increased before and after administration of ACTH (basal androstenedione = 24.7 ng/ml, reference range, 2.7 to 8.0 ng/ml; poststimulation androstenedione = 83.7 ng/ml, reference range, 3.0 to 10.0 ng/ml; basal estradiol = 80.7 pg/ml, reference range, 28 to 63 pg/ml; poststimulation estradiol = 79.1 pg/ml, reference range, 30 to 69 pg/ml; basal progesterone = 2.1 ng/ml, refer- ence range < 0.1 ng/ml; poststimulation progesterone = 16.1 ng/ml, reference range, < 1.2 ng/ml; basal 17- hydroxyprogesterone = 0.9 ng/ml, reference range, < 0.1 ng/ml; poststimulation 17-hydroxyprogesterone = 16.4 ng/ml, reference range, 0.4 to 1.2 ng/ml). Basal testosterone concentration was within reference range but was increased above reference range after ACTH administration (0.97 ng/ml; reference range, 0.3 to 0.6 ng/ml).
One week later, the right adrenal mass was removed via a standard midline celiotomy. Gross extension of the mass into surrounding structures was not apparent. The left adrenal gland appeared small. Intraoperative complications were not encountered, and the dog recovered without complications. During the immediate postoperative period, hydrocortisone was administered as a continuous IV infusion (0.65 mg/kg/h [0.295 mg/lb/h]). A tapering dose of pred- nisone was substituted once the dog tolerated medica- tion orally (0.4 mg/kg [0.18 mg/lb], PO for 5 days, then 0.4 mg/kg [0.18 mg/lb], PO, q 24 h for 5 days,
then 0.4 mg/kg [0.18 mg/lb], PO, q 48 h for 30 days). Histologic examination of the excised right adrenal mass was consistent with adrenocortical carcinoma. The neoplastic cells were round to polygonal mononu- clear cells with abundant pale foamy basophilic cyto- plasm and mild to moderate nuclear polymorphism. The mass invaded through the thin fibrous capsule of the adrenal gland. Also evident within the mass were multiple foci of extramedullary hematopoiesis, pre- dominantly erythropoiesis.
The dog was reexamined 2 months after surgery. No prednisone had been administered for 21 days. The owners reported resolution of the dog’s polyuria and polydipsia. On physical examination, the dog was slight- ly obese and had mild seborrhea. Results of a CBC were unremarkable with resolution of the previously noted polycythemia; the only abnormalities evident in the serum biochemical analysis were mild increases in cho- lesterol (358 mg/dl; reference range, 125 to 301 mg/dl) and total CO2 (28 mmol/L; reference range, 13 to 24 mmol/L) concentrations. An ACTH stimulation test was performed. The adrenal sex hormone concentrations had decreased to within reference ranges, and the post- stimulation cortisol concentrations had increased such that there was now an appropriate response to ACTH administration. The other measured adrenal hormone concentrations were also within reference ranges. Thirteen months after surgery the dog was clinically normal. Thoracic radiography and abdominal ultra- sonography revealed no evidence of metastatic disease.
Hyperadrenocorticism is defined as increased pro- duction of steroid hormones by the adrenal cortex. The term is usually associated with abnormal glucocorticoid production, a syndrome that is well described in dogs.2 However, HAC may also cause excessive secretion of mineralocorticoids and sex hormones by the adrenal cortex. Hyperadrenocorticism caused by excessive pro- duction of mineralocorticoids results in hypokalemia and hypertension. Synthesis of the mineralocorticoid hormone deoxycorticosterone by an adrenal tumor has been reported in a dog.3 Findings in the 2 dogs report- ed here illustrate the clinical and pathologic findings that result from HAC associated with excessive sex-hor- mone production. This syndrome has been described in humans,45 cats,6,7 and ferrets.8
Both the dogs of this report had clinical signs sug- gestive of excessive glucocorticoid production, includ- ing polyuria, polydipsia, and polyphagia. The discov- ery of an adrenal mass in both dogs appeared initially to support adrenal dependent HAC causing excessive production of glucocorticoids. However, cortisol con- centrations were below reference range in both dogs following ACTH administration. Iatrogenic HAC attributable to exogenous corticosteroid administra- tion could have accounted for the clinical signs and the low measured cortisol concentrations but not the pres- ence of the adrenal masses. No source of exogenous glucocorticoid could be identified in either dog.
A nonfunctional adrenal tumor was considered unlikely, because the dogs had clinical signs that could not be explained by the presence of a nonfunctional mass, and because cortisol concentrations were low following ACTH administration. A unilateral nonsecre-
tory adrenal tumor would not be expected to result in defective cortisol production. The subnormal endoge- nous ACTH concentrations in both dogs also support- ed the presence of an endocrinologically active tumor. This led us to theorize that the tumors in these 2 dogs were producing hormones that inhibited cortisol pro- duction by normal adrenal tissue. This was confirmed by measurement of increased concentrations of adren- al sex hormones. A number of dogs with adrenal tumors that were unresponsive to ACTH administra- tion have been reported.9,10 In some of these dogs, increased 17-hydroxyprogesterone concentrations were documented.1º Other possible causes for lack of response to ACTH include lack of receptors for ACTH or aberrant biosynthetic pathways for cortisol synthe- sis. The miniature Poodle of this report clearly responded to ACTH stimulation with an increase in androstenedione, progesterone, and 17-hydroxyprog- esterone concentrations, suggesting that the failure of the cortisol concentrations to increase in the normal manner was not attributable to a lack of ACTH recep- tors. In the Labrador Retriever of this report, response of sex hormone concentrations to ACTH stimulation was equivocal. Removal of the adrenal tumor from the miniature Poodle resulted in resolution of all clinical signs and normalization of the adrenal sex hormone and cortisol concentrations, thus confirming that HAC was responsible for the dog’s clinical signs of disease. Unfortunately, resolution of these clinical signs in the Labrador Retriever was never achieved.
The clinical signs in these dogs were attributed, at least in part, to high circulating sex hormone concen- trations. It is not possible, however, to exclude the pos- sibility that hypercortisolemia contributed to the clini- cal signs. It is possible that the total daily production of cortisol was inappropriately high despite the normal resting cortisol concentrations and the lack of response to ACTH administration, as long as cortisol production did not fluctuate according to the dog’s physiologic requirements. However, during mitotane treatment of hyperadrenocorticism, a good clinical response is achieved if cortisol concentrations are reduced to the concentrations similar to those detected in these 2 dogs.2
In humans with sex hormone secreting adrenal tumors, low dexamethasone testing is reported to be normal.11 In the Labrador, normal suppression of dex- amethasone occurred at 8 hours; however, complete suppression did not occur at 4 hours. The importance of the lack of suppression at 4 hours is unclear. It is pos- sible that this was a spurious result attributable to the stress of hospitalization. In hindsight, it would have been useful to repeat this test and also to perform a low dose dexamethasone test in the miniature Poodle.
Progesterone and 17-hydroxyprogesterone con- centrations were increased in both dogs before and after administration of ACTH. Little is known about the effects of high concentrations of naturally occur- ring progesterone because of its short half-life in circu- lation. What information is available comes from the study of progestins (synthetic therapeutics that mimic progesterone). Some progestins have considerable intrinsic glucocorticoid activity.12 In addition to their
direct glucocorticoid effects, high progesterone con- centrations may result in clinical signs of disease by displacing cortisol from cortisol binding protein.13 This results in high concentrations of free cortisol even though the total serum cortisol concentration is decreased. Because it is the free or unbound cortisol that is active and able to exert its effects on a variety of tissues, clinical signs of cortisol excess may develop. Administration of progestins to dogs results in pro- found suppression of the hypothalamic-pituitary adrenal axis.12 Thus, elevation of progesterone concen- tration over a prolonged period would be expected to result in decreased endogenous ACTH concentrations and a decreased response to ACTH administration, as was evident in both the dogs reported here.
Progesterones (progesterone and 17-hydroxyprog- esterone) are synthesized in the adrenal cortex from cholesterol via the intermediaries pregnenolone and 17-hydroxypregnenolone. However, only small amounts of progesterone are normally secreted into the circulation, with most being converted to mineralocor- ticoids, glucocorticoids, or to androgens (DHEAS and androstenedione). It is these androgens that normally constitute the main sex hormone product of the adren- al glands. These androgens, in turn, are the substrates for testosterone and estrogen production by peripheral tissues. In humans with adrenocortical neoplasms, aberrant biosynthetic pathways have been well charac- terized.11 Adrenal gland neoplasms may be deficient in the enzymes involved in normal steroidogenic path- ways, such as 21ß-hydroxylase or 11ß-hydroxylase. When such enzyme deficiencies develop, they cause accumulation of precursor steroids proximal to the blockade. These precursors may either be released and cause clinical signs directly or they may be shunted into other metabolic pathways, such as those for androgen biosynthesis, and result in clinical signs indi- rectly. Such enzyme deficiencies are most often thought to arise during development of the neoplasm; hormone synthesis in these instances is normal prior to the tumor developing.14 However, in a few cases it is thought that an inherited partial enzyme deficiency is present from birth but that it is clinically asymptomatic until the development of an adrenal tumor in later life.” In both the dogs of this report, the tumors were char- acterized as adrenocortical carcinomas. Two cases of sex hormone producing adrenal tumors have been reported in cats; these tumors were also both described histologically as adrenocortical carcinomas.6,7 In humans, virilizing adrenal tumors are almost invari- ably carcinomas and have a high propensity to metas- tasize.4 Further work is required to characterize pro- duction of progesterone, 17-hydroxyprogesterone, and DHEAS in dogs with benign versus malignant adrenal tumors. Studies suggest that adrenal sex hormones are also increased in dogs with pituitary dependent hyper- adrenocorticism.15
The Labrador Retriever of this report was treated with mitotane because the owners declined surgery. Mitotane causes selective progressive necrosis of the adrenal cortex and has been found to be an acceptable alternative to surgery in many dogs with cortisol secreting adrenocortical tumors, with 66% of dogs
treated having a good to excellent response.16 The find- ing of extensive necrosis of the adrenal gland in this dog may represent a partial response to mitotane treat- ment, although necrosis occurs commonly in large adrenocortical carcinoma even without treatment. In general, dogs with adrenocortical neoplasia require a high dose of mitotane for remission of clinical signs.1 It is possible that this dog would have responded to a higher dose of mitotane than it received, but the own- ers were discouraged by the continued growth of the tumor and further increase in sex hormone concentra- tions and refused additional treatment.
ªCortrosyn, Organon Inc, West Orange, NJ.
Lysodren, Bristol Myers Squibb, Princeton, NJ.
References
1. Schmeitzel LP, Lothrop CD Jr. Hormonal abnormalities in Pomeranians with normal coat and in Pomeranians with growth hor- mone-responsive dermatosis. J Am Vet Med Assoc 1990;197: 1333-1341.
2. Feldman EC, Nelson RW. Hyperadrenocorticism (Cushing’s syndrome). In: Canine and feline endocrinology and reproduction. 2nd ed. Philadelphia: WB Saunders Co, 1996;187-255.
3. Reine NJ, Hohenhaus AE, Peterson ME, et al. Deoxycorticosterone-secreting adrenocortical carcinoma in a dog. J Vet Intern Med 1999;13:386-390
4. Noton JA. Adrenal tumors. In: deVita VT, Hellman S, Rosenberg SA, eds. Cancer principles and practice of oncology. 5th ed. New York: Raven Press, 1997;1659-1675.
5. Nogeire C, Fukishima DK, Hellman L, et al. Virilizing adrenal cortical carcinoma. Cancer 1977;40:307-313.
6. Boord M, Griffin G. Progesterone secreting adrenal mass in a cat with clinical signs of hyperadrenocorticism. J Am Vet Med Assoc 1999;214:666-669.
7. Rossmeisl JH Jr, Scott-Moncrieff JC, Siems J, et al. Hyperadrenocorticism and hyperprogesteronemia in a cat with an adrenocortical adenocarcinoma. J Am Anim Hosp Assoc 2000;36: 512-517.
8. Rosenthal KL, Peterson ME. Evaluation of plasma androgen and estrogen concentrations in ferrets with hyperadrenocorticism J Am Vet Med Assoc 1996;209:1097-1102.
9. Reusch CE, Feldman EC. Canine hyperadrenocorticism due to adrenocortical neoplasia. Pretreatment evaluation of 41 dogs. J Vet Intern Med 1991;5:3-10.
10. Norman EJ, Thompson H, Mooney CT. Dynamic adrenal function testing in eight dogs with hyperadrenocorticism associated with adrenocortical neoplasia. Vet Rec 1999;144:551-554.
11. Orth DN, Kovacs WJ, DeBold CR . The adrenal cortex. In: Wilson JD, Forster DW, eds. Williams textbook of endocrinology. 9th ed. Philadelphia: WB Saunders Co, 1998;590-595.
12. Selman PJ, Mol JA, Rutteman GR, et al. Effects of progestin administration on the hypothalamic-pituitary-adrenal axis and glu- cose homeostasis in dogs. J Reprod Fertil Suppl 1997;51:345-354.
13. Juchem M, Pollow K. Binding of oral contraceptive progestogens to serum proteins and cytoplasmic receptor. Am J Obstet Gynecol 1990;163:2171-2183.
14. Toth M, Racz K, Adleff V, et al. Comparative analysis of plas- ma 17-hydroxyprogesterone and cortisol responses to ACTH in patients with various adrenal tumors before and after unilateral adrenalectomy. J Endocrinol Invest 2000;23:287-294.
15. Frank LA, Schmeitzel LP, Oliver JW. Steroidogenic response of adrenal tissues after administration of ACTH to dogs with hyper- cortisolemia. J Am Vet Med Assoc 2001;218:214-216.
16. Kintzer PP, Peterson ME, Mitotane treatment of 32 dogs with cortisol-secreting adrenocortical neoplasms J Am Vet Med Assoc 1994;205:54-61.
17. Feldman EC, Nelson RW, Feldman MS, et al. Comparison of mitotane treatment for adrenal tumor versus pituitary-dependent hyperadrenocorticism in dogs. J Am Vet Med Assoc 1992;200: 1642-1647.