Megestrol acetate is a progestational agent that is structurally similar to progesterone. The observation that patients with cancer who are treated with megestrol acetate gain weight has led to the use of this agent in patients who have cachexia associated with the acquired immunodeficiency syndrome (AIDS) [1-3]. Although the mechanism of action of megestrol acetate is unknown, treatment with this drug has been reported to be safe [1, 2, 4, 5]. At high doses, progestins have been found to have glucocorticoid activity [6], which is presumably due to cross-reactivity within the superfamily of steroid receptors [7-9]. If megestrol acetate has glucocorticoid activity, then high doses of this drug, although insufficient to cause overt Cushing syndrome, could cause suppression of the hypothalamic-pituitary-adrenal axis. Prolonged pituitary suppression of the secretion of adrenocorticotropic hormone (ACTH) could be expected to lead to secondary adrenal insufficiency. This adverse effect has been seen in patients treated with the progestational agent medroxyprogesterone acetate [10], but similar effects have not yet been ascribed to megestrol acetate. We are investigating the glucocorticoid activity of megestrol acetate in patients with AIDS, and we recently observed a patient with AIDS in whom megestrol acetate appeared to have induced adrenal insufficiency.

Case Report

Top Case Report Discussion Author & Article Info References

A 32-year-old man, who was diagnosed as positive for the human immunodeficiency virus (HIV) in April 1987 and who had cryptococcal meningitis in January 1991, presented with weight loss, nausea, vomiting, orthostatic hypotension, and profound fatigue of 2 weeks' duration. He indicated that he had been receiving at least 240 mg of megestrol acetate daily for 4 years but that he had stopped receiving it 14 days before presentation. He was taking additional medications, including fluconazole, 200 mg/d; clarithromycin, 500 mg twice daily; acyclovir, 800 mg four times daily; dapsone, 100 mg/d; and multiple vitamins. As a participant in AIDS Clinical Trials Group (ACTG) protocol 193A, the patient was also receiving one or more of the following in double-blind fashion: zidovudine, didanosine, zalcitabine, or nevirapine. On physical examination, he was hypotensive (blood pressure, 90/50 mm Hg), lethargic, and pale, with no hyperpigmentation. Laboratory evaluation showed normal levels of sodium and potassium (141 mEq/L and 4.3 mEq/L, respectively), a CO₂ concentration of 18 mmol/L, and a glucose level of 94 mg/dL. However, the patient's morning cortisol level was undetectable at less than 1.0 µg/dL (normal, 6 to 18 µg/dL), and his ACTH level was low at 4.0 pg/mL (normal, 9 to 52 pg/mL). One hour after receiving 250 µg of cosyntropin intravenously (Cortrosyn; Organon, Inc., West Orange, New Jersey), the patient's plasma cortisol level increased to only 2.4 µg/dL (normal, > 20 µg/dL), indicating adrenal insufficiency. Magnetic resonance imaging showed the adrenal glands to have a normal appearance.

The patient began to receive hydrocortisone, 50 mg/d; this resulted in the prompt and dramatic amelioration of his symptoms. The dose of hydrocortisone was soon decreased to physiologic replacement levels of 30 mg/d in divided doses. The patient then had a prolonged cosyntropin-stimulation test (250 µg intravenously daily for 3 days). By the third day, the patient's plasma cortisol level had increased from an undetectable baseline level to 27.1 µg/dL. Further attempts to wean the patient from hydrocortisone were met with the prompt return of symptoms.

Prospective Study

Methods

Four patients with AIDS who were selected by their primary physicians to begin receiving megestrol acetate (Megace; Bristol-Myers, Princeton, New Jersey) for cachexia were studied. Three were men and one was a woman; they ranged in age from 26 to 49 years. All were ambulatory; none were known to have an active opportunistic infection.

Baseline studies included a routine physical examination and measurement of morning cortisol and ACTH levels. Additional measurements of cortisol levels were obtained 1 hour after intravenous administration of 250 µg of cosyntropin. A 75-g oral glucose tolerance test was done before and within 1 week after commencement of therapy with megestrol acetate. One patient did not have an oral glucose tolerance test because of preexisting diabetes mellitus treated with daily insulin. After 1 month of therapy with megestrol acetate (80 mg three times daily), basal ACTH values were obtained and a short cosyntropin test was repeated. One patient became acutely ill, required hospitalization, and dropped out of the study before the cosyntropin test was repeated.

Statistical analysis was done using the paired Student t-test.

Results

All three patients who received megestrol acetate for at least 1 month had gains in total body weight: One had gained 9 pounds at 1 month, one had gained 7.5 pounds at 1 month, and one had gained 16 pounds at 3 months. All patients studied had normal basal cortisol and ACTH levels, and all had normal responses to stimulation with cosyntropin. One month after the initiation of therapy with megestrol acetate, the mean basal morning cortisol level had decreased from 11.0 ±1.8 µg/dL to 1.5 ±0.9 µg/dL (P < 0.01) (Figure 1). The stimulated cortisol levels had decreased from 27.3 ±3.3 µg/dL to 9.3 ±6.3 µg/dL (P = 0.01). Basal ACTH values had decreased to below normal, from 16.6 ±6.3 pg/mL to 6.3 ±3.3 pg/mL (normal, 9 to 52 pg/mL; P = 0.02).

View larger version (16K): [in this window] [in a new window]   Figure 1. Response of the pituitary-adrenal axis to megestrol acetate therapy in three patients. Top. Plasma adrenocorticotropic hormone (ACTH) values before and after 1 month of therapy with megestrol acetate. The dotted line represents the lower limit of the normal range (9 pg/mL). Bottom. Response to a 1-hour cosyntropin-stimulation test at baseline and after 1 month of therapy with megestrol acetate. The normal response is an increase in plasma cortisol level to more than 20 µg/dL.

In two patients, oral glucose tolerance testing indicated that insulin resistance had developed while the patients were receiving megestrol acetate. This was shown by increased glycemic response, elevated insulin levels, or both, relative to values before treatment. In a third patient, who had preexisting diabetes, insulin requirements increased precipitously within days of starting therapy with megestrol acetate. Despite an increase in daily insulin dose from 0.47 U/kg to more than 8.3 U/kg, daily glucose levels increased. When therapy with megestrol acetate was stopped 3 months later, insulin requirements returned to the levels seen before the administration of megestrol acetate.

Discussion

Top Case Report Discussion Author & Article Info References

The cortisol and ACTH levels of our prospectively studied patients decreased significantly while these patients were receiving megestrol acetate. After only 1 month of therapy, response to stimulation with cosyntropin had become subnormal. This finding indicates pituitary suppression and suggests that megestrol acetate has glucocorticoid properties. Some investigators have found evidence that other progestational agents have glucocorticoid activity [6, 11]. Medroxyprogesterone acetate is reported to have caused the Cushing syndrome in a woman [7] and mild adrenal suppression in children treated for precocious puberty [12]. Decreased plasma cortisol levels were recently observed in 14 patients receiving megestrol acetate for malignancy [13]; ACTH levels were not elevated in these patients, a finding that rules out primary adrenal insufficiency.

The glucocorticoid activity of megestrol acetate has been observed in animal studies, and withdrawal symptoms have been noted after abrupt discontinuation of therapy [14-17]. In one study, megestrol acetate (270 mg/d) led to worsening glucose tolerance in patients with breast cancer [18], and the development of diabetes in a patient with AIDS who was treated with megestrol acetate was recently reported [19].

Some authors have suggested that the adrenal suppression seen with megestrol acetate is not caused by glucocorticoid activity per se [13], and they cite as evidence the absence of reports of the Cushing syndrome developing during treatment with megestrol acetate or of adrenal insufficiency occurring after its withdrawal. However, it should be noted that in two recent studies of megestrol acetate in patients with AIDS, studies in which adrenal insufficiency was not reported, treatment duration was only 12 weeks [1, 2]. Further, patients with adrenal suppression might not be expected to develop overt adrenal insufficiency in the absence of acute stress. In the case presented here, we strongly suspected a diagnosis of adrenal insufficiency because we had been investigating the actions of megestrol acetate.

We believe that the patient we describe in our case report represents the first recognized case of adrenal insufficiency resulting from the withdrawal of therapy with megestrol acetate. This patient had prolonged suppression of the pituitary-adrenal axis after megestrol acetate was withdrawn. He responded to prolonged stimulation with cosyntropin, thus ruling out the possibility of primary adrenal insufficiency, which can be seen in patients with AIDS. He also had inappropriately low ACTH levels and responded well to replacement hydrocortisone. These observations suggest that megestrol acetate has glucocorticoid action sufficient to cause clinically significant adrenal suppression in some patients. Consistent with this hypothesis is the observation that, in the patients we prospectively studied, megestrol acetate appeared to induce peripheral insulin resistance. This was shown by a decreased glucose tolerance in one patient and increased insulin levels in another during the glucose tolerance test, and by a 10-fold increase in insulin requirements in a patient with preexisting diabetes.

The mechanism that causes weight gain in patients treated with megestrol acetate is unknown. We believe that this weight gain may be due in part to glucocorticoid activity because it is common in patients with glucocorticoid excess. Two recent randomized trials found that most of the weight gained by patients with AIDS who were given megestrol acetate was due to an increase in body fat [1, 2]. In one study [1], lean body mass did not increase. The investigators in this study [1] also noted that weight gain tended to be central, as is typical with the Cushing syndrome, and that one patient developed a "Cushingoid" appearance.

That megestrol acetate has glucocorticoid activity has important implications. Clinicians using progestins such as megestrol acetate must be alert to the potential for secondary adrenal insufficiency after withdrawal of these drugs after prolonged therapy.