Androgen administration, in contrast, causes a decrease in serum T₄-binding globulin levels and a subsequent decrease in circulating total thyroid hormone levels [4-9]. Although patients with no thyroid disease remain clinically euthyroid when given androgens, the same might not be true in patients receiving thyroid hormone supplementation. No published studies have addressed the potential alterations in thyroid hormone replacement doses in patients receiving androgens. Women with hormone-dependent metastatic breast cancer are often treated with androgens [10, 11]. The common use of levothyroxine as a prescription drug in the general population and particularly in women [1] suggests that approximately 2% to 6% of women treated with androgens may also be receiving thyroid hormone replacement. This investigation was prompted by the development of clinical and biochemical hyperthyroidism in a previously euthyroid woman who was maintained on a stable dose of levothyroxine several weeks after androgen administration. The primary goal of this study was to evaluate alterations in thyroid hormone levels in women treated with androgens and also to investigate the hypothesis that the administration of this gonadal steroid to women with hypothyroidism who receive replacement therapy reduces the required therapeutic dose of thyroid hormone. The study shows that as a result of androgen-induced reduction in circulating levels of T₄-binding globulin, thyroid hormone replacement doses had to be decreased by 25% to 50% to achieve similar therapeutic response.

Patients and Methods

The study comprised two groups of postmenopausal women who received androgens between 1990 and 1993 as a palliative treatment for metastatic, hormone-dependent breast cancer. Androgen therapy (fluoxymesterone, 10 mg orally twice daily) was continued for as long as it was effective in controlling measurable tumor (12 to 92 weeks). All patients were ambulatory and did not receive other medications known to alter thyroid hormone levels or metabolism. None of the patients had evidence of liver or kidney diseases, nor did they have nutritional deficits.

The first group included four women Table 1 who had had primary hypothyroidism for 5 to 20 years. At presentation, they were clinically euthyroid, and the replacement levothyroxine dose had not been altered for several years. All four patients received Synthroid (Boots Pharmaceuticals, Lincolnshire, Illinois) as the form of thyroid hormone replacement therapy and continued to receive the same dose and brand after androgen therapy was initiated.

The second group served as a control Table 1 and included seven consecutive women with no known thyroid disease who also received androgen therapy for metastatic breast cancer during the same interval (1990 to 1993). These patients were clinically and biochemically euthyroid before, during, and after androgen therapy.

Patients were monitored during androgen therapy and were examined for signs of thyroid dysfunction. Blood samples were obtained before treatment, every 2 to 4 weeks for 20 weeks, and every 6 to 8 weeks thereafter until a final sample was drawn 6 to 12 weeks after androgen therapy was stopped. Serum levels of total and free T₄, T₄-binding globulin, thyroid-stimulating hormone (TSH), triiodothyronine (T₃) resin uptake, and the free T₄ index were determined from collected blood samples. The levothyroxine dose given to patients with thyroid disease was altered depending on the clinical evaluation and the results of laboratory studies. Free T₄ levels were measured using commercially available radioimmunoassay kits purchased from INCSTAR Corporation (Stillwater, Minnesota), and TSH levels were determined using the fluoroimmunoassay technique (Wallac Oy, Turku, Finland). Normal serum levels of free T₄ range from 6.4 to 25.7 pmol/L, whereas those of TSH range from 0.5 to 5.0 mU/L. Total T₄ levels_andT₃ resin uptake were measured by using standard methods and kits purchased from Abbott Laboratories (Abbott Park, Illinois). The normal range for total T₄ levels_is 64 to 154 nmol/L; the normal range for T₃ resin uptake is 0.8 to 1.3 nmol/L. The free T₄ index was calculated as the multiplication product of total T₄ and T₃ resin uptake values (normal, 64 to 154). Serum levels of T₄-binding globulin were determined using specific radioimmunoassay methods by the Nichols Institute (San Pedro, California) (normal, 17 to 36 mg/L). The data are presented as the mean ±SD unless stated otherwise. Statistical differences between characteristics of groups 1 and 2 before treatment were analyzed by using the t-test for unpaired values. Statistical analysis of the data obtained after androgen administration to women with no thyroid disease (group 2) was done using the paired t-test. The 95% confidence intervals for the changes in serum levels of total T₄ and T₄-binding globulin during androgen therapy in the latter group were determined using the nonparametric analysis of paired cases [12].

Results

Top Results Discussion Author & Article Info References

Patients without Known Thyroid Disease

As shown in Figure 1, serum levels of total T₄ and T₄-binding globulin decreased by the 4th week (P < 0.001) and remained within a similar range throughout the treatment period. After 4 weeks of androgen therapy, serum levels of total T₄ had decreased by 34.2 nmol/L from baseline (CI, 16.7 to 50.8 nmol/L), and T₄-binding globulin levels had decreased by 10.3 mg/L from baseline (CI, 9 to 13 mg/L). These biochemical changes were clinically significant. The decrease in total T₄ and the increase in T₃ resin uptake levels were of similar proportions such that the calculated values of the free T₄ index were unchanged (Figure 1). Serum TSH levels decreased only slightly at 4 and 8 weeks (P = 0.057 and 0.054, respectively) and remained within the same range during treatment (Figure 1). Measured free T₄ levels were nearly unchanged throughout the treatment period. By the 20th week of treatment, serum levels of total T₄ had decreased from baseline by 46 nmol/L (CI, 34.8 to 57.9 nmol/L), and those of thyroxine-binding globulin had decreased by 12 mg/L (CI, 10 to 14 mg/L). Androgen therapy was discontinued at 20 weeks in three patients and at 22, 27, 63 and 92 weeks in the remaining four patients. Repeat measurement of all variables, which was done 6 to 12 weeks after androgen therapy was discontinued, showed values almost identical to those observed before treatment (data not shown).

View larger version (20K): [in this window] [in a new window]   Figure 1. Serial measurements of total thyroxine (T4) and free thyroxine (fT4), T4-binding globulin (TBG), triiodothyronine resin uptake (T3RU), free thyroxine index (fT4I), and thyroid-stimulating hormone (TSH) levels obtained during androgen administration to seven women with no history of thyroid disease. Each point represents the mean ±SD (error bars). * = P < 0.001 compared with levels before treatment. + = P < 0.001 compared with values before treatment.

Patients Receiving Thyroid Hormone Replacement Therapy

The data for each of the four patients studied are shown in Figure 2. Patient 1 was the index patient seen 20 weeks after androgen treatment was started. Eight weeks after androgen therapy was initiated, routine laboratory data obtained by the primary physician showed a low TSH level (<0.001 mU/L) and a high serum free T₄ level. When the patient was evaluated at 20 weeks, she was clinically thyrotoxic, and the serum levels of free T₄ and TSH were similar to those observed at week 8. Androgen therapy was discontinued within a few days because of disease progression, and symptoms of hyperthyroidism gradually resolved.

View larger version (30K): [in this window] [in a new window]   Figure 2. Free thyroxine (fT4) and thyroid-stimulating hormone (TSH) levels in four women who received long-term treatment for primary hypothyroidism measured before, during, and after androgen therapy for breast cancer. The upper portion of each graph shows the levothyroxine dose used; the arrows indicate the beginning and duration of androgen therapy. Levothyroxine dose was not altered in patient 1 but was reduced during androgen administration to patients 2, 3, and 4 to maintain euthyroidism.

The clinical course of the other three patients is also shown in Figure 2. In all three patients, serum free T₄ levels increased and TSH levels decreased within 2 weeks of treatment. The biochemical changes were associated with clinical symptoms that included tachycardia, hyperhidrosis, and insomnia. The T₄ dose was decreased in all three patients to restore a euthyroid condition. An increase in serum TSH levels and a decrease in free T₄ levels were noted when patients continued to receive the same dose of levothyroxine after androgen therapy was discontinued.

Serum levels of T₄-binding globulin decreased 2 weeks after androgen therapy was initiated and returned to baseline values 8 to 10 weeks after therapy was discontinued. The changes were similar to those seen in the seven euthyroid women who were not receiving thyroid hormone replacement therapy (see Figure 1).

Discussion

Top Results Discussion Author & Article Info References

The data indicate that androgen administration to postmenopausal women caused significant alterations in thyroid hormone levels. The major change was a reduction in the level of T₄-binding globulin that resulted in decreased total T₄ levels and increased T₃ resin uptake determinations. Patients with no history of thyroid disease remained clinically and biochemically euthyroid during androgen therapy. Although serum TSH levels decreased slightly in these patients, all values remained normal. In contrast, patients receiving thyroid hormone replacement therapy had higher free T₄ and lower TSH levels after 2 weeks of treatment. Clinical and biochemical euthyroidism was re-established in these patients after the levothyroxine dose was reduced by 25% to 50%. The process was reversible in both groups of patients as evidenced by the changes in serum levels of TSH, free T₄, and T₄-binding globulin after androgen therapy was discontinued. The reciprocal changes in serum TSH and free T₄ levels during and after androgen therapy and the absence of other illnesses or drugs known to alter thyroid function indicate that these changes were androgen induced. The mechanisms mediating the observed androgen-induced changes are not completely known. Although the decrease in serum T₄-binding globulin levels might be an attractive mechanism, it is unknown whether fluoxymesterone altered thyroid hormone metabolism. Such a possibility was not investigated in this study but could be assessed in the future by using T₄ turnover determinations.

Although androgen administration has been known to lower serum T (₄-binding) globulin levels [4-9], most of the published reports preceded the availability of specific radioimmunoassay for the binding protein, and treatment with this gonadal steroid was given for only a brief period. None of the published reports address effects in patients receiving thyroid hormone replacement. This is the first report to address this issue in detail and show the need to lower thyroid hormone dose after androgen administration. A unique feature of this study was the use of an androgen, fluoxymesterone, that can not be aromatized [13, 14]. This feature limited the alterations in thyroid function in this group of postmenopausal women to alterations induced by androgens. Although the conclusions of this study were based on data obtained from patients receiving thyroid hormone replacement, similar alterations may be seen in patients with autonomous thyroid hormone production, as might occur in those with toxic goiter or Graves disease. It is possible that androgen administration to such patients could worsen or unmask states of mild hyperthyroidism.

Previous studies have reported a decrease in the thyroxine binding capacity of T₄-binding globulin and an increase in the binding to T₄-binding prealbumin (transthyretin) after a brief period of androgen treatment [6, 7]. The study population in the reported studies was not as uniform as in this investigation, and the duration of treatment was less than 2 weeks, making it difficult to compare the results. Although this study did not investigate possible alterations in transthyretin, it is unlikely that such changes were clinically significant because the decrease in serum levels of total T₄ correlated with the decrease in T₄-binding globulin levels. Furthermore, patients receiving long-term thyroid replacement therapy developed clinical and biochemical signs of hyperthyroidism.

The exact mechanisms mediating the effects of androgens on T₄-binding globulin are unknown. A recent study has elucidated the mechanism by which estrogens increased the serum levels of this binding globulin [15]. The study showed that estrogens caused increased sialylation of newly synthesized T₄-binding globulin, a process that decreased clearance of the protein [15]. Although estrogens were also shown to increase biosynthesis of this binding globulin [16], it is currently believed that the predominant mechanism mediating the increase in T₄-binding globulin is that involving alterations of sialylation. No comparable studies have investigated androgen effects. It is unknown whether androgens decrease sialylation of newly synthesized T₄-binding globulin, a process that might cause increased clearance and decreased circulating levels. That the effects lasted for months suggest that inhibition of biosynthesis of this binding globulin could be an additional mechanism mediating androgen action.