Octreotide Therapy for Thyroid-Stimulating Hormone-Secreting Pituitary Adenomas: A Follow-up of 52 Patients

Methods

Patients

Twenty-seven men and 25 women who were 16 to 84 years old (median age, 44 years) and had a TSH-secreting adenoma were evaluated. Studies were done at 24 medical centers in nine countries (Appendix). Data were collected from reports published in the English or French language between 1987 and 1991 [2,10-36]. The list of published articles and abstracts of papers presented at various national and international meetings was provided by the Centre de Documentation of Laboratoires Sandoz, Rueil-Malmaison, France. The authors of these studies were asked to provide us, when possible, with follow-up information on their patients [2, 10-36]. In 6 of these patients (patients 30 to 35), available data were too imprecise or incomplete, particularly regarding TSH levels, to allow pertinent evaluation [29-32]. However, some data on these patients were analyzed. Data on 15 additional patients (patients 38 to 52) whose studies were done in various centers (see Appendix) and not previously reported were included in the final analysis. In all patients, the diagnosis of thyrotropinoma had been made on the basis of an inappropriate secretion of TSH (that is, the coexistence of increased serum thyroid hormone levels and an increased or normal [but at least unsuppressed] TSH serum level [3]. In all but 5 patients, who had surgically proven microadenoma, macroadenoma generally associated with suprasellar or laterosellar extension was shown by computed tomographic (CT) scan or magnetic resonance imaging (MRI).

Before octreotide therapy, 23 patients had undergone pituitary adenomectomy; 9 of these patients had also had radiotherapy of the pituitary area without resolution of hyperthyroidism. At the time octreotide therapy was initiated, some patients were receiving antithyroid medication because of hyperthyroidism or thyroid hormone replacement because of previous thyroidectomy or radioiodine treatments, and only fragmentary information could be obtained in these cases. In all but 4 of the remaining patients (90%), hyperthyroidism was present before treatment. Basal serum TSH levels were increased (5.2 to 129 mU/L) in 23 of 49 patients and normal (but inappropriately unsuppressed in the presence of high thyroid hormone levels) in the remaining patients. Thirty patients had increased serum levels (1.5 to 311 g/L) of free -subunitthe uncombined subunit of glycoprotein hormones often secreted in excess together with TSH by TSH-secreting adenomas [3]whereas 4 of the 37 patients studied had normal levels. Nine of the patients had mixed growth hormone-TSH-secreting adenoma and also had the classic features of acromegaly.

Treatment Regimen

The initial dose of octreotide (Sandostatin, Sandoz Pharma, Ltd., Basel, Switzerland) ranged from 50 to 100 g subcutaneously two or three times per day. The dosage was increased according to individual biochemical responses to a maximum level of 500 g three times per day. The median octreotide dose at the final evaluation was 300 g/d. In 25 patients, treatment was extended beyond 3 months (range, 3 to 61 months; mean, 20 17 months).

Evaluation

Patients were evaluated at various intervals determined by the protocol of the medical center at which they were treated. Data from the pretherapeutic evaluation were available in 49 patients. The response to the first injection of octreotide (50 or 100 g), although assessed in 40 patients, was provided in 35. Evaluation of TSH levels was done after 1 to 2 weeks of octreotide therapy in 33 patients. Evaluation was done again after 1 month of therapy in 23 patients. In the 25 patients receiving long-term therapy (> 3 months), evaluation was done at 3-month intervals during the first year and thereafter at 6-month intervals.

For each evaluation, we attempted to retrieve measurements of serum TSH, free -subunit, and thyroid hormone levels done on blood samples collected in the morning or, when available, hourly during a 3- to 12-hour profile; mean values were used for analysis. Levels of TSH, -subunit and thyroid hormone (generally as free thyroxine, free triiodothyronine, or both) were measured by immunoradiometric assay or radioimmunoassay using commercially available kits at the individual study centers. In consideration of the different assays used, the normal ranges for TSH and free thyroxine levels were set at 0.1 to 5 mU/L and 10 to 20 pmol/L, respectively. Methods of serum free -subunit determination were too heterogeneous to allow within-group comparison. Thus, the individual course of -subunit was assessed according to the normal reference range indicated by the individual medical center. In general, hormonal responses are indicated as a percentage decrease in hormone levels or expressed in terms of whether they returned to within the normal range (as defined by the investigators).

Serial anatomic evaluation of the pituitary gland by either CT scan or MRI was available in 26 patients. In 7 patients, the percentage change in tumor volume was calculated.

Abdominal ultrasound examinations were done at regular intervals and at the final evaluation in 19 of the 25 patients receiving long-term therapy.

Statistical Analysis

Statistical analysis was done using a Wilcoxon test for paired data. Results are expressed as mean SD.

Results

Hormonal Effects

The acute response of serum TSH levels to the first injection of octreotide (50 to 100 g) was evaluated in 35 patients (Figure 1). The TSH level decreased in all but 2 of these patients (mean decrease for the whole group, 55.8% 27%), the nadir occurring between 3 and 6 hours after injection. The -subunit level decreased in 15 of the 19 patients assessed (mean decrease, 37.5% 24%).

Thirty-three patients were evaluated for the short-term effect of octreotide (Figure 2). The mean pretreatment TSH concentration was 14.7 25.9 mU/L. Reduction of TSH levels after 1 to 2 weeks of octreotide therapy (50 or 100 g two or three times daily) was observed in 30 patients (mean decrease for the whole group, 74.1%). Serum TSH levels decreased to a mean level of 3.8 7.4 mU/L (P < 0.01). Among patients whose basal serum TSH levels were supranormal, 88% showed a reduction of more than 50% in the TSH level; in 72%, the TSH level returned to normal. The -subunit levels, assessed in 19 patients, showed a similar pattern, with a mean reduction of 64.3% (24%). In about two thirds of the patients, the TSH response after 1 to 2 weeks was better than after the first dose; this analysis included 5 patients who did not respond to acute injection with a reduction in TSH level of more than 50%. Thyroid hormone levels decreased in all patients and returned to normal in 73% of patients. Patients had similar TSH and thyroid hormone responses to octreotide, regardless of whether they had a pure TSH-secreting adenoma or a mixed growth hormone-TSH-secreting adenoma. After 1 month of treatment, serum TSH (n = 23) and -subunit (n = 11) levels were similar to those seen after 1 week of treatment (data not shown). In 3 patients, despite the lack of normalization of serum TSH levels, thyroid hormone levels returned to normal after 1 week of therapy [10, 17]. The relation between individual dosage and the hormonal response was impossible to assess because of the different octreotide doses used (ranging from 100 to 300 g/d). Although 2 of the patients studied showed persistently suppressed TSH levels, thyroid hormone levels that initially returned to normal demonstrated an escapedefined as the re-increase in serum levels despite increasing dosageearly between weeks 2 and 4 of treatment. This led to the discontinuation of treatment in these two patients.

Octreotide therapy was continued beyond 3 months (range, 3 to 61 months; mean, 20 17 months) in 25 patients, with daily doses ranging from 100 to 1500 g. Overall, the response to treatment, as assessed by TSH and -subunit levels, was better than or similar to that after short-term treatment. Based on the persistence of normal thyroid hormone levels, efficacy was maintained in 21 of the 25 patients (84%) as long as the treatment was continued. Tachyphylaxisdefined as the necessity to increase doses to maintain normal TSH levelswas observed in 5 of the 21 patients. Thyroid hormone levels did not increase, however, and, in fact, remained normal in 4 of these patients. In all cases, these alterations resolved with doubling of the octreotide dose. By contrast, 3 of the 25 patients receiving long-term therapy showed true escape; their thyroid hormone (n = 3) and TSH (n = 1) levels increased despite incremental doses of octreotide (reaching a dose of 1500 g/d in two cases). Last, 1 patient remained resistant to treatment, with respect to both thyroid hormone and TSH levels.

In all patients in whom treatment was temporarily interrupted or permanently discontinued, serum TSH, -subunit, and thyroid hormone levels returned to pretreatment levels.

Some investigators tried to reduce the octreotide dose. One patient whose dose was reduced from 450 to 300 g per day showed increased serum TSH levels, which resolved when the dose was increased again. Another patient had an increase in serum TSH and thyroid hormone levels when the octreotide dose was reduced from 600 to 400 g per day. Conversely, a reduction in daily dose from 300 to 100 g did not modify the effects of the drug in 5 other patients.

Other Clinical Effects

Improvements in visual field defects after octreotide therapy were remarkable in two patients [11, 24]. In one, the improvement was noticeable by hour 3 of treatment [24]. Despite initial improvement, the visual fields in the other patient worsened when treatment was discontinued. Reinstitution of therapy in this patient was again accompanied by a recovery of visual defects. In this patient, tumor volume as assessed by CT scan remained unchanged. The improvement in these two patients was sustained during the 2 years of treatment; an increase in the octreotide dose was necessary in only one of them [24].

Reduction of tumor volume as assessed by CT scan or MRI was observed in 10 of the 25 patients receiving long-term therapy. In 7 patients, the tumor shrinkage was obvious, approaching 30% to 70%. In the 3 other patients, decreases in tumor volume were either slight or unspecified. An 11th patient showed reduced tumor volume after only 20 days [20]. Positron emission tomographic scanning, which is probably a better means to assess the effects of treatment on tumor-viable tissue, showed changes in tumor metabolism during octreotide therapy in 2 patients [23, 31]. No patients showed increased tumor volume during octreotide therapy.

Side effects were reported by 4 of the 15 new patients and were mentioned in only a few previous reports [10, 12, 13, 16, 18, 24, 28]. Side effects were generally limited to transient abdominal discomfort and loose stools. Of the 21 patients receiving long-term therapy who underwent systematic ultrasound examination before and during the treatment period, 3 developed asymptomatic gallstones. Gallstones occurred in 2 patients after 6 and 12 months of treatment and disappeared within 4 months of therapy with ursodeoxycholic acid. Carbohydrate intolerance developed in 3 patients [10, 22]. Conversely, glucose concentrations improved in 1 diabetic patient [10].

Discussion

Our aim was to assess the effects of octreotide on TSH and thyroid hormone hypersecretion and tumor volume in patients with TSH-secreting adenomas. Such assessment required a large series of patients. Because of the rarity of TSH-secreting adenomas, we combined our unpublished results with updated data on previously reported cases.

Octreotide was clearly effective in decreasing TSH levels from TSH-secreting adenomas; only three patients in this series failed to respond. This finding is probably related to the high number of somatostatin-binding sites on the surface of the adenomatous cells. Increased numbers of binding sites were found on tumors from two patients [22, 25]. The heterogeneous distribution of somatostatin-binding sites and their variable number or affinity [37] may explain the varying response to octreotide observed in vivo.

The progressive suppression in TSH response to octreotide observed between the first injection (55% reduction) and the end of the first or second week of treatment (74% reduction) leads us to advocate delaying final evaluation of the efficacy of octreotide therapy until the patient has received treatment for 1 to 4 weeks. In addition, even if TSH levels remain elevated, thyroid hormone levels may decrease or return to normal during continued octreotide therapy. This finding suggests the possible effect of octreotide on TSH bioactivity and may be of value in patients with severe thyrotoxicosis due to a TSH-secreting pituitary tumor. Indeed, antithyroid drugs may be contraindicated in these situations because of their potential activation of tumor growth via hypothyroid-induced TSH production. This phenomenon occurred in one patient from this series before octreotide therapy was instituted.

The long-term effects of treatment were sustained in most of the study participants. Only 12% of patients had a true escape, showing an increase in thyroid hormone levels (with or without an increase in TSH levels) sufficient to prompt termination of treatment. Five of 21 patients developed tachyphylaxis based on TSH level, thyroid hormone level, or both. Adjusting the dosage adequately maintained the effects of the drug on hormonal levels in all patients. It must be emphasized that the control of TSH and thyroid hormone levels was achieved in most patients with doses lower than those used to treat patients with acromegaly [38]. In half of the patients, an octreotide dose of only 100 or 200 g daily was necessary. However, a dose of 300 g/d was used in most patients who showed tumor shrinkage.

Tumor shrinkage, observed in 10 of the 25 patients who had long-term follow-up, is important in view of the size and invasiveness of TSH-secreting pituitary adenomas. In all but 1 patient, in whom the shrinkage was striking [24], reduction in tumor size was similar to that seen in acromegalic patients who received octreotide therapy [38]. Tumor shrinkage, however, was not as dramatic as that seen in patients with prolactinoma who received bromocriptine therapy. Thus, one must ask whether static measurements of tumor volume by CT scan or MRI are good indicators of the effects of treatment on tumor-viable tissue. In fact, certain patients may show clinical benefits, such as improvement in visual fields, without evidence of a change on CT scan or MRI [11]. This may relate to the presence of fibrotic and nonviable tumor tissue as well as to the edema that may persist during treatment and obscure modifications occurring in viable tissue. In contrast, measurement of metabolic activity by positron emission tomographic scan seems much more sensitive to the relevant tumor tissue, as shown in 2 patients [23, 31]. Thus, given the growth potential of these tumors, the lack of change on MRI or CT scan might still be consistent with shrinkage or, at least, with absence of growth of the tumor. Only a controlled, therapeutic trial could answer these questions in a definitive fashion, but such a trial is probably not feasible, given the rarity of these tumors. Nonetheless, considering the low rates of cure with other treatments (surgery and radiotherapy), octreotide therapy for TSH-secreting adenoma remains a viable solution.

Twenty-four medical centers in nine countries were involved in this investigation, and the heterogeneity in diagnosis, management, and evaluation of patients may have been a limiting factor in the study. However, because of the rarity of TSH-secreting pituitary adenomas, the analysis of heterogeneous data was the only way to provide a meaningful assessment.

In conclusion, octreotide should be considered an effective treatment for TSH-secreting adenoma when surgery and radiotherapy have failed to cure the disease. Octreotide reduced TSH secretion in almost all patients treated and normalized thyroid hormone levels in 73%. Partial tumor shrinkage, observed in 40% of the patients receiving long-term therapy, may be of value in the management of these rare but often large and invasive tumors. Whether this treatment has a role as an alternative to surgery in selected patients remains to be determined.

Appendix

The following is a list of clinical investigators who participated in the International Multicenter Study on TSH-secreting Adenomas.

Investigators providing data on new cases included P. Chanson, A. Warnet, and P. J. Guillausseau: Centre Hospitalier Universitaire, Lariboisiere, Paris, France; J. Bertherat: Inserm U159, Paris, France; P. Bouchard and G. Schaison: Centre Hospitalier Universitaire Bicetre, Kremlin-Bicetre, France; D. Dewailly and P. Fossati: Centre Hospitalier Regional et Universitaire, Lille, France; A. Navaranne and P. Roger: Centre Hospitalier Regional et Universitaire, Bordeaux, France; A. Beckers and A. Stevenaert: Centre Hospitalier Universitaire, Liege, Belgium; G. Moreiro: Hospital Son Dureta, Palma Mallorca, Spain; and M. Montini: Ospedale Riuniti, Bergamo, Italy.

Investigators providing follow-up data for previously reported cases included B. Weintraub: National Institutes of Health, Bethesda, Maryland; W. Malarkey: Ohio State University, Columbus, OH; G. Francia: Ospedale Policlinico, Verona, Italy; A. Levy: Westminster Hospital, London, United Kingdom; C. Chapman: The General Infirmary, Leeds, United Kingdom; M. Linquette and J. L. Wemeau: Centre Hospitalier Regional et Universitaire, Lille, France; and C. Houdent: Centre Hospitalier Regional et Universitaire, Rouen, France.