Ectopic Prolactin Secretion From a Uterine Leiomyoma

Abstract

Ectopic hormone production is well recognized, but ectopic production of prolactin has been reported infrequently. We report here the case of a 47-year-old woman who had hyperprolactinemia (213-224 ng/mL) causing galactorrhea and hypogonadism. Cabergoline treatment, 1.0 mg twice a week, did not lower the prolactin level at all, but excision of a large uterine leiomyoma corrected the hyperprolactinemia and the hypogonadism. The excised leiomyoma tissue exhibited immunostaining for prolactin, confirming by this method for the first time that a uterine leiomyoma was the cause of hyperprolactinemia. This case illustrates the need to consider an ectopic source of prolactin in a patient who has hyperprolactinemia that is not associated with a large sellar mass and is completely resistant to cabergoline.

Ectopic production of hormones is well recognized, but clinically significant hyperprolactinemia due to ectopic production of prolactin is not common. Only a few reports describe hyperprolactinemia associated with uterine leiomyomas [1-4]. We report here a patient who had clinically significant hyperprolactinemia that was corrected by excision of a large uterine leiomyoma. This is the first case in which positive immunostaining for prolactin confirmed that the leiomyoma was the source of the prolactin.

1. Case Report

A. Clinical History

A 47-year old woman presented in June 2018 because of amenorrhea for one year and galactorrhea for one month. She also reported hot flashes and decreased libido. She was not taking prescription medications or dietary supplements. On examination, she had expressible galactorrhea bilaterally. Her serum prolactin concentration was 213 ng/mL (Table 1). Free prolactin was 97.2% of the total. The estradiol was 19 pg/mL, follicle-stimulating hormone (FSH) 5.4 mIU/mL, luteinizing hormone (LH) 1.1 mIU/mL, free thyroxine 1.1 ng/dL, cortisol 11.3 µg/dL, and insulin-like growth factor-1 177 ng/mL (Z score, 0.5). (All laboratory tests were performed at Quest Diagnostics, Horsham, PA.) Magnetic resonance imaging (MRI) revealed a sella turcica that was slightly large but a pituitary gland that was small and flattened against the sellar floor.

Table 1.

Prolactin and Estradiol Responses to Cabergoline and Hysterectomy in a Woman Whose Uterine Leiomyoma Caused Hyperprolactinemia

Date	Prolactin	Estradiol	Cabergoline Dose
	(ng/mL)	(pg/mL)	(mg/week)
6/22/18	213	20	-
8/3/18	203	-	1.0
10/12/18	213	16	2.0
6/24/19	224	18	-
8/14/19	Hysterectomy
9/20/19	2.5	-	-
11/5/19	8.9	114	-

She was treated with cabergoline, 0.5 mg twice a week, and 6 weeks later, the prolactin level remained elevated, 203 ng/mL (Table 1), so the dose was increased to 1.0 mg twice a week. Ten weeks later, the prolactin level remained elevated, 213 ng/mL, so cabergoline was discontinued. Seven months later, the prolactin level was 224 ng/mL.

During the time that the hyperprolactinemia was being evaluated, her gynecologist was monitoring her for a uterine leiomyoma. When ultrasonography showed that the leiomyoma had increased from 8.3 × 8.6 × 9.8 cm in June 2018 to 13.9 × 10.4 × 11.8 cm in June 2019, the gynecologist performed a total abdominal hysterectomy and salpingectomy but not ovariectomy. One month after surgery, the galactorrhea and hot flashes had resolved, and the prolactin level was normal, 2.5 ng/mL (Table 1). Six weeks later, the prolactin level remained normal, 8.7 ng/mL, and the estradiol level had increased to normal, 114 pg/mL, accounting for resolution of her hot flashes. The FSH was 3.7 mIU/mL, and LH was 4.9 mIU/mL.

B. Pathological Examination and Immunohistochemical Staining (Fig. 1)

Figure 1.

Pathologic examination of a large leiomyoma excised from a 47-year-old woman who had hyperprolactinemia. The upper panel shows hematoxylin and eosin staining, demonstrating dense proliferation of spindled cells growing in a fascicular pattern. The lower panel shows immunostaining for prolactin, demonstrating focal but intense positive cytoplasmic staining.

Pathological examination showed that the excised uterus contained a 12 × 10 × 6 cm mass. Hematoxylin and eosin staining of sections of the mass showed a neoplasm composed of a dense proliferation of spindled cells growing in a fascicular pattern admixed with thick-walled vessels, consistent with smooth muscle. Cell nuclei were elongated and ovoid. There was no evidence of increased mitotic activity or tumor cell necrosis. Immunohistochemical staining for prolactin showed focal but intense, positive punctuate staining.

2. Discussion

A 47-year-old woman had persistent clinically significant hyperprolactinemia that was resistant to cabergoline but that resolved completely after excision of a large uterine leiomyoma. Immunocytochemical staining confirmed that the leiomyoma was the source of the excessive production of prolactin.

Hormone secretion from an ectopic site is well recognized and is usually of single chain peptide hormones. Clinically significant ectopic hormone production by a uterine leiomyoma is uncommon, but secretion of parathyroid hormone-related peptide [5] and erythropoietin [6], as well as prolactin [1-4], have been reported. Ectopic prolactin production has also been reported from other tumors, including gonadoblastoma [7], ovarian teratoma [8], perivascular epithelioid cell tumor [9], uterine cervical carcinoma [10], and colorectal adenocarcinoma [11].

Although uterine leiomyomas have rarely been reported to cause hyperprolactinemia, cultured leiomyoma cells [12, 13], cell lines derived from leiomyoma muscle cells [14], and even normal myometrial cells in culture [15, 16] have been shown to produce prolactin messenger ribonucleic acid and to secrete immunoreactive prolactin into culture medium. The secreted prolactin is similar to that secreted by the pituitary gland immunologically, as shown by parallel dilution curves, and by size, as shown by gel filtration chromatography [12, 13]. Synthetic thyrotropin-releasing factor (TRH), however, did not stimulate prolactin secretion [14].

The patient presented here is similar in many ways to the few cases previously reported of patients who had hyperprolactinemia associated with a uterine leiomyoma: hyperprolactinemia of a modest degree (<250 ng/mL), but persistent and causing hypogonadism; resistance to dopamine agonist treatment; and cure of hyperprolactinemia by surgical excision of the leiomyoma [1-4]. Also similar to previously reported cases, this patient’s leiomyoma was very large, >8 cm in diameter, suggesting that these tumors produce and secrete prolactin inefficiently [1-4]. In the patient presented here, for the first time, immunostaining confirmed that the leiomyoma was the source of the prolactin.

The significance of this case is that it illustrates the need to consider ectopic prolactin secretion when hyperprolactinemia occurs in the absence of a pituitary macroadenoma and is completely resistant to dopamine agonist treatment. The vast majority of patients who have hyperprolactinemia as the result of a lactotroph adenoma experience a reduction in prolactin in response to dopamine agonist treatment. Although a well-documented minority of patients who have lactotroph adenomas exhibit some degree of resistance to dopamine agonist treatment, most lactotroph adenomas that are resistant are >1 cm in diameter and, therefore, recognizable by MRI, and are only relatively resistant in that increasing doses of cabergoline do reduce prolactin levels to some degree [17]. When hyperprolactinemia is the result of most other causes, such as stalk compression, the lactotroph cells are normally responsive to cabergoline.

We conclude that finding hyperprolactinemia that is not associated with a large sellar mass and is completely resistant to cabergoline treatment warrants consideration of an ectopic source of prolactin secretion, such as a uterine leiomyoma.

Glossary

Abbreviations

FSH

follicle-stimulating hormone

LH

luteinizing hormone

MRI

magnetic resonance imaging

Additional Information

Disclosure Summary: The authors have nothing to disclose.

Data Availability: Data sharing is not applicable to this article as no datasets were generated or analyzed during the current study.