Journal Logo

A Fatal Malignant Macroprolactinoma in a Female Patient: Case Report and Review of the Literature

Giusti, Massimo M.D.; Foppiani, Luca M.D.; Guido, Roberta M.D.; Losa, Marco M.D.; Terreni, Maria Rosa M.B.; Valenti, Sandra M.D., Ph.D.; Giordano, Giulio M.D.

Author Information
  • Free

Abstract

Introduction

The clinical presentation of prolactinoma varies greatly depending on the patient’s sex. Women usually present with oligomenorrhea and galactorrhea secondary to a microadenoma, while men show symptoms such as decreased libido, impotence and symptoms due to mass effect from a macroadenoma [1]. Malignant prolactinomas are uncommon [2,3].

Pituitary carcinomas are rare adenohypophyseal tumors whose prevalence is thought to be 0.1–0.2% of all pituitary tumors [2]. The definition, diagnosis, therapy and prognosis of these neoplasms are still controversial. The diagnosis of malignancy is based on the pattern of growth and spread rather than on histological criteria [3]. An additional criterion based on increased proliferation indexes, such as proliferating cell nuclear antigen (PCNA) and Ki-67 may be helpful [4]. To date, a few dozen pituitary carcinomas have been reported in literature, of which 25–30% were prolactinomas [4]. Various therapeutic strategies (surgery, dopamine-agonists, radiotherapy and chemotherapy) were used to treat these neoplasms, but the results were palliative rather than curative [5–7].

A slight increase in the prevalence of these malignant neoplasms occurs in men [3]. We present a fatal case in a young woman with a PRL-secreting macroadenoma. Progressive resistance to dopaminergic therapy coupled with an intracranial metastasis and a rapid growth of the pituitary mass after several years of satisfactory response to common dopaminergic treatment led to her demise.

Case Report

A 27-year-old woman with a long history (14 years) of oligomenorrhea, irregularly treated with estro-progestins, and a 3-year history of mild galactorrhea and moderate visual field defects underwent trans-sphenoidal adenectomy for a large sellar and suprasellar pituitary adenoma. Hormonal evaluation before surgery showed low-normal gonadotropin levels, normal thyroid and adrenal function and PRL levels ranging between 90–180 μg/L (normal values <20 μg/L).

The neoplasm was histologically classified as a sparsely granulated-cell adenoma with invasion of the dura mater. No immunohistochemical study was performed. Post-surgical CT revealed no tumor. PRL levels returned to normal (11–15 μg/L) Menses, however, did not return.

Two years later, she complained polyuria, weakness, amenorrhoea and hypotension. Baseline PRL levels ranged from 70–90 μg/L. The response to TRH testing (200 μg i.v.) peaked at 130 μg/L. Baseline gonadotropin levels were low (LH: 0.5–1 IU/L, FSH: 1.5–2 IU/L) and did not respond to GnRH. TSH levels were in the low-normal range and associated to low thyroxine levels. Urinary free cortisol was normal as was urine osmolarity. Visual fields were normal. Post-surgical panhypopituitarism was diagnosed. Therapy with L-thyroxine (50 μg/day), cortone acetate (12.5 mg/ day) and desmopressin (5 μg/day) was started. PRL levels fell to normal with bromocriptine (5 mg/day) and clinically she was much improved.

A few months later, visual field defects were found. CT-scan recurrence of the neoplasm. PRL levels ranged between 25–35 μg/L during bromocriptine therapy. A second transcranial operation was performed, followed by radiotherapy (total dose 50 Gy). Hyperprolactinemia (30–35 μg/L) persisted after both treatments and bromocriptine (5 mg/day) was restarted. In the following seven years, CT and magnetic resonance imaging (MRI) examinations showed no evident recurrence of the neoplasm, and PRL levels remained within the normal range (2–15 μg/L) with bromocriptine therapy.

When the patient was 37 years old, a slight increase in PRL levels was observed (38–40 μg/L), which continued to rise during the following months despite increasing bromocriptine dose (10 mg/day). MRI revealed a recurrence of the macroadenoma with invasion of the left sphenoidal sinus. Because of the bromocriptine resistance, and intolerance to cabergoline (nausea, postural hypotension), therapy was changed to the non-ergot dopamine-agonist quinagolide. Dosages up to 600 μg/day failed to lower PRL levels (155–187 μg/L).

Six months later, MRI examination revealed a left frontal mass (Fig. 1) in addition to the known sellar recurrence, which now invaded the cavernous sinus and compressed the internal carotid and the left trigeminal nerve. The frontal neoplasm, initially supposed to be a radiotherapy-induced meningioma, was surgically removed. Histological analysis of the mass was surprisingly compatible with a metastasis from the pituitary tumor being rich in PRL by immunohistochemistry (Fig. 2). Stained with Ki-67, a sensitive marker of cellular replication [4] selectively expressed during the G1, S1, G2 and M phases of the cell cycle, the metastasis showed a proliferation index of 16%, a value similar to that of other malignant non-pituitary neoplasms. After surgery, PRL levels declined to normal values (2–5 μg/L) on quinagolide (600 μg/day) treatment.

Figure 1
Figure 1:
Figure 1. MRI scan (coronal section) showing the voluminous mass in the left frontal lobe, which proved to be a metastasis of the malignant prolactinoma.
Figure 2
Figure 2:
Figure 2. Hematoxylin-eosin staining (left side) of the left frontal lobe lesion together with the surrounding dura mater. The tumoral cells are small and uniform in shape. There is clear evidence of dural infiltration by the neoplastic cells. The picture is compatible with a metastasis from a pituitary tumor (original magnification × 62). Immunohistochemistry for PRL (right side) of the left frontal lobe tumor. Positive staining for PRL (dark-brown staining of the cells) is present in about 40–50% of the neoplastic cells, thus confirming the diagnosis of a metastasis from a PRL-secreting tumor (original magnification × 275).

Isoelectrofocusing analysis of PRL isoforms [8] performed on sera from the patient during the quinagolide treatment after the appearance of resistance to dopaminergic treatment revealed no difference from those present in a large group of micro-and a few invasive macroprolactinomas before treatment. However, after removal of the metastasis and recovery of PRL response to therapy, the isoforms showed a nearly flat profile, which was comparable to those of normal subjects and hyperprolactinemics with different etiologies on successful therapy (Fig. 3). This finding suggests that the “synthesis” of PRL by the metastasis was similar to that present in micro-and macroprolactinomas used as controls.

Figure 3
Figure 3:
Figure 3. Isoelectrofocusing profiles of PRL isoforms before and after excision of the metastasis in the patient studied (upper inset: isoelectrofocusing profiles of PRL isoforms in micro-and macroprolactinomas off-therapy; lower inset: isoelectrofocusing profiles of PRL isoforms in micro-and macroprolactinomas on-therapy and in healthy subjects).

Prolactin levels continued to rise (38–68 μg/L), however, and expansion of the pituitary mass was seen by MRI. The patient underwent trans-sphenoidal debulking of the neoplasm to enable radiosurgery to be performed. Histology confirmed a pituitary adenoma, and immunohistochemical studies stained positively for PRL in about 50% of the cells (Fig. 4). About 8% of the pituitary tumor cells stained positive for Ki-67. Cytogenetic studies on tumor specimens showed various genetic abnormalities: clonal rearrangement of chromosome 1, such as t (1;3) and i (1q) as well as telomeric association between the short arms of chromosomes 14 and 19: tas (14 p19 p) [9]. After surgery, PRL levels were unchanged and quinagolide was restarted at a dose of 600 μg/day. Three months later, MRI showed that the macroadenoma now compressed the pons.

Figure 4
Figure 4:
Figure 4. Hematoxylin-eosin staining (left side) of the pituitary tumor. The neoplastic cells are arranged in a typical glandular architecture. The picture is compatible with a pituitary adenoma (original magnification × 62). Immunohistochemistry for PRL (right side) is positive (dark-brown staining of the cells) in approximately 30–40% of the neoplastic cells, thus confirming the diagnosis of PRL-secreting pituitary tumor (original magnification × 275).

Stereotactic radiosurgery (gamma knife) was administered. Two months later, after the onset of sudden rhinorrhea, severe headache and bilateral visual loss, the patient was transferred to a neurosurgical division. Coma quickly ensued. MRI showed a hemorrhage into the pituitary tumor. PRL levels were in the normal range (2–4 μg/L). Antibiotics and corticosteroids partially improved the state of consciousness but blindness persisted. The patient’s physical condition stabilized but did not improve. Two months later, MRI showed partial shrinkage of the pituitary neoplasm with signals on T2 compatible with vascular damage to the pons and the cerebral peduncle. Finally, one month later, the patient died of pulmonary embolism. No post-mortem studies were authorized. The profile of PRL levels throughout the clinical history of the patient is shown in Fig. 5.

Figure 5
Figure 5:
Figure 5. Profile of PRL levels throughout the clinical history of the patient. X axis shows the years of disease and each point of the plot represents the mean ±SE of different PRL measurements. TNS = trans-sphenoidal surgery; TC = trans-cranic surgery; RT = radiotherapy; Br = bromocriptine; MTS = metastasis; SR = stereotactic radiosurgery.

Discussion

Prolactinomas are usually larger in men than in women at the time of diagnosis. This finding was previously attributed to a longer delay in diagnosis in males, but a recent literature report [1] has modified this concept. According to this paper, the prevalence of large prolactinomas in men is due to a high frequency of rapidly growing tumors (as assessed by high indexes of proliferating cells by Ki-67 and PCNA), which are often invasive and resistant to dopaminergic treatment. For this reason, the difference in tumor dimension between the sexes is present even in young age.

The current definition of pituitary carcinomas requires the demonstration of metastasis. However, high mitotic index [4] and p-53 immunoreactivity [4,10] suggest the diagnosis and appear to be of prognostic help. Although prolactinomas are the most common functioning pituitary tumors and have the highest recurrence rate after neurosurgery, they are the least likely to show malignant behaviour [3]. When they do metastasize, however, they usually show a higher tendency to systemic spread than the other pituitary carcinomas [4]. PRL levels in carcinomas range from normal to thousands of μg/L [4].

We found 30 malignant prolactinomas described in the literature [2–4,6,7,11–21]. There is a slight prevalence in males (57%) over females (43%). Median age at presentation in women was 48 years. All tumors were invasive macroadenomas at the time of initial diagnosis and there was a long latency period (years) between the first symptoms and tumor recurrence or progression of a residual tumor. The median interval for the occurrence of metastases was 8 years. The median survival after metastasis was only 1 year. The long interval between the diagnosis and the occurrence of metastasis in most cases (including ours), might support the hypothesis of an adenoma-to-carcinoma transformation. For the time being, cytogenetic characteristics of the tumor do not allow a clear-cut correlation between the karyotype and clinical findings [9]. An overview of the literature on the features of malignant prolactinoma in women is shown in Table 1.

Table 1
Table 1:
Table 1 Malignant prolactinomas in female patients: overview of the literature.C = craniotomy; S = transsphenoidal surgery; R = radiotherapy; D = dopamine agonists; CH = chemotherapy; L = laminectomy; SH = shunt; SS = stereotactic surgery; AWMD = alive with metastatic disease.*in brackets: time span between primary tumor and metastases

In this paper, we describe the dramatic story of a young women with an invasive macroprolactinoma which metastasized several years after diagnosis. The initial response to dopaminergic treatment was satisfactory, but subsequent loss of sensitivity to bromocriptine was associated with the appearance of a frontal metastatic mass. This fact may be viewed as the first sign of real dedifferentiation of the neoplasm, with consequent modifications of D2 membrane receptors on the metastasis and/or post-receptor signalling. This was confirmed by the reappearance of the response to dopaminergic treatment after excision of the metastasis. Despite the high PRL levels, which were unresponsive to dopaminergic treatment, isoelectrofocusing studies did not reveal any qualitative difference in PRL isoforms secreted by the metastasis.

Loss of sensitivity to dopaminergic drugs may be the first signal of malignant transformation in a macroprolactinoma. In agreement with a previous report in the literature [12], the metastasis produced intact PRL and was able to secrete the hormone evidenced by the sudden increase in serum PRL levels. The malignant transformation of the tumor was confirmed by the high proliferation index of the metastasis, as assessed by Ki-67 staining. In our case, a possible role of the previous (i.e., 9 years before) radiotherapy on the malignant transformation of the macroadenoma over time cannot be excluded. This is also suggested in the literature [19].

The treatment of malignant prolactinoma involves combined therapy such as surgery, external beam therapy [5], bromocriptine [6,16], quinagolide [17], tamoxifen, octreotide and chemotherapy [7,13,22]. Combined therapy, however, is not usually curative. Recent efforts to administer minimally toxic chemotherapy agents (lomustine/5-fluorouracil) have shown some benefit in patients with malignant carcinomas, at least in achieving temporary remission or delay of progression [7].

In our case, the various approaches (surgery, different dopamine agonists, radiotherapy and stereotactic radiosurgery) used over the years resulted in delaying but not arresting neoplastic growth, which eventually proved fatal. Malignant prolactinomas are a rare but important entity whose prognosis, despite different approaches is very unfavorable.

Acknowledgment

This work was in part supported by grant CNR 98.00462.CT 04 to Prof. M. Giusti.

References

1. Delgrange E, Trouillas J, Maiter D, Donckier J, Tourniare J: Sex-related difference in the growth of prolactinomas: a clinical and proliferation marker study. J Clin Endocrinol Metab 1997; 82: 2102–17.
2. Doniach I: Pituitary carcinomas. In Clinical Endocrine Oncology, edited by Blackwell, pp. 225–7, Oxford, Science, 1997.
3. Popadic A, Witzman A, Buchfelder M, Eiter H, Komminoth P: Malignant prolactinoma: case report and review of the literature. Surg Neurol 1999; 51: 47–55.
4. Pernicone PJ, Scheithnauer BW, Sebo TJ, et al.: Pituitary carcinoma: a clinicopathologic study of 15 cases. Cancer 1997; 79: 804–12.
5. Grossman A, Cohen BL, Charlesworth M, et al.: The treatment of prolactinomas with megavoltage radiotherapy. Br Med J 1984; 288: 1105–9.
6. Popovic EA, Vattuone JR, Siu KH, Busmanis I, Pullar MJ, Dowling J: Malignant prolactinomas. Neurosurgery 1991; 29: 127–30.
7. Kaltas GA, Mukherjee JJ, Plowman PN, Monson JP, Grossman AB, Besser GM: The role of cytotoxic chemotherapy in the management of aggressive and malignant pituitary tumors. J Clin Endocrinol Metab 1998; 83: 4233–8.
8. Guido R, Valenti S, Foppiani L, De Martini D, Cossu M, Giusti M: Prolactin decrease and shift to a normal-like isoform profile during treatment with quinagolide in a patient affected by an invasive prolactinoma. J Endocrinol Invest 1997; 20: 289–93.
9. Bettio D, Rizzi N, Giardino D, et al.: Cytogenetic study of pituitary adenomas. Cancer Genet Cytogen 1997; 96: 1–6.
10. Thapar K, Scheithauer BW, Kovacs K, Pernicone PJ, Laws ER: P 53 expression in pituitary adenomas and carcinomas: correlation with invasiveness and tumor growth fractions. Neurosurgery 1996; 38: 763–70.
11. Martin NA, Hales M, Wilson CE: Cerebellar metastasis from a prolactinoma during treatment with bromocriptine. Journal of Neurosurgery 1981; 55: 615–9.
12. Cohen DL, Diengdoh JV, Thomas DGT, Himsworth RL: An intracranial metastasis from a PRL secreting pituitary tumor. Clin Endocrinol 1983; 18: 259–64.
13. Landgraf R, Rieder G, Schmiedek P, Clads D, Bise K, von Werder K: Hormone-active intradural spinal metastasis of a prolactinoma: a case report. Klin Wochensch 1985; 63: 379–84.
14. Plangger CA, Twerdy K, Grunert V, Weiser G: Subarachnoid metastases from a prolactinoma. Neurochirurgie 1985; 28: 235–7.
15. Scheithauer BW, Randall RV, Laws ER, Kovacs KT, Horvath E, Whitaker MD: Prolactin cell carcinoma of the pituitary. Clinicopathological immunohistochemical and ultrastructural study of a case with carnial and extracranial metastasis. Cancer 1985; 55: 598–604.
16. Muhr C, Bergstrom M, Lundberg PO, et al.: Malignant prolactinoma with multiple cranial metastases studied with positron emission tomography. Neurosurgery 1988; 22:374–9.
17. Davis JRE, Sheppard MC, Heath DA: Giant invasive prolactinoma: a case report and review of nine further cases. Q J Med 1990; 74:227–38.
18. Berezin M, Gutman I, Tadmor R, Horowitz A, Findler G: Malignant prolactinoma. Acta Endocrinol 1992; 127: 476–80.
19. Rainov NG, Burkert W: Malignant growth of a recurrent macroprolactinoma after radiation therapy. Neurochirurgia 1993; 36: 172–5.
20. Alexander MJ, Desalles AA, Tomiyasu U: Multiple radiation-induced intracranial lesions after treatment for pituitary adenoma. Case report. J Neurosurg 1998; 88: 111–5.
21. Walker JD, Grossman A, Anderson JV, et al.: Malignant prolactinoma with extracranial metastases: a report of three cases. Clin Endocrinol 1993; 38: 411–9.
22. Bayindir C, Balak N, Gazioglu N: Prolactin-secreting carcinoma of the pituitary: clinicopathological and immunohistochemical study of a case with intracranial and intraspinal dissemination. Br J Neurosurg 1997; 11: 350–5.
Hurel SJ, Harris PE, Mcnicol AM, Foster S, Kelly WF, Baylis PH. Metastatic prolactinoma: effect of octreotide, cabergoline, carboplatin and etoposide; immunocytochemical analysis of proto-oncogene expression. J Clin Endocrinol Metab 1997; 82: 2962–5.
© 2001 Lippincott Williams & Wilkins, Inc.