Giant prolactinomas correspond to 0.5–4.4% of all pituitary tumors, based on retrospective analysis1,2,3,4,5. They represent 2–3% of all prolactin-secreting tumors and are more prevalent in men with a male to female ratio of 9:1 and a mean age at diagnosis of 40 years. The linear relationship between prolactin levels and tumor size can be lost in this group of patients, either because of the hook effect or due to the presence of necrotic or hemorrhagic components in the tumor6. Hook effect is caused by an excess of soluble antigen binding separately to both the solid-phase and the soluble antibody in one-step immunometric assays. The ‘sandwich’ in which the antigen molecule binds simultaneously to both antibodies is not formed and during the washout phase these antigen-soluble antibody complexes are removed therefore resulting in a lower than expected reading. The hook effect can be eliminated by using a two-step process or more usually by performing analysis at two sample dilutions7. Criteria for the definition of a giant prolactinoma vary among authors. Recently Maiter and Delgrange6 suggested that these tumors should be defined on the basis of the following: 1) Size – largest diameter of 4 cm or more in any direction with massive extrasellar extension; 2) prolactin (PRL) levels at diagnosis usually above or equal to 1,000 µg/L using a modern and well-standardized assay; 3) Exclusion of concomitant GH and ACTH secretion. The therapeutic goals are the same as for the non-giant prolactinomas, however rapid relief of neurologic symptoms is usually critical. There are no specific evidence-based recommendations for the management of these tumors and most of the literature consists of case reports or small case series.
The purpose of this case series is to report eight patients with giant prolactinomas followed at the Endocrinology Clinics of the Hospital Universitário Antônio Pedro, adding information of such a rare condition to the literature.
METHODS
A review of the medical records of patients with prolactinomas from the endocrinology outpatient clinics was performed with special interest to the patients with tumors with maximum diameter of more than 4 cm. The study was approved by the local Ethics Committee.
Eight patients (six men; male/female ratio 3:1) were included in the analysis.
At diagnosis median age was 29 years (18–54) and median PRL concentration was 4,562 ng/mL (1,543–18,690 ng/mL; Reference values: male 2.5–17; female 1.9–25). All patients had hypogonadotropic hypogonadism and additional pituitary hormone deficiencies were observed in three patients (#4, #7, and #8), the latter two being affected in all adenohypophyseal axis.
RESULTS
Median maximum tumor diameter was 50 mm (41–60). Five patients (62.5%) had visual field defects (detailed in Table 1). Regarding the presenting features, the two female patients (#7 and #8) had the classic amenorrhea/galactorrhea syndrome and in one patient (#8), additionally, significant visual symptoms were also noted. One patient (#3) sought medical care in the context of an erectile dysfunction and laboratory tests confirmed hyperprolactinemia. Three patients (#2, #4 and #5) had symptoms predominantly related to the tumor mass effect with headache and visual field defects (suprasellar extension). Patient #2 also manifested left oculomotor nerve paresis (cavernous sinus extension). One patient (#6) had a very rare presentation due to large anterior tumoral extension. Giant prolactinoma was discovered in the context of the investigation of chronic nasal obstruction unresponsive to usual treatment (“incidental” finding of paranasal sinus computed tomography). Finally, one patient (#1) presented as an incidentaloma in imaging studies ordered in the context of an acute ischemic stroke.
Table 1 Baseline characteristics.
Patient # | Sex | Age at diagnosis | PRL levels at diagnosis (ng/mL) | Initial symptoms | Hormone deficiencies | Visual field defects | Tumor size (mm) |
---|---|---|---|---|---|---|---|
1 | M | 50 | 18,690 | Incidental / acute ischemic stroke | Gonadotrophins | Right eye amaurosis + left eye temporal hemianopia | 55 |
2 | M | 18 | 1,543 | Headaches + visual field defects + left ptosis (oculomotor nerve compression) | Gonadotrophins | Left eye amaurosis +III cranial nerve paresis | 60 |
3 | M | 54 | 5,746 | Erectile dysfunction | Gonadotrophins | No | 50 |
4 | M | 27 | 3,404 | Headaches + visual field defects | Gonadotropic and thyrotropic axis | Bitemporal hemianopia | 50 |
5 | M | 30 | 13,039 | Headaches + visual field defects | Gonadotrophins | Bitemporal hemianopia | 59 |
6 | M | 50 | 4,128 | Incidental / chronic nasal stuffiness | Gonadotrophins | No | 41 |
7 | F | 23 | 4,996 | Amenorrhea/galactorrhea syndrome | All axis | No | 41 |
8 | F | 28 | 2,000 | Amenorrhea/galactorrhea syndrome + visual field defects | All axis | Left eye amaurosis + Right eye temporal hemianopia | 41 |
F: female; M: male; PRL: prolactin.
Data are summarized in Table 1.
Treatment and follow up
All patients but one (#4) were initially treated with cabergoline (CAB). Patient #4 underwent transfrontal surgery at diagnosis, in 1998, in another center, for unclear reasons.
During follow up, six patients treated with CAB achieved normal PRL levels after a median time of 10.5 months (7–84), with a median dose of 2.0 mg/week (1.0–3.0 mg/week). Patients #3 and #4 only reached normalization of PRL levels after 49 and 84 months, respectively, due to noncompliance with treatment. Two patients did not normalize PRL levels (#2 and #5). Nevertheless patient #2 had 93% reduction in serum PRL levels, with the last imaging study showing a partial empty sella. Patient #5 developed a cerebrospinal fluid (CSF) leak diagnosed on the basis of a clear rhinorrhea which started two months after CAB was initiated (1.5 mg/week). As he refused surgical correction of the CSF leak, CAB dose was reduced (0.5 mg/week) to allow tumor re-growth as an attempt to stop CSF leak, however it was unsuccessful and after two months CAB dose was progressively increased (3.0 mg/week) to attain PRL level reduction/normalization and tumor shrinkage. Patient was constantly told to keep the head of his bed elevated, among other conservative measures, as well as repeatedly informed of the risks and gold standard surgical treatment for his condition. During follow up, CSF leak resolved spontaneously after seven months of continued CAB treatment with no intercurrent infection.
Of the five patients who had visual field defects, three showed improvement during treatment. Only one patient (#3) recovered gonadal function. No other patients with hypopituitarism at diagnosis experienced recovery of the pituitary function. At the final evaluation, median maximum tumor diameter was 22 mm (patients #2 and #4 showed partial empty sella and empty sella, respectively). All patients showed significant tumor shrinkage (reduction ≥ 25% in tumor volume and/or ≥ 30% in tumor diameter) (Figures 1 and 2).

Figure 1 (A) Patient #1. Sella turcica MRI coronal view T1WI, showing a large tumor extending from the sella to supra, infra and parasellar regions, with heterogeneous contrast enhancement, measuring 4.8 x 4.0 x 5.5 cm. (B) After two years of treatment with cabergoline 3.0 mg/week, residual lesion, with cystic areas, irregular enhancement of contrast, measuring 1.0 x 2.2 x 2.4 cm.

Figure 2 (A) Patient #2. Sella turcica MRI coronal view T1WI, showing an expansive sellar tumor measuring 6.0 x 5.2 x 5.3 cm, extending to the suprasellar region and to the left temporal lobe and periventricular region, with massive left cavernous sinus invasion. In the superior portion a hyperintensity area suggests acute bleeding inside the tumor. (B) Partial empty sella (after seven years of treatment with cabergoline 3.5 mg/week).
Data are summarized in Table 2.
Table 2 : Treatment and follow up.
Patient # | Initial treatment | CAB dose (mg/week) | PRL level (ng/mL) / % reduction | Time for PRL normalization (months) | Visual field defects | % reduction in tumor size |
---|---|---|---|---|---|---|
1 | DA | 3.0 | 6.81 / 99.9 | 7 | Bitemporal hemianopia (improvement in right eye) | 95 |
2 | DA | 3.5 | 110 / 93 | - | Left eye hemianopia (improvement) +III cranial nerve paresis | Partial empty sella |
3 | DA | 1.0 | 17 / 99.7 | 49 | No | 54 |
4 | Surgery | 2.0 | 15 / 99.6 | 84 | Bitemporal hemianopia | Empty sella |
5 | DA | 3.0 | 157 / 98.8 | - | None (bilateral improvement) | 45 |
6 | DA | 2.0 | 4.86 / 99.8 | 10 | No | 67 |
7 | DA | 2.5 | 8.6 / 99.8 | 7 | No | 49 |
8 | DA | 2.0 | 3.61 / 99.8 | 11 | Left eye amaurosis + Right eye temporal hemianopia | 76 |
CAB: cabergoline; DA: dopamine agonist; PRL: prolactin.
DISCUSSION
Giant prolactinomas show presenting features different from non-giant tumors, generally reflecting massive tumor extension into surrounding structures6.
In the present case series, two patients (#1 and #6) had presentations that are worthy to note. In one patient, giant prolactinoma was an incidental finding of a head MRI ordered in the context of an acute ischemic stroke (#1). In the other (#6), giant prolactinoma was discovered during the investigation of chronic nasal obstruction which was due to large anterior tumor extension into the nasopharynx.
Patient (#1) is a 50-year-old man who was admitted to the emergency department with a right hemiplegia. In addition to an acute ischemic stroke in topography in the semiovale centrum, corona radiata, caudate nucleus, internal capsule and insula on the left, head MRI showed an expansive sellar lesion with upper extension to the suprasellar cistern and below to the sphenoid sinus as well as a lateral extension, especially to the right, involving the cavernous portion of the right carotid artery, measuring 4.8 x 4.0 x 5.5 cm. Patient reached normalization of PRL levels and significant reduction of tumor diameter with CAB (data are summarized in Tables 1 and 2). A few cases of stroke related to giant pituitary tumours have been described in the literature. However, they are often associated with pituitary apoplexy (PA)8. Multiple hypotheses have been proposed to explain a stroke in PA, but the most accepted is the extraluminal compression of the main brain vessels caused by the acute growth of the tumor. This seems not to be the case of the present patient since no clinical or radiological evidence of PA were present at the time of acute ischemic stroke.
In patient #6, as previously stated, giant prolactinoma was discovered during the investigation of chronic nasal obstruction which was due to large anterior tumor extension into the nasopharynx. There are few cases in the literature reporting this type of presentation. In one of them9 a 14-year-old boy presented with progressive weight gain, nasal obstruction, snoring, and a polypoid mass in his left nostril. Imaging studies revealed a homogenously enhancing mass extending from the floor of sella to the nasopharynx, measuring in the maximum diameter 9.9 cm. Only after biopsy and laboratory analysis the diagnosis of giant prolactinoma was done and successful treatment with CAB was instituted. In another case10 a 30-year-old man presented with recurrent sinusitis not relieved by several courses of antibiotics. Imaging studies revealed an extensive skull base mass with a nasopharyngeal component which suggested a nasopharyngeal carcinoma. Patient was submitted to a biopsy and the histological pattern suggested a low grade neuroendocrine tumour. Only after laboratory tests and specific staining techniques, the diagnosis of a giant invasive prolactinoma was made. As previously seen, sometimes the diagnosis of prolactinoma may be made through the biopsy of a nasal polypoid mass. These cases should remind the clinician that giant pituitary tumors, and notably giant prolactinomas, may extend far from the sella and cause symptoms different from cranial nerve compression6.
Regarding the follow up, it is important to mention that patient #5 developed a CSF leak, a possible complication of dopamine agonist (DA) treatment, after two months of starting CAB (1.5 mg/week). There have been several reports of development of CSF rhinorrhea following the initiation of medical treatment for pituitary adenomas. The classical description of this phenomenon has been in the setting of DA therapy for prolactinomas11. Lam et al.12 reviewed the literature between the years 1980 and 2011, describing 52 patients with spontaneous or medically induced CSF leaks in the setting of pituitary adenomas. In 38 patients (73%), CSF rhinorrhea developed following initiation of medical therapy. Forty-two patients (81%) had prolactinomas and the medical agents associated with CSF leakage were DA (97%) and somatostatin analogs (3%; one patient). The average time from initiation of medical treatment to onset of rhinorrhea was 3.3 months (range 3 days–17 months).
Another complication from medical treatment of prolactinomas is empty sella which occurs quite frequently and may cause or aggravate pre-existent visual disturbances. In the present series, patients #2 and #4 had respectively partial and total empty sella, both asymptomatic and not associated with worsened visual field defects. Secondary empty sella usually develops following radiation therapy or surgery of the sellar region or during dopamine agonist treatment for prolactinomas13,14,15. In that scenario, visual disturbance is caused by traction of the chiasm or the optic nerves toward the sella floor by the adhesion scar or pituitary stalk16,17,18. Gradual dose reduction of the dopamine agonist and systematic visual field evaluation may be attempted as a medical treatment to recover visual field defects19. Alternatively, surgical treatment should consist of chiasmapexy or sellar packing2.
In conclusion, the giant prolactinoma is a rare subtype of macroprolactinoma with diagnostic criteria still not clearly defined. It is proposed that the most important criterion is the maximum tumor diameter greater than 4 cm. Regarding the presenting features, this series reports the first case of a giant prolactinoma presenting as a true incidentaloma, to the best of our knowledge. Also it reminds the importance of considering unusual presentations resulting from massive extrasellar tumor extension. Finally, pharmacological treatment with dopamine agonists is an effective anti-secretory and anti-proliferative option and its use may result in massive tumor shrinkage which may complicate with CSF leak.