Correspondence: Mônica R. Gadelha, Rua Prof. Rodolpho Paulo Rocco, 255, 9th floor, Ilha do Fundão, Rio de Janeiro 21941-913, Brazil. Tel./Fax: 55-21-2562-2111; E-mail: firstname.lastname@example.org
Giant prolactinomas are an unusual subset of macroprolactinomas and are more commonly found in men. The goal of this review is to propose a giant prolactinoma definition and discuss the available therapeutic options for biochemical and tumour volume control.
A comprehensive search of all published studies was performed between April and November 2012 in electronic databases (PubMed and Ovid).
A giant prolactinoma should be defined as an adenoma with a maximum diameter of more than 4 cm that is associated with serum prolactin above 5300 mIU/l. Regarding treatment, cabergoline is the preferred dopamine agonist for medical management of giant prolactinomas because of its excellent efficacy and tolerability. Normalization of prolactin level and significant tumour reduction may be achieved in the majority of patients. Combined therapy, particularly cabergoline and surgery, may be necessary due to the large tumour load. Radiotherapy and temozolomide may be used for patients with aggressive giant prolactinomas in whom tumour volume control is not achieved with cabergoline and surgery.
There is a scarcity of large studies about the management of giant prolactinoma. Cabergoline is the first-line treatment. However, caution should be exercised when comparing efficacy rates among the different treatment modalities due to the variability in study design and data quality. In this scenario, a ‘standard’ definition for giant prolactinomas and larger series may be helpful to assess the real efficacy and safety of each therapeutic modality.
Prolactinomas are the most common pituitary tumours in adults, accounting for up to 45% of pituitary tumours. Microprolactinomas are more common in women, whereas macroprolactinomas are more frequent in men. Giant prolactinomas are an unusual subset of macroprolactinomas and are more commonly found in men.[2, 3]
Patients harbouring a giant prolactinoma typically respond to dopamine agonist (DA) therapy, despite the large tumour size and invasion of adjacent structures.[4, 5] However, due to the large tumour volume, multimodal treatment may be necessary to normalize the serum prolactin (PRL) level and control tumour volume. In addition, a subgroup of giant prolactinomas exhibits aggressive clinical behaviour, and biochemical and tumour volume control in this subgroup is a significant challenge.
We (A.B.M. and C.M.S.S.) conducted a literature review from April 2012 to November 2012 with the intention of including studies that were most applicable to the clinical scenario of patients with giant prolactinomas. Electronic database (PubMed and Ovid) searches of all years of publication were performed using the following search string: giant prolactinoma OR prolactinoma + dopamine agonist OR prolactinoma + surgery OR prolactinoma + hook effect OR giant invasive prolactinoma OR large prolactinoma. Only articles written in English were included. Original studies, case reports, reviews and current guidelines were included in this review. All PRL units expressed in ng/ml were converted to mIU/l by multiplying by 21·2.
In this review, we propose a definition of giant prolactinoma and discuss the epidemiological, clinical, pathological and therapeutic aspects of giant prolactinomas, with particular focus on the available multimodal treatments for biochemical and tumour volume control.
Definition of a giant prolactinoma and diagnostic pitfalls
There is no consensus on the definition of a giant prolactinoma.[2, 4, 6-11] Therefore, we propose that giant prolactinoma should be defined as an adenoma with a maximum diameter of more than 4 cm (an arbitrary size) that is associated with serum PRL above 5300 mIU/l.[2, 11] Macroprolactinomas are characteristically associated with serum PRL > 5300 mIU/l, according to the Endocrine Society's recent guidelines.
In most patients, there is an association between serum PRL level and tumour volume. Therefore, patients with giant prolactinomas often have very high serum PRL. However, this association is not always consistent, and tumour mass and PRL level may be dissociated. The hook effect is a possible cause for this discrepancy and has been reported to occur in 20% of giant prolactinomas. The hook effect, an assay artefact, occurs when excessively elevated serum PRL interferes with antibody–antigen–antibody sandwich complex formation. The final effect is lower signal detection by the assay than expected, causing erroneously low serum PRL to be detected. To overcome this flaw, the assessment of PRL should be repeated on a 1:100 dilution of the serum sample; if the pituitary tumour is a prolactinoma, PRL will be as high as expected (proportional to the tumour size); however, if the sellar lesion is not a prolactinoma but a clinically nonfunctioning pituitary adenoma, for example, PRL will remain at a similar level after the dilution. Failure to recognize this artefact has led to misdiagnoses of giant prolactinomas or large macroprolactinomas as nonfunctioning pituitary adenomas with normal PRL level or with mild/moderate hyperprolactinaemia due to stalk disruption and has thus also led to unnecessary neurosurgical intervention. Thus, in giant pituitary adenomas with seemingly normal or mildly/moderately elevated PRL, the assay should be repeated on a 1:100 serum sample dilution to circumvent a potential hook effect and thus prevent incorrect diagnosis of nonfunctioning pituitary adenomas.
Cystic giant prolactinomas are another potential cause of the discrepancy between PRL level and tumour volume because although the tumour volume is large, the number of viable lactotroph cells is not excessively high due to cystic degeneration.
Epidemiological, clinical and pathological features
The prevalence of giant prolactinomas is not well established. There are no data regarding the prevalence of giant prolactinomas among prolactinomas, but based on retrospective analyses, it may be estimated to range from 0·5 to 4·4% of all pituitary tumours.[2, 4, 15-17] The majority of patients included in the largest published series of giant prolactinomas are men, with a male/female ratio of 6·5/1. In addition, giant prolactinomas in men are usually more aggressive than in women. Among series with giant prolactinomas, the median age at diagnosis is 42 years (24–59).[2, 5, 6, 9]
In general, giant prolactinomas cause clinical symptoms as a result of mass effect and/or hyperprolactinaemia. These tumours often cause visual field defects and/or ophthalmoplegia due to compression of the optic chiasm and/or cranial nerves, respectively, as well as headaches. Although the most common site of extrasellar extension is into the suprasellar cistern, large tumours can also extend inferiorly into the sphenoid sinus or laterally into the cavernous sinuses. In rare instances, the temporal or frontal lobes are invaded, causing seizures and personality disorders. Other rare presentations include invasion of the skull base, mimicking primary bone dysplasia, as well as invasion of the nasopharynx, which may cause epistaxis.[18, 19] Moreover, the development of intracranial hypertension as a consequence of obstructive hydrocephalus can rarely occur. Orbital invasion by a pituitary adenoma is the least common form of extrasellar extension. Because these patients may have hypopituitarism, it is important to evaluate their pituitary function. Hyperprolactinaemia typically presents with symptoms and signs such as decreased libido, impotence, infertility, galactorrhoea, oligomenorrhoea or amenorrhoea and gynaecomastia.
Several criteria have been described to define pituitary adenomas as aggressive/invasive. Aggressiveness is defined by the clinical characteristics of the tumour. Giant prolactinomas may have clinically aggressive behaviour, including resistance to DA therapy and high recurrence rates after surgery. The criteria used to define invasiveness are radiological, surgical and pathological. Giant prolactinomas are typically invasive. For pituitary adenomas, microscopic invasion of the dura mater is frequent, but it has not been associated with tumour recurrence or aggressiveness. On the other hand, macroscopic invasion of the perisellar tissues (sphenoidal or cavernous sinuses), defined as a radiographic or gross surgical finding, is considered a more consistent prognostic factor. So far, there is no agreement regarding the best radiological classification of tumour invasiveness, and some classifications have been proposed, such as Knosp-Steiner, SIPAP and Hardy.[26-28] The classification of Hardy modified by Bates is the most commonly used criterion.[25, 26]
In 2004, the World Health Organization (WHO) classified endocrine tumours as typical or atypical adenomas in an effort to identify tumours with potentially aggressive clinical behaviour. The atypical pituitary adenoma was defined as an invasive tumour characterized by a Ki-67 labelling index > 3%, extensive p53 positivity and an elevated mitotic index. To date, according to WHO, only prolactinomas with distant metastases are termed carcinomas, which are very rare. However, Zemmoura et al. suggest that the association of certain clinical signs (male sex, dopamine-resistant hyperprolactinaemia), radiological characteristics (invasive macro or giant tumour on MRI) and histological features (Ki-67 > 3%, p53 positive, mitoses > 2 per high-power field, angiogenesis, vascular invasion, up-regulation of genes related to invasion and proliferation, and allelic loss of chromosome 11) might indicate aggressiveness and be suspicious of malignancy before the appearance of metastasis. Thus, prompt detection of an aggressive phenotype of a prolactinoma should allow earlier establishment of the best therapeutic approach associating surgery and radiation therapy to delay or prevent metastasis.
Several markers of proliferation have been investigated in invasive prolactinomas. To date, there is no robust marker to identify pituitary adenomas with the potential to grow, invade or metastasize. However, in clinical practice, the proliferation marker used most frequently is Ki-67. A high Ki-67 labelling index (>3%) may suggest an aggressive tumour, whereas tumours with low Ki-67 (<3%) may or may not exhibit aggressive behaviour.
Therapeutic management of giant prolactinomas
Different therapeutic approaches, such as DA alone or in combination with surgery (or early surgery, as indicated for patients exhibiting apoplexy with severe clinical symptoms or intracranial hypertension) and radiotherapy, may be necessary to reach the therapeutic goals in patients with giant prolactinomas, which include control of tumour volume, normalization of PRL level and restoration of eugonadism. Temozolomide (TMZ) has been used in aggressive prolactinomas that are unresponsive to the above-mentioned therapies. It is important to emphasize that, due to large tumour volume and commonly invasive behaviour, control of mass effect should be a priority in the management of patients with giant prolactinomas.
Medical therapy with DA is the first-line treatment for giant prolactinomas, even in patients with visual field defects,[5, 12] because DAs are able to induce tumour shrinkage within a few days after initiating therapy. However, close monitoring of patients with visual field disturbances is advisable. Table 1 summarizes the series that evaluated the impact of DAs on tumour shrinkage. It is important to mention that caution should be used when comparing efficacy rates among the different DAs due to variability in study design and data quality.
Table 1. Impact of dopamine agonists on tumour shrinkage
Number of patients
Diagnostic criteria for giant prolactinomas
Mean dose of DA (range)
Mean duration of treatment (months)
At least 50% tumour reduction (% of patients)
Visual field improvement
BRC, Bromocriptine (mg/day); CAB, cabergoline (mg/week); DA, dopamine agonist; NR, not reported; PRL, prolactin; *with suprasellar extension; †high aggressiveness; ‡invasion of the cavernous sinus to an extent corresponding to grade III or IV in the classification scheme of Knosp; §atypical symptoms (nasal obstruction, hypoacusia, psychiatric symptoms, retro-orbital pain); ¶in the nonsurgery group, four patients had visual field improvement (data in the surgery group are not available); **diagnostic criteria for giant prolactinomas were one or more of major diameter > 4 cm, PRL > 63 600 mIU/l and markedly invasive growth; ††mass effect and/or clinical signs may or may not be present for diagnosis of giant prolactinoma. ‡‡the average percentage of tumour reduction was not calculated because the series are heterogeneous regarding the imaging methods (computed tomography or magnetic resonance imaging) and tumour size parameters used (volume, larger diameter, etc.), making the series not comparable.
Cabergoline (CAB) is the preferred DA for medical management of giant prolactinomas because of its excellent efficacy and tolerability.[5, 9, 36] Data on the efficacy of CAB have been reported in five series of giant prolactinomas, with mean doses that varied from 1·4 mg/week to 4·5 mg/week.[4, 5, 9, 37, 38] In 30 patients with giant prolactinomas who were primarily treated with CAB, 19 (63%) achieved normal PRL levels.[4, 5, 9, 37, 38]
According to the series of Corsello et al., persistent control of PRL level in patients primarily treated with CAB was attained in four of six patients. In the remaining two patients, PRL values decreased by more than 97%. In the Cho et al. series, among 10 patients, five did not exhibit normal PRL concentrations with CAB as primary therapy. However, all patients displayed PRL reductions of at least 89% during CAB therapy. However, higher rates of PRL normalization (80%) were observed by Shimon et al. among patients primarily treated with CAB.
Significant tumour shrinkage is typically achieved with CAB therapy. In five series (38 patients) with giant prolactinomas who were treated with CAB, 26 patients (68%) experienced at least a 50% reduction in tumour size (Table 1).[4, 5, 9, 37, 38] Corsello et al. found that the tumour volume decreased in nine of 10 patients with giant prolactinomas. The tumour volume reduction was > 95% in three patients, between 50 and 95% in four, and between 30 and 50% in two. One patient did not achieve tumour volume reduction after 12 months of CAB therapy. In another study, tumour shrinkage was observed (median 87%; 57–98) in all 10 patients with giant prolactinomas after 6 months of CAB treatment. Long-term CAB treatment (>12 months) was more efficient regarding tumour size reduction compared to treatment for < 12 months.
Most patients with giant prolactinomas have visual field improvement with CAB therapy. One study evaluated the effect of CAB on visual fields in 12 patients, of whom nine had visual field defects at presentation: the visual field normalized in three patients (within 1–3 months) considerably improved in five patients and did not change in one patient.
Corsello et al. found complete normalization of sexual potency and libido in two of eight patients, and this normalization was associated with control of PRL and normalization of testosterone levels. The remaining six patients had clinical improvement of both symptoms; however, they did not achieve normalization of testosterone, and three of the patients did not attain control of PRL. Infertility was present in one patient, and the sperm sample showed an improvement in number and motility after normalization of PRL. The severity of headaches decreased in all five patients with this complaint. Only one patient presented with galactorrhoea, which disappeared during treatment.
To the best of our knowledge, there are no data in the literature about CAB withdrawal in patients with giant prolactinomas. Two studies that evaluated withdrawal of CAB in patients with macroprolactinoma in general showed rates of 11% and 47% of persistent normoprolactinaemia.[39, 40]
Bromocriptine (BRC) is not currently used as frequently as CAB, mainly due to the higher incidence of drug intolerance. The efficacy of BRC in giant prolactinomas, with a mean dose of 10–40 mg/day, varies.[2, 6-8, 37, 41, 42] In 45 patients (five series) with giant prolactinomas who were treated with BRC as primary therapy, 15 (33%) achieved normal PRL levels.[2, 7, 8, 41, 42]
Twelve patients treated primarily with BRC achieved a mean reduction in tumour volume of 91%. In another series, all 10 patients showed tumour shrinkage, with a mean tumour volume reduction of 69% (52–82%). Yu et al. observed visual field improvements in 50% of the patients, while no changes were observed in 42%, and worsening of the visual field was observed in 8%. In contrast, all of the patients with visual field defects reported improvement in the series of Shrivastava et al.
Complaints regarding gonadal function improved in 80% of patients according to Shrivastava et al. Wu et al. found that headaches disappeared in 40% of patients and improved in 10%. Sexual function improved to different degrees in nine patients with decreases in PRL level, and menstruation returned in one of four patients with amenorrhoea.
In patients with giant prolactinomas, Saeki et al. showed that PRL levels were maintained within the normal range after BRC withdrawal in only 1 of 10 patients, whereas Sieck et al. observed normal PRL levels in 2 of 24 patients (one following radiotherapy). Orrego et al. reported a patient with a giant prolactinoma who had a good response to BRC (tumour volume shrinkage of 53% within a month) but rapid re-expansion to its original dimensions 1 week after discontinuation of BRC.
Quinagolide is available in some countries for the treatment of prolactinomas. Currently, few studies with this drug have been reported in patients with hyperprolactinaemia or prolactinoma. Quinagolide has not been studied in giant tumours.[45, 46]
Side effects of dopamine agonists
Cabergoline is associated with fewer side effects than BRC. These side effects tend to occur at the beginning of treatment and with dose increases but can be minimized by introducing the drug at a low dose and increasing the dosage gradually. Its most common side effects are postural hypotension, which may result in dizziness and syncope, headaches, drowsiness, nausea, vomiting, constipation, dry mouth, dyspepsia and symptoms suggestive of reflux esophagitis. Other less frequent side effects include leg cramps, flushing and nasal congestion. In rare instances, psychiatric symptoms, dyskinesia, paraesthesia, nightmare, blurred vision, diplopia and a pleuropulmonary inflammatory-fibrotic syndrome can occur. Gambling and similar obsessive behaviours have been described with high-dose, long-term DA use.
Several studies have focused on the association between the use of DA and valvular heart disease.[49, 50] Unlike Parkinson's disease, where the use of these therapies has been associated with clinically significant valvular regurgitation, the treatment of prolactinoma with DAs has not been associated with clinically significant valvular heart disease in recent series.[49, 50] However, during long-term treatment with DAs, subclinical lesions may occur, such as aortic calcifications and mild or moderate tricuspid regurgitation.[49, 50] All patients eligible for CAB therapy should undergo cardiovascular evaluation, including an echocardiogram, to assess the potential presence of asymptomatic valvular disease. If fibrotic valvular disease is detected, the patient should not be treated with CAB. During CAB therapy, the first echocardiogram should be performed within 3–6 months after treatment initiation and, thereafter, the frequency of echocardiographic monitoring should be guided by the signs and symptoms of cardiac disease but should occur at least every 6–12 months (for more details, see http://www.mhra.gov.uk/home/groups/l-unit1/documents/websiteresources/con014958.pdf).
Complications during treatment with dopamine agonists
Some relevant complications during medical treatment with DAs may be observed, and clinicians should be aware of this possibility. Possible complications include herniation of the frontal lobe and optic chiasm into the pituitary sella, pneumocephalus and cerebrospinal fluid (CSF) leaks.[8, 51, 52] The latter may increase the risk of meningitis. Skull base bone erosion caused by prolactinomas may be unmasked by tumour shrinkage, resulting in CSF leaks and pneumocephalus. Although previous reports have described DA therapy as a precipitating factor for apoplexy, Moller-Goede et al. did not find an increased risk of apoplexy in patients undergoing DA treatment.
Resistance to dopamine agonists
Dopamine agonist resistance has been defined as failure to normalize PRL on maximally tolerated doses of DA and the absence of tumour size reduction ≥ 50%. In series with giant prolactinomas, resistance to CAB varies between 16% and 50%.[4, 5, 38] In patients with giant prolactinomas using BRC, the resistance rate may be as high as 55%.[2, 6, 8, 42] However, DA resistance differs from DA intolerance. A patient who suffers side effects from DAs that preclude their use even at a low dose is intolerant but may have a DA-sensitive tumour.
Regarding giant prolactinomas, the term ‘insufficient response to DA’ may be better than ‘resistance to DA’ because, although the tumours typically do respond to the drug, a normal PRL level may not be reached due to the large tumour load and very high PRL serum level at the time of diagnosis. True resistance to DA (complete absence of response) is rare, suggests tumour aggressiveness and may be indicative of malignancy. However, in the literature, the term ‘resistance to DA’ typically encompasses both patients with insufficient response and patients with true resistance.
The molecular mechanism of DA resistance is not completely understood. Reduced dopamine type 2 receptor (D2R) expression as well as possible alterations in intracellular signal transduction in the tumoural cells may be involved.[56, 57] The D2R encoded by the dopamine receptor D2 gene (DRD2) exists as two alternatively spliced isoforms, short (D2S) and long (D2L). Studies in vivo demonstrated that reduction in D2L isoform mRNA is correlated with DA resistance.[56, 57] In addition, recent data suggest that DA resistance may be related to reduced expression of filamin A.
What should we do about patients with insufficient response to dopamine agonists?
High doses of CAB/switch to CAB
Typically, 2 mg/week of CAB is sufficient to control PRL and tumour volume in patients with prolactinomas in general. A recent study by Vroonen et al. evaluating the efficacy of an exclusive high-dose CAB regimen (higher than 3·5 mg/week) in a subset of 19 patients who were resistant to DAs observed normal PRL levels in 26% (5/19) of the patients as well as some degree of tumour reduction of 53% (10/19) in the patients. Importantly, we should be aware of the possible risk of valvular cardiac disease when high CAB doses are necessary.
In patients with BRC-resistant tumours, switching to CAB is recommended. Seventy per cent of patients with BRC-resistant prolactinomas may exhibit normalized PRL during CAB therapy.
Cytoreductive surgery should be considered part of the multidisciplinary approach to patients with DA-resistant prolactinoma. Two studies have shown the importance of cytoreductive surgery in the treatment of giant prolactinomas.[41, 42] These authors reported that cytoreductive surgery is a good option and it can allow for better control of tumour volume and PRL level with DA administration after surgery.[41, 42] Complete surgical removal of a giant prolactinoma is challenging; tumour persistence after surgery occurs frequently, and biochemical cure is rare.
Yu et al. evaluated surgery as first-line therapy in 18 patients with giant prolactinomas (five transsphenoidal and 13 transcranial). Complete surgical removal and PRL normalization were not observed in any patients. Diabetes insipidus was described in 61% (11/18) of cases, visual deterioration occurred in 39% (7/18), and cranial nerve palsy occurred in 33% (6/18). Other surgical complications, such as CSF leakage and mortality, occurred in two patients (11%) and one patient (6%), respectively.
Regarding the type of surgery, in most cases, the transsphenoidal approach is the first choice for giant prolactinomas. A systematic review comparing traditional open or transsphenoidal microscopic surgery with the extended endoscopic endonasal transsphenoidal surgery for giant pituitary adenomas, including 19 giant prolactinomas, showed that the patients who underwent the endoscopic approach had higher rates of total resection (47% vs 9·6%, considering all pituitary adenomas) and improved visual defects (91% vs 46%, considering all pituitary adenomas) compared to the group that underwent the open approach. Furthermore, the endoscopic group had a higher rate of total resection (47% vs 31%, respectively) and better visual outcomes (91% vs 35%, respectively) than the microscopic transsphenoidal group.
In patients who underwent transsphenoidal microscopic surgery, Komotar et al. observed hypopituitarism in 9·5% (9/95), permanent diabetes insipidus in 8·7% (12/138), CSF leaks in 5·1% (13/256) and meningitis in 6·1% (10/165) of patients. Perioperative mortality occurred in 2·0% of the patients (5/256). In patients who underwent endonasal transsphenoidal surgery, postoperative complications included permanent diabetes insipidus in 4·7% (5/106), haemorrhage in 2·1% (2/94), sinusitis in 3·2% (3/94) and pulmonary embolus in 2·1% (2/94) of patients. No cases of CSF leak or meningitis were reported. In general, these rates are higher than those typically observed in surgery for nongiant prolactinomas.
Other therapies (Temozolomide – TMZ)
Temozolomide is an oral second-generation alkylating chemotherapeutic agent that inhibits all phases of tumour cell growth.
O6-methylguanine–DNA methyltransferase (MGMT) constitutes part of a DNA repair system that reverses alkylation at the O6 position of guanine. MGMT function is the principal mechanism involved in TMZ resistance because it neutralizes the effects of TMZ, which alkylates DNA at this position. TMZ may be a therapeutic option for patients with clinically aggressive giant prolactinoma or tumours that remain uncontrolled in terms of mass effect despite multiple conventional treatment modalities. Whitelaw et al. reviewed 11 patients with benign aggressive prolactinomas, including giant prolactinoma, who were treated with TMZ. Nine patients achieved a good response regarding tumour shrinkage. Studies have used several protocols for the dose regimen (number of cycles) and the duration of treatment (from 6 to 26 months). A possible association between negative/low MGMT status and a good response to TMZ was observed. In fact, Raverot et al. described an overall prediction of 67% for response to TMZ based on MGMT expression. Therefore, MGMT expression has been considered a test to define the chance of tumour response to TMZ. However, the positive predictive value of the test has been inconsistent, mainly because of the lack of a clear cut-off value. The complexity in scoring a tumour (intra- and interoperator heterogeneity) and the use of different antibodies and tissue preservation methods among studies lead to difficulty in comparing their results and make this protein a suboptimal predictor of efficacy. Thus, MGMT expression should most likely not be taken as grounds to deny these patients the potential benefit of TMZ treatment, taking into account the lack of other available treatments.
Usually, TMZ is well tolerated. Nausea is common but typically controlled with an antiemetic. Fatigue is also frequent and may be an important limiting side effect. Leukopenia and thrombocytopenia may occur in up to 4% and 17% of patients, respectively. Haematological malignancies and hepatotoxicity have been reported after TMZ use.
Radiotherapy should be used in patients with prolactinomas in general who do not achieve disease control with DA therapy and surgery, particularly regarding mass effect. To the best of our knowledge, only two series of cases have reported the efficacy of radiotherapy in patients with giant prolactinomas.[8, 66] One included four giant prolactinomas that were primarily treated with conventional radiotherapy (41·4–50 Gy) prior to BRC. All tumours responded to BRC therapy, and PRL decreased to the normal range in three patients and to just above the reference range in the fourth. Wu et al. observed the clinical outcomes of conventional radiation therapy in 11 of 20 patients with invasive giant prolactinoma treated with BRC. In this series, radiotherapy was indicated to accelerate tumour volume reduction and to enhance PRL control. There was no significant difference in tumour volume reduction between the groups with and without adjuvant radiotherapy.
In prolactinomas, in general, it has been suggested that suspending DA therapy prior to radiation therapy may increase tumour cell susceptibility to radiation; however, this relationship must be confirmed. The most frequent side effect of radiotherapy is hypopituitarism, which may affect all patients during long-term follow-up. In rare instances, cranial nerve damage or the development of a second tumour may occur. Cerebrovascular accident risk is higher in patients with pituitary adenoma. However, the precise reason for this side effect is not fully understood and seems to involve multiple factors, such as hypopituitarism, radiation injury to the vascular endothelium, local effects of the tumour and perioperative damage.
A summary of the management strategies for patients with giant prolactinoma is shown in Fig. 1.
Hypogonadism in patients with giant prolactinomas
The mechanisms of hypogonadism in giant prolactinomas appear to involve decreased serum levels of luteinizing hormone (LH) and follicle-stimulating hormone (FSH), due to low levels of gonadotrophin-releasing hormone (GnRH) reaching the gonadotrophic cells as a result of hyperprolactinaemia and stalk compression. In addition, the gonadotrophic cells may be compromised by mass effect.[71, 72] Because GnRH neurons are stimulated by kisspeptin neurons, Sonigo et al. studied the role of kisspeptin as a major regulator of GnRH neurons. Kisspeptin neurons express PRL receptor, and elevated PRL inhibits kisspeptin expression. Therefore, hyperprolactinaemia causing diminished kisspeptin expression affects hypothalamic release of GnRH, which in turn decreases LH and FSH secretion.
Hypogonadism in males
At least 50% of male patients with giant prolactinomas on DA treatment have persistent hypogonadism, a condition that is very disturbing for the patients. Hypogonadism is defined by total testosterone below 10·4 nmol/l and low or low-normal LH and FSH with impairment of spermatogenesis.
Persistent hypogonadism is treated with testosterone. Although there is a theoretical risk of increase in tumour size during androgen replacement, this condition has rarely been reported. The aromatization of testosterone to oestradiol appears to be responsible for this effect on tumour volume because oestradiol may stimulate both PRL synthesis and release as well as lactotroph cell proliferation. Gillam et al. reported a case of a patient with giant prolactinoma who initiated testosterone replacement and anastrozole therapy, and no tumour enlargement or side effects were documented. However, a long-term complication of anastrozole therapy is an elevated risk of osteoporosis and fractures, which may limit its use. Thus, if hypogonadism persists despite DA therapy, testosterone replacement should be initiated after excluding contra-indications (for example, prostate cancer and severe sleep apnoea), but close follow-up should be performed with serum PRL measurements and MRI.
When reproduction is desired but not achieved following normalization of PRL (with DA alone or DA and surgery), assisted reproductive technology treatment, such as gonadotrophin therapy, may be required.
Hypogonadism in females
The effects of oestrogen on PRL secretion and proliferation of lactotroph cells are well known, and the roles of the inhibitory properties of progesterone on PRL levels and on the effect of oestrogen have also been described.[78, 79] Thus, theoretically, oestrogen in combination with progesterone could be used to treat hypogonadism because the progesterone counteracts the effect of oestrogen on lactotroph proliferation. Patients with microprolactinoma are allowed to use contraceptive pills, although these may mask the clinical consequences of hyperprolactinaemia. In macroprolactinomas, including giant prolactinomas, the possibility of prescribing oestrogen must be evaluated on case-by-case basis, and the impact of the drug on the adenoma should be closely monitored. In women who cannot use oestrogen, the use of preventive strategies for bone loss should be considered.
For women who continue to be infertile and fail to restore eugonadism with standard therapy, additional hormonal therapy to induce ovulation, such as clomiphene citrate, recombinant gonadotrophins, or pulsatile GnRH therapy, may be required.
New molecules may constitute alternative tools in the therapeutic management of patients with prolactinomas, particularly those in whom disease control proves difficult or who relapse after standard therapy.
The mechanism of action of PRL receptor antagonists involves competition with endogenous PRL for receptor binding. These antagonists may have a potential role in the treatment of patients with prolactinomas whose tumour load is not an issue and whose PRL level does not achieve normalization. Therefore, these compounds should restore the GnRH pulsatility and thus fertility. These drugs could also be used alongside DAs in patients in whom DAs have resulted in tumour shrinkage but not PRL normalization.
In addition to the D2 receptor, prolactinomas express somatostatin receptors (SSTR), particularly SSTR1 and SSTR5, while SSTR2 is expressed at low levels.[81, 82] Hofland et al. compared the in vitro effects of pasireotide vs octreotide in three DA agonist-sensitive prolactinomas. They demonstrated that the somatostatin analogue that binds mainly to SSTR2 is useless in suppressing PRL secretion, whereas pasireotide considerably inhibits PRL secretion due to its high affinity for SSTR5. However, Fusco et al. showed that BIM-23268 (SSTR5-preferential) and pasireotide were unsuccessful in decreasing PRL secretion in cultures from prolactinomas resistant to DAs. Postreceptor defects in the transduction pathways may explain this combined absence of responsiveness to both the somatostatin analogues and DAs. Thus, the use of SSTR5-targeting compounds remains under exploration for patients who are intolerant to DAs.
Fukuoka et al. demonstrated that lapatinib, a tyrosine kinase inhibitor, suppressed PRL expression and colony formation on cell cultures. Indeed, in rats with prolactinomas, oral lapatinib therapy produced tumour shrinkage and serum PRL suppression. In cell cultures from human prolactinoma, lapatinib suppressed both PRL mRNA expression and secretion. These results indicate that the tyrosine kinase inhibitors may be a new drug class for patients with resistance to, or intolerance of, DAs.
Epigenetic mechanisms such as increased CpG island-associated methylation and enrichment of histones may be responsible for the low expression of D2R, a possible mechanism of DA resistance. A demethylating agent (zebularine) alone or in combination with a histone deacetylase inhibitor (trichostatin A) may induce the expression of D2R. This approach may constitute a new therapeutic strategy in patients who are resistant to conventional therapies.
Regarding treatment of hypogonadism, clomiphene citrate, a selective oestrogen receptor modulator that increases gonadotrophin secretion via a hypothalamic–pituitary action or gonadotrophins may be used to restore eugonadism. Ribeiro and Abucham found that the chronic administration of clomiphene citrate (50 mg/day) restored gonadal function (erectile function and sperm motility) in 10 of 14 patients with persistent hypogonadotrophic hypogonadism on DA treatment. The PRL level was not a predictive factor of response to clomiphene citrate. Testosterone increased significantly and peaked at 4 weeks (from 7·0 ± 0·8 to 15·9 ± 1·3 nmol/l), whereas oestradiol, LH and FSH increased progressively over 12 weeks. However, as soon as the clomiphene was withdrawn after 12 weeks, testosterone returned to the hypogonadal range. More studies with larger series are necessary to determine whether clomiphene citrate is an efficacious and safe long-term treatment for hypogonadism in these patients.
A recent study in hyperprolactinaemic rats demonstrated that kisspeptin administration restored LH pulsatility due to GnRH neuron recovery as well as improvement of ovarian cyclicity. Therefore, the administration of kisspeptin might also serve as an alternative therapeutic approach to restore the gonadal status in patients who fail to restore eugonadism with standard therapy, when there is no stalk compression or compromise of gonadotrophic cells compromised by mass effect.
Giant prolactinoma is an uncommon condition, mainly found in men. Given the rarity of this disease, previous studies have mostly been case reports or retrospective series with small numbers of cases. There is no consensus regarding the definition of a giant prolactinoma, and we propose that the most important factor to classify a macroprolactinoma as being giant is a maximum diameter of more than 4 cm. Dopamine agonists, mainly CAB, are the first-line medical therapy for patients with giant prolactinomas because the majority of cases achieve a normal PRL level and significant tumour reduction. Due to its invasiveness, total gross tumour resection is almost unfeasible, but surgery should be considered as part of this multidisciplinary approach, especially in those patients who still have a large tumour load despite DA therapy. Radiotherapy and/or TMZ may be reserved for a subgroup of giant prolactinomas that exhibit an aggressive clinical behaviour in terms of mass effect. Restoration of eugonadism should be attained. Finally, new molecules are under investigation and might become important therapeutic tools in the management of patients with giant prolactinoma.
We would like to thank Prof. Dr. Alberto M. Pereira from the Leiden University Medical Center and Prof. Dr. Jacqueline Trouillas from Lyon University for their helpful discussions about the management of patients with giant prolactinomas and the pathological aspects of this subset of prolactinomas.