Craniopharyngiomas are rare benign tumors derived from cell remnants of Rathke’s pouch along a line from the nasopharynx to the diencephalon. The majority of these epithelial tumors occur in the sellar, para-, and suprasellar region (Jane and Laws, 2006; Karavitaki et al., 2006; Garnett et al., 2007). In the United States, craniopharyngiomas constitute approximately 2.5–3% of brain tumors with an incidence rate (IR) of 1.1–1.3 per million (Bunin et al., 1998; Jane and Laws, 2006; Karavitaki et al., 2006). In other parts of the world, like Japan or certain parts of Africa, IR seem to be higher (Stiller and Nectoux, 1994). A recent analysis calculated a worldwide overall IR of 1.34 in all ages (Nielsen et al., 2011). About half of the cases occur in adults with a peak IR between 40 and 44 years according to data from a Danish study (Nielsen et al., 2011). The male to female ratio is 1.1–1.4:1 (Nielsen et al., 2011). Although craniopharyngiomas are regarded as benign tumors they often infiltrate adjacent structures like the pituitary, hypothalamus, optic nerves, blood vessels, or the third ventricle thereby causing significant morbidity and mortality. As a consequence patients present with a wide range of symptoms which are often permanent or might even be exacerbated by therapy. Because of their infiltrative growth behavior and their high tendency for recurrence, the treatment is often challenging. In the present article we present an overview on craniopharyngiomas with special emphasis on patients with adult-onset of the disease.

Craniopharyngiomas are divided in two main histological subtypes, the adamantinomatous and the papillary type, but transitional and mixed variants have also been reported (Weiner et al., 1994; Crotty et al., 1995; Louis et al., 2007). The adamantinomatous type may develop at all ages, whereas the papillary type almost exclusively occurs in adults comprising about 14–50% of the tumors in this age group (Adamson et al., 1990; Weiner et al., 1994; Karavitaki et al., 2006). The papillary type rarely presents with calcifications, is generally well-circumscribed and compared to the adamantinomatous type tumor infiltration of surrounding tissue is less frequent (Weiner et al., 1994; Crotty et al., 1995). Macroscopically craniopharyngiomas are predominantly cystic or mixed lesions although solid lesion might also occur. The tumor size varies from small, solid, and well-circumscribed tumors to large multilocular cysts invading the sella turcica and displacing adjacent cerebral structures (Crotty et al., 1995; Karavitaki et al., 2005). Most tumors are located in the sellar/parasellar region with suprasellar tumor mass extension. A smaller number of tumors are confined to the sella, or arise in the third ventricle or within the optic system (Van Effenterre and Boch, 2002).

The molecular pathogenesis of craniopharyngiomas remains widely unknown. In patients harboring the adamantinomatous variant mutations in the beta-catenin gene CTNNB1 resulting in degradation-resistant mutant forms of beta-catenin have been identified (Sekine et al., 2002; Kato et al., 2004; Buslei et al., 2005; Campanini et al., 2010). Studies in rodents showed that permanent activation of the Wnt signaling pathway, which leads to high expression levels of beta-catenin, causes formation of pituitary tumors that resemble the human adamantinomatous craniopharyngioma. Thus, reactivation of the Wnt signaling pathway may be one factor in the pathogenesis of craniopharyngiomas of the adamantinomatous type (Gaston-Massuet et al., 2011). The formation of cysts and their size appears to be associated with the expression of carbonic anhydrase IX, an enzyme causing fluid production. The specific tumor-associated regulating mechanisms, however, of this enzyme have not been clarified yet (Proescholdt et al., 2011).

The clinical manifestations are multiple and depend on the location of the tumor, its size, growth pattern, and relationship to adjacent cerebral structures (Table 1). As craniopharyngiomas are in general slowly growing, symptoms may develop gradually – a circumstance which may contribute to the reported delay of 1–2 years between symptom onset and diagnosis (Garnett et al., 2007). In adults the most common presenting clinical symptoms are visual field deficits and signs of hypopituitarism (Hoff and Patterson, 1972). In about 40–80% of the patients suprasellar tumor extension pressurizes the optic chiasm causing loss of visual acuity and visual field abnormalities (mostly asymmetric bitemporal hemianopsia; Crotty et al., 1995; Karavitaki et al., 2005, 2006). In a series of 78 adult patients 57% of the female patients reported about menstrual irregularities or amenorrhea and 28% complained about impaired sexual function (Karavitaki et al., 2005). Other symptoms like nausea and vomiting (26%), poor energy (32%), and lethargy (26%) are also frequent in the adult patient (Karavitaki et al., 2005). These symptoms are indicative for anterior pituitary dysfunction. Overall, growth hormone (GH) deficiency is most common, followed by gonadotropin, adrenocorticotropic hormone (ACTH), and thyroid stimulating hormone (TSH) deficiency, which are present in 86, 74, 58, and 42% of adult cases, respectively (Karavitaki et al., 2005). Compression of the pituitary stalk causes diabetes insipidus (DI) with polyuria and polydipsia in 17–38.5% of cases (Karavitaki et al., 2005; Mortini et al., 2011). Hypothalamic involvement may cause significant weight gain, which is a presenting symptom in 13–15.4% of adult patients (Karavitaki et al., 2005; Mortini et al., 2011). Severe headaches are also frequent (56%) and may be caused by raised intracranial pressure due to the tumor mass itself, to obstructive hydrocephalus resulting from compression of the third ventricle or due to leaked out cyst content which could lead to meningeal irritation (Karavitaki et al., 2006). Parasellar tumor extension with infiltration of the cavernous sinus could result in cranial nerve palsies with diplopia and paresis of ocular muscles. Involvement of the temporal lobe might trigger seizures and especially in the elderly, large tumors may cause deteriorating cognitive abilities and personality changes (Karavitaki et al., 2006).

The suspicion of a craniopharyngioma is based on clinical and radiological findings. The diagnosis is finally proven by histology. A proposal for the initial diagnostic work-up in patients suspicious for craniopharyngioma is given in Table 2.

Most adult patients present with visual field deficits and signs of hypopituitarism (Hoff and Patterson, 1972). Visual field and visual acuity assessment is therefore important to determine any deficit and to establish baseline values. In addition, an examination of the optic disks is required to rule out papilledema and visual evoked potentials are recommended. An endocrine evaluation should include the assessment of anterior and posterior pituitary functions. Signs and symptoms indicative for hypopituitarism including DI should be asked for. Basal ACTH and cortisol, TSH and free T4, LH and testosterone in men and FSH and estradiol in women, IGF-1 and prolactin as well as serum electrolytes, serum, and urine osmolality are determined. Further endocrine testings should be performed according to the results and the individual clinical situation.

Cranial computer tomography (CT) and/or magnetic resonance imaging (MRI) with and without contrast enhancement are part of the neuroradiological evaluation of craniopharyngiomas. Calcifications and bony structures are best evaluated on CT, while the delineation of tumor size and the involvement of neighboring structures can be assessed most accurately by MRI. The MRI scan is therefore the method of choice in order to prepare a surgical approach (Garnett et al., 2007). According to a recent study, there are no significant differences in neuroradiological characteristics of craniopharyngiomas between children and adults with the exception of lower rates of tumor calcification in adult patients (Mortini et al., 2011). In some studies hydrocephalus occurred less frequently in adults than in children, however other authors reported no differences in this regard (Van Effenterre and Boch, 2002; Karavitaki et al., 2005). Although several attempts have been made, no reliable discrimination between the two histological subtypes is possible based on neuroradiological criteria (Crotty et al., 1995; Eldevik et al., 1996; Sartoretti-Schefer et al., 1997). In some cases the presence of a craniopharyngioma is suggested by an incidental finding of a mass lesion on neuroimaging scans. The differential diagnosis comprises other sellar and parasellar tumors like Rathke’s cleft cysts and other non-malignant cysts, pituitary adenomas, optic and hypothalamic gliomas, meningiomas, teratomas, germinomas, lymphomas, metastases, and infiltrative diseases, like sarcoidosis or systemic histiocytosis (Garnett et al., 2007).

Prospective and controlled studies comparing different treatment strategies in adult patients are missing. Most data come from retrospective analyses of mixed populations of patients with childhood- and adult-onset disease. There are no evidence-based guidelines or a clear consensus for best treatment of primary or recurrent craniopharyngiomas in adults. Like in children a radical approach with complete tumor resection and potential cure has to be balanced with a more conservative approach to avoid substantial treatment-associated long-term morbidity. Therefore, optimal treatment should always be individualized taking into account, e.g., the patient’s symptoms, age, tumor localization, and extension. Endocrine and metabolic complications should be treated prior to any tumor-directed therapy such as surgery. This is especially important in case of adrenal or thyroidal insufficiency, as well as DI. It is generally agreed that a multidisciplinary team comprising endocrinologists, neurosurgeons, radiotherapists, and ophthalmologists is essential for optimized treatment of patients of any age (Garnett et al., 2007; Garre and Cama, 2007; Honegger and Tatagiba, 2008).

Surgery is the first-line approach to the treatment of craniopharyngiomas. Complete surgical resection is the goal of initial therapy and because of major advances in neurosurgical techniques, it is increasingly feasible. However, due to their size, their often irregular margins and intense peritumoral gliosis with adherence to vital neighboring neurovascular structures this is often difficult and potentially dangerous. According to the literature complete tumor removal by radical surgery can be achieved in 18–84% of selected childhood and adult cases (Weiner et al., 1994; Fahlbusch et al., 1999; Van Effenterre and Boch, 2002; Karavitaki et al., 2005, 2006). Aggressive surgery with gross total resection (GTR) of the tumor, however, may result in significant and devastating peri- and postoperative morbidity especially after resection of tumors invading the hypothalamus (Rutka, 2002). If complete tumor removal is unlikely to be achieved or too hazardous, subtotal, or partial resection (PR) of the tumor, e.g., to reduce pressure effects on adjacent structures like the optic pathways and/or to re-establish circulation of the cerebrospinal fluid with or without subsequent radiotherapy is an alternative approach (Habrand et al., 1999; Cama et al., 2006). In any case the benefits of surgery must be balanced against the risks of treatment-related morbidity. For pediatric patients a preoperative grading system has been proposed according to the degree of hypothalamic involvement (Puget et al., 2007). Using this classification a significant relationship between the preoperative tumor grade and post-surgical morbidity could be demonstrated. GTR should only be attempted in patients without hypothalamic involvement (grade 0) or with a distorted or elevated but still visible hypothalamus (grade 1). In grade 2 craniopharyngiomas (hypothalamic structures not discernible) a subtotal resection leaving the hypothalamic part is recommended (Puget et al., 2007). Similar studies in adults are lacking. Nevertheless it appears reasonable to follow these recommendations established in pediatric patients. For craniopharyngiomas with major cystic portions stereotactic cyst decompression is another treatment option. This can be used for acute symptomatic therapy, prior to definite craniopharyngioma resection, or may be recommended as intermittent therapy whenever a total cyst excision is not possible (Fahlbusch et al., 1999; Honegger and Tatagiba, 2008).

Radiotherapy (RT) is a therapeutic option in patients for whom surgery is contraindicated, with residual tumor tissue after subtotal or partial surgical resection, as well as in patients with recurrent disease. Advances in RT techniques allow higher treatment precision with less long-term toxicity by focusing the ionizing radiation on the tumor and limiting the exposure of adjacent tissues to a minimum. Techniques used for the treatment of craniopharyngiomas include stereotactic fractionated radiotherapy, radiosurgery, intensity modulated radiation therapy, and proton beam therapy (Fitzek et al., 2006; Minniti et al., 2009; Niranjan et al., 2010). For fractionated conformal RT schedules the probability of local tumor control is best with radiation doses above 54–55 Gy (Regine et al., 1993) but less than 61 Gy to lower the risk of radiation-associated side-effects such as visual impairment, pituitary deficiency, impaired cognitive function or development of secondary malignancies (Cavazzuti et al., 1983; Merchant et al., 2006). Application of radiosurgery with delivery of a single high-dose radiation to the tumor is an attractive technique but restricted to smaller tumor volumes and requires sufficient safety margins of adjacent critical structures. Although reports published so far are promising (Minniti et al., 2009; Iwata et al., 2012), further studies are needed to define its role in the prevention of tumor recurrence and its potential long-term adverse effects. Intracavitary irradiation with stereotactically guided instillation of radioisotopes is another approach to treat mono- or multicystic tumors (Voges et al., 1997; Barriger et al., 2011).

Other treatment options for predominantly cystic craniopharyngiomas comprise intracystic chemotherapy or immunological therapy with bleomycin or interferon alpha, respectively (Steinbok and Hukin, 2010). The experience regarding these treatments is mainly restricted to pediatric patients. In general, these options were used for temporary tumor control with the expectation for further therapies like surgery or radiotherapy. Systemic chemotherapy is generally regarded as ineffective (Karavitaki et al., 2006).

Recurrence of craniopharyngiomas remains a problem which highly impacts on the long-term prognosis of the patients. In a recent study with 106 patients including 78 adults, 24.5% experienced a relapse within a median follow-up of 83 months (Mortini et al., 2011). The risk of recurrence appears highest in the first 3 years after surgery and then tends to reach a plateau (Fahlbusch et al., 1999; Karavitaki et al., 2005). After GTR the risk of tumor recurrence has been reported to range between 0 and 26% (Maira et al., 1995; Fahlbusch et al., 1999; Duff et al., 2000). This is significantly lower than in patients with subtotal or partial tumor resection in whom tumor recurrence occurs in 25–100% of the patients (Fahlbusch et al., 1999; Duff et al., 2000; Karavitaki et al., 2005). In these patients adjuvant RT significantly improves the tumor control rates with recurrence rates ranging from 10 to 63% at 10-years follow-up (Karavitaki et al., 2006). Thus, patients with subtotal resection without subsequent RT have the highest risk of relapse (Duff et al., 2000; Karavitaki et al., 2005). In patients treated with RT alone tumor growth or recurrence was observed in 0–23% of the patients at 10 years follow-up (Hetelekidis et al., 1993; Rajan et al., 1993). The data about the prognostic significance of age at diagnosis, gender, tumor size, location, or consistency, as well as the histological subtype are inconsistent (Karavitaki et al., 2005, 2006; Mortini et al., 2011), but in general these factors do not seem to affect the risk of recurrence (Karavitaki et al., 2006).

Recurrent disease is difficult to treat. Because of scars and adhesions due to previous surgery or radiation, the surgical success rates for recurrent disease are significantly lower than for primary surgery and peri- and postoperative morbidity and mortality is significantly increased (Yasargil et al., 1990; Wisoff, 1994; Fahlbusch et al., 1999; Karavitaki et al., 2005). Therefore, the severity of clinical symptoms has to be taken into account and it has been suggested that repeated surgery should only be performed when acute pressure effects occur (Karavitaki et al., 2006). In patients not previously treated with RT irradiation is another option either instead of or following a second surgery, and it appears that RT is equally effective in the control of recurrent tumors as for primary treatment (Jose et al., 1992). Further options comprise cyst controlling procedures like aspiration, intracystic irradiation or application of bleomycin, salvage surgery in case of life-threatening solid lesions, or radiosurgery. In any case treatment decisions for recurrent disease should be made on an individualized basis.

Control of tumor growth and the disease- or treatment-related complications determine the long-term prognosis. Overall survival rates reported range from 89 to 94% at 5-years and from 85 to 90% at 10-years follow-up (Van Effenterre and Boch, 2002; Karavitaki et al., 2005). Nevertheless, overall mortality appears three to five times higher than those of the general population (Sherlock et al., 2010). Currently, it is unclear which therapeutic strategy (e.g., GTR or subtotal or PR followed by radiotherapy) is associated with a better survival (Karavitaki et al., 2005; Sherlock et al., 2010) and whether the histological subtype is of relevance (Weiner et al., 1994). Recurrent disease, however, is significantly associated with increased morbidity and mortality, resulting in lower 10-year survival rates (29–70%; Karavitaki et al., 2005; Honegger and Tatagiba, 2008). Apart from mortality caused by the tumor itself or by surgical treatment, mortality from cardio- and cerebrovascular as well as respiratory causes is increased (Bulow et al., 1998; Tomlinson et al., 2001). It is therefore essential to optimize the treatment of any endocrine and metabolic sequelae of the disease in order to reduce mortality in this high risk patient population.

Although craniopharyngiomas are generally benign their location, size, and tendency to infiltrate adjacent cerebral structures makes their management rather demanding and may lead to sometimes devastating complications. According to the data available adult- and childhood-onset craniopharyngiomas behave similar in many aspects. Treatment decisions for primary or recurrent disease need to consider long-term tumor control as well as disease and treatment-related morbidity and they are often reached on an individualized basis. Surgery is the initial treatment approach, which should remove as much tumor as safely possible, while avoiding severe treatment-induced complications. There is some controversy as to whether a more radical surgical approach with gross total tumor resection should be aimed at or a more limited approach with reducing the tumor mass followed by radiotherapy in order to minimize long-term morbidity. As some patients with subtotal or partial tumor resection may have stable disease and since RT may cause long-term complications it is currently unclear, whether all patients with residual tumor should receive immediate postoperative RT. Likewise it is unknown which therapeutic strategy is best in patients with tumor recurrence despite RT. Posttreatment follow-up should include neuroimaging with MRI, visual assessment, and monitoring of endocrine functions and metabolic status. Endocrine and metabolic alterations need to be treated adequately. There is no doubt that for optimal results a lifelong surveillance of an experienced multidisciplinary team is essential.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.