Craniopharyngioma 

Craniopharyngiomas are dysontogenic tumors with benign histology and malignant behavior. These lesions have a tendency to invade surrounding structures and to recur after a seemingly total resection (see the image below). (See Etiology and Treatment.)

Craniopharyngiomas most frequently arise in the pituitary stalk and project into the hypothalamus. They extend horizontally along the path of least resistance in various directions, as follows:

• Anteriorly - Into the prechiasmatic cistern and subfrontal spaces
• Posteriorly - Into the prepontine and interpeduncular cisterns, cerebellopontine angle, third ventricle, posterior fossa, and foramen magnum
• Laterally - Toward the subtemporal spaces.

The tumors can even reach the sylvian fissure. In rare cases, the tumors can occur extradurally or extracranially, developing as nasopharyngeal or pure posterior fossa craniopharyngiomas or as craniopharyngiomas extending down the cervical spine. Purely intraventricular craniopharyngioma is usually of the squamous-papillary (metaplastic) type; it occurs very rarely.

Craniopharyngioma usually presents as a single large cyst or multiple cysts filled with a turbid, proteinaceous, brownish yellow material that glitters and sparkles because of a high content of floating cholesterol crystals. Because of its appearance, this material has been compared with machinery oil. (See Etiology and Workup.)

Clinical behavior and the choice of surgical approach are dictated by the primary location of the tumor and its extension pattern.^[1] Prechiasmatic craniopharyngiomas (extending into subfrontal spaces) and retrochiasmatic craniopharyngiomas (expanding into the posterior fossa) may reach large sizes before being diagnosed. (See Presentation and Workup.)

Vascular supply

The craniopharyngioma’s vascular supply is dependent on different sources, usually all of which come from the anterior circulation. The anterior portion of the tumor is supplied by small perforators coming off of A1 (ie, anterior cerebral artery); lateral portions receive perforators from the proximal portion of the posterior communicating artery; and the intrasellar part is supplied by branches of the intracavernous meningohypophyseal arteries. Craniopharyngiomas are rarely supplied with blood coming from the posterior circulation, unless the anterior blood supply for the anterior hypothalamus and floor of the third ventricle is lacking.

Recurrence

Recurrences usually occur at the primary site. Ectopic and metastatic recurrences are extremely rare but have been reported after surgical removal. The two possible mechanisms of seeding are dissemination of tumor cells along the surgical paths during the procedure and migration of tumor cells through the subarachnoid space or Virchow-Robin spaces (which would explain ectopic recurrences distant from the surgical bed and within brain parenchyma).

In one metastatic case, after removal of a suprasellar (adamantinomatous) craniopharyngioma, 2 peripheral lesions were identified 7 years later, adjacent to the dura and contralateral to the initial craniotomy site. They proved to be composed of adamantinomatous tissue, raising the possibility of meningeal seeding.

In another reported case, an adamantinomatous craniopharyngioma recurred at different intervals, at different sites, along the operative track of the initial surgical procedure, as well as distant to it, within the brain parenchyma; this suggests that both seeding mechanisms were involved in these recurrences.

In a large, retrospective review, the histopathologic type of craniopharyngioma and/or brain invasion did not correlate with the risk of recurrence.^[2]

A craniopharyngioma is a slow-growing, extra-axial, epithelial-squamous, calcified, cystic tumor arising from remnants of the craniopharyngeal duct and/or Rathke cleft and occupying the (supra)sellar region. Two main hypotheses—embryogenetic and metaplastic—explain the origin of craniopharyngioma. These hypotheses complement each other and explain the craniopharyngioma spectrum.

Embryogenetic theory

This theory relates to development of the adenohypophysis and transformation of the remnant ectoblastic cells of the craniopharyngeal duct and the involuted Rathke pouch. The Rathke pouch and the infundibulum develop during the fourth week of gestation and together form the hypophysis. Both elongate and come in contact during the second month. The infundibulum is a downward invagination of diencephalon; the Rathke pouch is an upward invagination of the primitive oral cavity (ie, stomodeum).

The craniopharyngeal duct is the neck of the pouch, connecting to the stomodeum, which narrows, closes, and separates the pouch from the primitive oral cavity by the end of the second month. Thus, the pouch becomes a vesicle, which flattens and surrounds the anterior and lateral surfaces of the infundibulum. Walls of this vesicle form different structures of the hypophysis. Finally, this vesicle involutes into a mere cleft and may disappear completely.

The Rathke cleft, together with remnants of the craniopharyngeal duct, can be the site of origin of craniopharyngiomas.

Metaplastic theory

This theory relates to the residual squamous epithelium (derived from the stomodeum and normally part of the adenohypophysis), which may undergo metaplasia.

Dual theory

This theory explains the craniopharyngioma spectrum, attributing the adamantinomatous type (most prevalent in childhood) to embryonic remnants, and the adult type (ie, squamous papillary) to metaplastic foci derived from mature cells of the anterior hypophysis. (Prevalence of the adult type increases with each decade of life and is almost never found in children.)

Other cystic lesions may originate from remnants of the stomodeum and pharyngohypophyseal duct as well, such as Rathke cleft cysts, epithelial cysts, epidermoid cysts, and dermoid cysts.

Genomic and molecular biology of craniopharyngiomas

Comparative genomic hybridization (CGH) studies have been reported with conflicting results. CGH sensitivity is limited to deletions of the order of several megabases; thus, smaller deletions and balanced alterations can be missed.^[3]

Some suggest that chromosomal imbalances^[4] do not play a significant role in tumorigenesis of papillary and adamantinomatous craniopharyngiomas. Others report a small subset of adamantinomatous craniopharyngiomas showing a significant number of genetic alterations and abnormal deoxyribonucleic acid (DNA) copy number, thus suggesting a monoclonal origin driven by the activation of oncogenes located at specific chromosomal loci.^[5]

Adamantinomatous craniopharyngiomas have been consistently reported to show alterations in beta-catenin gene expression.^{[6, 7, 8]} Expression of beta-catenin correlates with some of the hallmarks ("wet" keratin, calcifications and palisading cells) of adamantinomatous craniopharyngiomas. This abnormality has not been reported in papillary craniopharyngiomas.

Beta-catenin is a transcriptional activator of the Wnt signaling pathway and a component of the adherence junction. The Wnt signaling pathway has been proven to play a crucial role in embryogenesis and cancer. Wnt signaling is involved in determination of cell fate, proliferation, adhesion, migration, polarity, and behavior during development and plays an intricate role in the temporal and spatial regulation of organogenesis.

The Wnt complex is made up of 3 different pathways: canonical, noncanonical, and Wnt/Ca⁺². The canonical pathway regulates cell fate determination and primary axis formation through gene transcription. The noncanonical pathway regulates cell movements through modification of the actin cytoskeleton. The Wnt/Ca⁺² pathway is involved in regulation of both cell movement and fate determination.

Immunohistochemistry for beta-catenin in adamantinomatous craniopharyngiomas showed an abnormal cytoplasmic and nuclear accumulation. The normal membranous staining was present in adamantinomatous and papillary craniopharyngiomas.

Sequencing analysis revealed beta-catenin gene mutations in adamantinomas, while none were found in papillary craniopharyngiomas. All mutations were missense mutations involving the serine/threonine residues at glycogen synthase kinase-3beta (GSK-3beta) phosphorylation sites or an amino acid flanking the first serine residue. These mutations are believed to lead to beta-catenin accumulation as a result of impaired proteosome degradation, this degradation itself being due to ineffective phosphorylation by a mutated GSK-3beta.

Furthermore, the Wnt/beta-catenin signaling pathway has been shown to prevent differentiation (of mouse embryonic stem cells) through convergence on the LIF/Jak-STAT (leukemia inhibitory factor/Janus kinase ̶ signal transducer and activator of transcription) pathway at the level of STAT3.^[9] Interferons are known modulators of Jak/STAT pathways, thus revealing the possible molecular basis for interferons as a therapeutic option in adamantinomatous craniopharyngiomas.

Some craniopharyngiomas express insulinlike growth factor receptors (IGF-1Rs) and sex hormone receptors (estrogen receptors [ERs] and progesterone receptors [PRs]).^{[10, 11]} Despite reported sporadic expression of IGF-1R in 2 large, retrospective reviews (including children and adults) in which the mean treatment duration was 6 years and the mean follow-up period was approximately 10 years, no evidence was found to suggest increased recurrence rates in patients who received growth hormone supplementation.^{[12, 13]}

ER and PR expression in 1 correlative study was linked to higher differentiation and a decreased incidence of tumor recurrence and was proposed as a tool for recurrence risk stratification.

Other markers have been proposed for noninvasive clinical monitoring. Urinary matrix metalloproteinases (MMPs, nonspecific tumor invasion markers) in one case were reported to be a useful predictor of disease activity and risk of recurrence.^[14]

Expression of human minichromosome maintenance protein 6 (MCM6) and DNA topoisomerase 2 alpha (DNA Topo 2 alpha) were proposed as histologic markers associated with a higher risk of recurrence in adamantinomatous craniopharyngiomas.

Occurrence in the United States

Data from the Central Brain Tumor Registry of the United States (CBTRUS), collected between 1990 and 1993, revealed an average of 338 cases diagnosed annually, with 96 occurring in children aged 0-14 years.^[15] Other results were as follows:

• Overall incidence was 0.13 per 100,000 person years
• No variance by gender or race was found
• Craniopharyngioma accounted for 4.2% of all childhood tumors (ages 0-14 years)
• Distribution by age was bimodal, with peak incidence in children aged 5-14 years and adults aged 65-74 years

International occurrence

The incidence of craniopharyngioma is 0.5-2 per 100,000 per year. Overall, craniopharyngioma accounts for 1-3% of intracranial tumors and 13% of suprasellar tumors. In children, craniopharyngioma represents 5-10% of all tumors and 56% of sellar and suprasellar tumors. No definite genetic relationship has been found, and very few familial cases have been reported.

Race-, sex-, and age-related demographics

Higher frequencies of all intracranial tumors have been reported from Africa, the Far East, and Japan; they are 18%, 16%, and 10.5%, respectively.

A slight male predominance exists in all age groups (55%). The age of diagnosis varies widely; cases have been reported in fetuses and in the elderly. Age distribution is bimodal–the first peak is in children aged 5-10 years and a second one is in adults aged 50-60 years.

In the United States, data collected during the periods 1985-1988 and 1990-1992, coinciding with the introduction of computed tomography (CT) scanning, for the National Cancer Data Base (NCDB), indicate that survival rates for craniopharyngioma were 86% at 2 years and 80% at 5 years after diagnosis.

Survival rate varied by age group, with excellent rates for patients younger than 20 years (99% at 5 years). Survival rate was poor for those older than 65 years (38% at 5 years).

Female sex has been reported as an independent predictor of increased cardiovascular, neurologic, and psychosocial morbidity.^[16]