Craniopharyngioma Treatment & Management

Essentially, two main management options are available for craniopharyngiomas: (1) attempt a gross total resection or (2) perform a planned subtotal resection followed by radiotherapy or some other adjuvant therapy.

No firm consensus exists concerning the appropriate management of craniopharyngiomas, and no guidelines have been established yet.

Most of the accepted management strategies are from retrospective reviews; no prospective, randomized clinical trials have been conducted to compare the various therapeutic modalities.

Although no consensus exists, most authors maintain that successful management is determined by the ability to preserve independent social functioning, prevent symptomatic recurrence, and increase survival rate.

Neuropsychological deficits represent the major limiting factor for independent social functioning because (1) patients often can overcome minor neurologic deficits and (2) hormone replacement therapies are widely available. The degree of psychosocial impairment correlates directly with the degree of hypothalamic injury sustained at the time of surgery.

There has been significant debate in recent years regarding the outcomes of GTR (Gross total removal) in the pediatric population given the high risk for hypothalamic injury and deficits, which can be life-altering in children (i.e., extreme obesity, deterioration in educational abilities).

Attempts at employing systemic chemotherapy in the treatment of craniopharyngiomas have been unsuccessful. Systemic biologic therapies currently under investigation include interferon (IFN) alpha-2a for progressive or recurrent craniopharyngiomas, with promising results.

Research has led to advances in medical treatment of craniopharyngiomas (eg, with BRAF and MEK inhibitors), suggesting that neoadjuvant therapy for these lesions should be considered in appropriate patients.

Inflammatory cytokines and biomodulation

Several inflammatory cytokines have been shown to be elevated in the craniopharyngioma cyst fluid in comparison to CSF. Interleukin (IL)–1alpha and tumor necrosis factor (TNF)–alpha levels may be significantly elevated. The concentration of IL-6 may be over 50,000 times greater in the cystic fluid than in the CSF. ^[26]  These findings support the hypothesis that biomodulation of the cytokine profile can lead to prolonged stability and even tumor regression.

IFN-alpha exerts diverse influences mainly on cytokine antagonists and soluble adhesion molecules. It has been shown to play a role in the treatment of craniopharyngioma after systemic as well as local, direct intracystic delivery. ^[27]

Postsurgical follow-up should be planned in 1-2 weeks for all patients. Patients with subtotal resections who are candidates for radiation therapy should start radiation usually within 3 weeks of surgery. Patients with either complete resections or completed radiation should be seen every 3 months for the first postsurgical year, every 6 months for the second and third years, and yearly thereafter. Strict follow-up is advised.

Each follow-up visit should include a brain MRI to be used for comparison with previous films and to correlate imaging with the clinical exam and neurocognitive testing results. Neuroendocrine and neuroophthalmology status should be followed up as well.

Neurocognitive testing must be considered for preoperative and postoperative patients, as well as patients who have undergone subtotal resection followed by radiation. All patients should have neurocognitive testing if performance at school or workplace drastically declines or clinical examination reveals worsening neurocognitive deficits (i.e., problem solving, language, memory, apraxia). ^{[22, 28]}

In some patients, deficits encountered are related to radiation injury. These are identified by specific MRI findings and correlated with neurocognitive testing results. Subsequently, specific treatments can be used. Close monitoring of endocrine dysfunction as evidenced by symptoms and confirmatory laboratory tests are recommended for all patients. Most patients require multiple hormonal supplements and adjustments during their postsurgical/postradiation phase and even years later.

Preventive management of long-term and multisystem morbidities is key for a successful outcome. A comprehensive multidisciplinary approach is strongly recommended. Panhypopituitarism was reported in almost 90% of patients followed for more than 10 years. Long-term follow-up with endocrinology is strongly recommended.

Other prevalent morbidities include neurologic (49%), psychosocial (47%), and cardiovascular (22%) abnormalities. The female sex is reported as an independent predictor of increased cardiovascular, neurologic, and psychosocial morbidity. Long-term follow-up should include appropriate hormonal replacement ^[29]  (including estrogen in premenopausal women) and aggressive control of cardiovascular risk factors (blood pressure, weight, lipids, and glucose). 

Other prevalent morbidities include neurologic (49%), psychosocial (47%), and cardiovascular (22%) abnormalities. The female sex is reported as an independent predictor of increased cardiovascular, neurologic, and psychosocial morbidity. Long-term follow-up should include appropriate hormonal replacement ^[29] (including estrogen in premenopausal women) and aggressive control of cardiovascular risk factors (blood pressure, weight, lipids, and glucose).

Recurrence

Immunohistochemical studies and case reports suggest higher incidence of recurrence in patients receiving growth hormone and/or sex hormone replacement, as some craniopharyngiomas express insulin-like growth factor receptors (IGF-1Rs), estrogen receptors (ERs), and progesterone receptors (PRs).

Despite the sporadic expression of IGF-1Rs, two large retrospective reviews assessing children and adults, in which the mean treatment duration was 6 years and the mean follow-up was 10 years, reported no evidence of increased recurrence rates in patients who received growth hormone supplementation. ^{[14, 15]}  Imaging follow-up every 4-6 weeks and close clinical monitoring are indicated with sex hormone and/or growth hormone replacement. ^[30]

Agents/modalities used in the treatment of craniopharyngioma include (1) radiation therapy applied as external fractionated radiation, stereotactic radiation, or brachytherapy (intracavitary irradiation) ^{[31, 32, 33, 34, 35]} and (2) bleomycin for local intracystic chemotherapy. ^{[36, 37, 38]}

Radiation therapy

Radiation creates free oxygen ions that damage cellular DNA. The cells’ ability to repair DNA is lower for tumor cells than for normal cells, and, subsequently with each cycle of mitosis, a higher cumulative effect in tumor cells results in apoptosis.

Proton beam radiation

This radiation modality became more and more common in recent years for this kind of tumor, mainly because of the Bragg peak effect, which means the energy beam peak occurs immediately before the particles come to rest. In a review published by Bishop and colleagues, ^[39] they didn’t find any difference between conformal photon radiotherapy and proton beam therapy in terms of overall survival and solid and cystic control among pediatric population with craniopharyngioma.

External fractionated radiation

This offers a dual advantage by (1) allotting normal cells more time for repair and (2) amplifying a higher cumulative effect of DNA damage in more rapidly dividing tumor cells.

Radiation following a partial resection offers excellent long-term results (80% at 20 years). When compared, the results of giving radiation after partial resection are superior to those achieved when radiation is delayed until the time of recurrence. Recurrence is less frequent after imaging-confirmed total resection (10-30% recurrence rate), in which case, radiation should be delayed.

Stereotactic radiation

Stereotactic radiation has been used primarily as first-line of treatment for rapidly expanding or symptomatic, solid, and small craniopharyngiomas (< 25-30 mm in diameter). Stabilization or reduction of the cystic cavity after radiosurgery is achieved in more than 60% of patients. ^[40]

Stereotactic radiation has also been used for further treatment of residual solid tumor after brachytherapy.

Brachytherapy/radioisotopes

Brachytherapy is recommended for solitary cystic craniopharyngiomas and consists of stereotactic aspiration of cystic content followed by instillation of beta-emitting isotopes (ie, phosphorus 32, rhenium 186, gold 198, yttrium 90).

Brachytherapy is highly feasible because about 60% of craniopharyngiomas occur as single large cysts. Early refilling is common, requiring intermittent aspiration either by stereotactic puncture or Ommaya reservoir.

Intracystic chemotherapy

Intracystic injection of bleomycin ^[41]  and internal irradiation with radioisotopes have been reported to control the tumor cysts, yet numerous side effects have been described. ^[42]

Antibiotic with anti-tumor activity: Bleomycin

Bleomycin is a mixture of glycopeptides extracted from the Streptomyces species. There is ongoing research regarding the utility and toxicity of using intracystic bleomycin, especially in the pediatric population. For nearly 30 years, bleomycin has consistently demonstrated objective tumor response and disease control in 20% to 50% of patients. ^[43] In 2016, a Cochrane database review summarized that a conclusion cannot be made because there is not enough high-quality data regarding this treatment as a whole and especially among kids. ^[44]

In combination with other drugs, chemotherapeutic agents are used frequently and systemically against epithelial tumors. In the early 1970s, bleomycin was shown to effectively inhibit craniopharyngioma tissue growth in vitro. Intracavitary bleomycin reduces cyst size and thickens the cyst wall, facilitating surgical excision of the cystic membrane, which may otherwise fragment at the time of surgery. However, reports of intracystic bleomycin use are limited.

The toxicity of bleomycin depends on the age of the patient and the cumulative dose of the drug. Systemic administration may cause pneumonitis, which can progress to fatal pulmonary fibrosis.

When administered systemically, bleomycin does not produce significant bone marrow toxicity. Toxicity with local administration results from systemic contamination (associated with anaphylactoid reaction, transient fever, nausea, and vomiting) and leakage into surrounding neural tissue.

Fatal outcomes have been reported with leakage, related to diffuse diencephalon and brainstem edema. Transient local toxicity involving the surrounding brain parenchyma may be reversible with high-dose steroids.

Alpha interferon

This is another potential intracystic treatment modality. Few publications in the past showed a good response in some of the cases, but research is still ongoing. The literature states that fatigue is the most frequent side effect and the main limiting factor of alpha interferon treatment. ^[45] The efficacy of alpha interferon against squamous cell carcinoma of the skin, in which it induces apoptosis, is well established. ^[46] Jakacki et al. ^[47]  was the first group to use systemic alpha interferon in the treatment of either recurrent disease or patients with craniopharyngioma that did not respond to conventional therapy. This study was a phase II study with a small cohort of pediatric patients (less than 20 years of age), and they were able to show that for patients that had a predominantly cystic lesion the response to treatment was very good. However, all the patients experienced episodes of fever in the first weeks of treatment, as well as muscle cramps and myalgia, and almost 50% of the patients developed significant signs and symptoms of alpha interferon toxicity, which led to either the interruption of treatment or a reduction in the doses administered. This treatment’s benefit, safety, and long-term efficiency is yet to be determined.

There are a variety of microsurgical and endoscopic approaches that can be applied to craniopharyngiomas. There is an ongoing debate as to weather the neurosurgeon should even attempt a gross total resection, or just perform a biopsy and decompression of the tumor (usually the cyst) followed by a referral for adjuvant radiotherapy. The major surgical approaches to craniopharyngiomas can be summarized into five categories: (1) anterolateral transcranial, (2) midline transcranial, (3) extended endoscopic endonasal, (4) intraventricular, and (5) lateral transcranial. While each approach has its advantages and limitations, an individualized approach tailored to each patient based on multiple factors is crucial in determining the optimal treatment strategy. Nowadays, the best treatment can be achieved in facilities where knowledge and expertise in both microsurgical and endoscopic endonasal techniques are present, and the specific approach or combination can be tailored to each patient. ^[48]

Gross total resection

Gross total surgical resection has traditionally been the treatment of choice for craniopharyngiomas. Note panels A and B in the image below.

Coronal views of T1-weighted MRI for a patient with craniopharyngioma before gross total resection (A) and at postoperative follow-up evaluation (B). There was no sign of tumor recurrence, and the patient was neurologically and endocrinologically intact.

View Media Gallery

There are several surgical approaches, as mentioned above. Different considerations are important when choosing the approach for a specific case, including size of the lesion, extension to nearby anatomical structures (i.e., temporal lobes, 3^rd ventricle, vascular structures, etc.), amount of parenchymal edema, and more.

Local inflammation can lead to tumor adhesion to surrounding vascular structures. Tumor adhesion represents the most common cause of incomplete tumor removal. Fusiform dilatations of large surrounding vessels have been reported after attempts at radical dissection of the tumor capsule due to injury to the vasa vasorum leading to weakening of the adventitia.

For many years, complete resection was considered the treatment of choice for optimal tumor control and lower recurrence rates. More recent studies have suggested that a tissue-sparing (yet aggressive) near-total resection followed by radiotherapy may be a suitable alternative to gross total resection, as the rates of tumor control are similar, but the risk of endocrine and behavioral morbidity is less than with more aggressive surgery. Many investigators have associated very aggressive attempts at total tumor removal with significant endocrinopathies. Permanent diabetes insipidus occurs in 68-75% of adults and 80-93% of children. Panhypopituitarism occurs in 75-100% of patients who undergo resection, and replacement of two or more of the anterior pituitary hormones is necessary in 80-90% patients. Hypothalamic obesity occurs in 40-50% of patients postoperatively.

A list of potential perioperative morbidities includes the following:

• Seizures
• Visual deficits (including blindness)
• Hypothalamic injury
• Stroke
• Cerebrospinal fluid (CSF) leakage

Craniopharyngiomas have a high rate of recurrence, mostly in the first three years after surgery. Overall, recurrence rates range from 0-17% after gross total resection and from 25-63% after subtotal resection with radiotherapy. However, two studies have reported recurrence rates of 53-62% even after apparent complete removal of the tumor. One series assessed only pediatric patients and the other included patients younger than 25 years. Therefore, young age may be a risk factor for tumor recurrence independently of the degree of tumor excision. Ultimately, if left untreated, these recurrences may cause death through aggressive local behavior.

Recent studies propose subtotal resection with postoperative radiotherapy as the management paradigm of choice for craniopharyngiomas, especially in the pediatric population. Goals of this approach include pathologic confirmation of the tumor and surgical decompression of the optic chiasm. Surgery is followed either by proton beam radiotherapy or by external beam radiation, at a dose of 5400-5500 cGy delivered at 180 cGy/fraction. More advanced radiotherapy modalities currently under investigation include Gamma Knife and CyberKnife radiosurgery. ^{[49, 50]}  As mentioned above, in 2014, Bishop and colleagues compared the efficacy and safety of postoperative radiation therapy for two modalities – proton beam therapy (PBT) and intensity-modulated radiation therapy (IMRT). ^[39] In their work, they didn’t find any significant difference between the two groups, although the follow-up time for the PBT was much shorter. Interestingly enough regarding the tumoral cyst, they found that during therapy, 40% of patients had cyst growth (20% requiring intervention), a third of the patients had cyst growth immediately after therapy and that was seen more commonly in the IMRT group. Toxicity did not differ between the two groups. Their final conclusion did not find any significant difference between these two modalities and they recommended that in any case strict follow-up needs to be done in regards to cyst dynamics. 

The incidence of tumor progression after subtotal surgical resection and radiotherapy ranges from 12-25% and is similar to rates associated with failed gross total resection and radiotherapy (4-25%).

Radiotherapy delivered after recurrence (salvage radiotherapy) is effective, with a posttreatment progression rate of 29%. Recurrence following radiotherapy has been associated with a 50-80% mortality rate.

Complete surgical removal of craniopharyngiomas can be achieved with reasonable safety in the majority of patients. Aggressive attempts at total tumor removal may lead to increased rates of anterior hypopituitarism, diabetes insipidus, growth disturbances, and behavioral and feeding abnormalities. On the other hand, subtotal resection with adjuvant radiotherapy can provide tumor control rates essentially similar to those for gross total resection while limiting hypothalamic and hypophyseal morbidity.

For selected patients with suprasellar craniopharyngiomas, an extended endonasal endoscopic approach could provide a viable alternative to transcranial approaches. ^{[51, 29, 52]}

Other approaches that can be useful in the management of giant craniopharyngiomas, especially at the time of recurrence, include (1) intermittent aspiration by stereotactic puncture or Ommaya reservoir placement, (2) intracystic injection of bleomycin, ^[41]  and (3) internal irradiation with radioisotopes. The latter two treatment modalities have been reported to control the tumor cysts in 90-100% of cases.

Another important consideration, especially with suprasellar tumors, is the need for CSF diversion (i.e., ventriculoperitoneal shunt). 

There is data supporting the suspicion clinicians have had for a long time that adamantinomatous and papillary craniopharyngioma (AC and PC, respectively) are different entities and hence may respond to different treatment modalities. Analyses of the craniopharyngioma genome and transcriptome as well as analysis of the DNA methylation patterns of craniopharyngiomas have provided considerable insight into the origin of these tumors and into what drives their growth. ^[48]

Adamantinomatous craniopharyngioma

The main pathway under investigation in this subgroup is that of β-catenin and the WNT/Wingless Pathway. β-catenin is a key member in the WNT pathway. The CTNNB1 gene encodes this protein and it plays a critical role in development, cellular proliferation, differentiation, and cell migration. ^{[53, 54, 55]}

When WNT pathway is activated, an intracellular signaling cascade starts that ultimately prevents formation of the β-catenin destruction complex 31. Without the destruction complex, the β-catenin protein accumulates within the cell, binding another protein called fascin and ultimately changing the genomic transcription and facilitating uncontrolled cellular proliferation. ^{[56, 57]} The accumulation of β-catenin eventually can reduce E-cadherin expression, which may reduce cell adhesion and results in cells that are more motile, which eventually can lead to increased invasive potential. Researchers are looking at the possible inhibition of either fascin or β-catenin accumulation that could potentially lead to re-activation of the destruction complex, preventing the progression of the cell towards uncontrolled proliferation and cell migration.

Nuclear accumulation of β-catenin results from mutations within exon 3 of the CTNNB1 gene. While mutations have been identified at a number of different codons, these all affects the binding of GSK3b. ^{[56, 58, 59, 9]} As a result of this mechanism, nuclear accumulation of β-catenin is a histological hallmark of CTNNB1 mutation. ^{[53, 54, 60]} Interestingly enough, when examining AC for this specific mutation, it is very common and happens at a rate higher than 70% of the cases. ^{[61, 62]} The clinical implications of this knowledge have yet to be discovered and developed.

The Sonic hedgehog (SHH) pathway plays an integral role in the maintenance of adult stem cells and in the normal development of several organs, including the pituitary gland and Rathke’s pouch. It has been linked to different pathologies in the brain including medulloblastoma, basal cell carcinoma, and even meningiomas. ^{[63, 64]} The therapeutic relevance of SHH protein can be significant since it is highly unregulated in AC, even in comparison to other brain tumors, especially in the pediatric population. ^[62]  This raises the possibility of the inhibition of SHH pathway as a clinical tool for treating AC. Preclinical animal studies of the smoothened inhibitor vismodegib are ongoing. ^[65] SHH expression opens up the possibility for better understanding of the difference between AC and PC, since as Hölsken and colleagues ^[66]  demonstrated, there is a significant difference in the expression of SHH protein between them, with AC tumors having high overexpression. The identification of cilia throughout the epithelium of AC also further intensifies the link to the subgroup of tumors of SHH that have cilia as well (i.e., Medulloblastoma). This again is a potential treatment path for AC. ^[67]

Epidermal growth factor has been described in a variety of tumors, as well as the clinical implications for its inhibition. It has been described as factor that promotes cell growth and infiltration. In AC, we can see downstream upregulation of this pathway. This pathway is usually regulated by the epidermal growth factor receptor (EGFR). ^[68] The involvement of EGFR in the regulation of the expression of stem cell markers in AC, and the presence of the activated EGFR pathway in β-catenin accumulating cells, suggests a potential role for inhibition of the cell proliferation and migration through this pathway. As for now, the study is ongoing.

For AC, there is a robust amount of ongoing research as was described for the pathways mentioned above and many others (such as the use of Dasatinib as part of inhibition of tyrosine kinase pathway). ^[65]

Papillary craniopharyngioma

A possible treatment of PC has the potential of changing the role of neurosurgeons in treating this subtype of craniopharyngioma, which usually affects adults. In PC, in contrast to AC, β-catenin localizes to the cell membrane, similar to the pattern of localization in other CTNNB1 wild-type tumors of the sellar region ^{[56, 54]}  and throughout the body. ^{[65, 48]} The most distinct pathway in PC is MAP kinase. Brastianos and colleagues identified the BRAFv600e mutation in 92.8% of PC specimens. ^[69] Later publications found incidence rate to be close to 100%. ^{[70, 71]} This revealed the potential for a diagnostic tool and BRAF inhibitors as a possible effective treatment. BRAF mutation upregulates MAP kinase signaling and propagate cell division and proliferation. This mutation was found to have multiple subtypes in a variety of tumors with the most known being substitution of valine by glutamate at codon number 600, termed the BRAFv600e mutation. In terms of diagnosis, recognition of BRAF v600e mutation in a sellar mass can help differentiate PC from other potential diagnoses. ^{[71, 72, 48]} In a subset of PC, there is a combination of the genomic mutation CTNNB1 and the BRAF mutation. Several publications describe a very good response to BRAF inhibitor (i.e., vemurafenib), MEK inhibitor (trametinib) and RAF inhibitor (dabrafenib). One interesting point regarding the inhibition of either BRAF or MEK/RAF cycle is that, like with gliomas, when the treatment is stopped, the tumor tends to recur and sometimes will not respond again for the same treatment. The significant reduction in the size of the tumors and the cystic component after the treatment raises the possibility of using tool in the future as a neoadjuvant treatment before surgery or as an adjuvant treatment after the first surgery and before a second one if needed.