Immunohistochemical staining techniques are crucial for identifying the neuronal and astrocytic features within these tumors. Positive staining for synaptophysin, neuropeptides, and biogenic amines are associated with a neuronal phenotype. Similarly, positive staining for glial fibrillary acid protein (GFAP) identifies the astrocytic component. Electron microscopy is also helpful to identify additional neuronal features such as dense core granules and synaptic junctions (Hirose et al. 1997; Gelabert-Gonzalez et al. 2010).

Molecular cytogenetic data regarding ganglioglioma are scarce, but a few abnormal karyotypes have been observed. Specific cytogenetic abnormalities include a ring chromosome 1, trisomy of chromosomes 5–7, and deletion of chromosome 6 (Neumann et al. 1993; Xu et al. 2014). Analysis for microsatellite marker instability in tumor DNA from six gangliogliomas found no abnormalities (Zhu et al. 1996). One series of ganglioglioma patients reported a comparatively higher frequency of splice-site-associated single-nucleotide polymorphism in the tuberous sclerosis 2 gene (TSC2) (Platten et al. 1997). This may suggest an underlying genetic susceptibility for sporadic ganglioglioma, although the underlying biologic mechanism is unknown. More recently described variants, papillary glioneuronal tumor and rosette-forming glioneuronal tumor (WHO grade I), have been mostly described in adults, although a recent systematic review also identified a number of children with these tumors (Schlamann et al. 2014). Chromosomal and structural alterations involving only chromosome 7 with breakpoints at 7p22 have been reported in the papillary glioneuronal tumor variant (Faria et al. 2008).

Seizure is the most common presenting symptom of gangliogliomas. The seizure history is often longstanding, with a mean duration prior to diagnosis ranging from 6 to 25 years (Prayson et al. 1995; Luyken et al. 2003; Southwell et al. 2012). In one series of patients, temporal lobe gangliogliomas represent 40 % of all tumors causing chronic temporal lobe epilepsy (Blumcke et al. 1999). Patients with brain stem lesions commonly present with involvement of the motor tracts: weakness, spasticity, and gait disturbance (Gopalakrishnan et al. 2013). Gangliogliomas of the spinal cord may involve the entire spinal cord and typically produce scoliosis, gait disturbance, and progressive weakness (Hamburger et al. 1997; Jallo et al. 2004). These symptoms can be very longstanding before a diagnosis is reached. Patients with midline tumors may develop symptoms and signs of hydrocephalus, such as headache, papilledema, alterations in the level of consciousness, and nausea/vomiting (Haddad et al. 1992; Deling et al. 2013; Zhang et al. 2013a).

Gangliogliomas are indolent, slow-growing tumors. Without resection, patients often have prolonged courses of disease, depending on the location of the primary mass (Selch et al. 1998). Anaplastic glial changes in ganglioglioma, as well as high MIB-1 labeling indices, may be markers for more aggressive tumor behavior (Hirose et al. 1997; Scoccianti et al. 2012). Malignant transformation is relatively rare, occurring in less than 3 % of gangliogliomas (Hakim et al. 1997; DeMarchi et al. 2011).

Neuroimaging is the initial step in diagnosis, as no specific laboratory tests are available. A CT scan, if performed as a screening test, reveals an iso- to hypodense solid or cystic mass (Adachi and Yagishita 2008). Cysts may be associated with a mural nodule, although both cyst and nodule are well circumscribed. Calcifications may be present and contrast enhancement is usually seen, but occasionally can be minimal or absent (Lagares et al. 2001). MRI is the best imaging modality and is required to adequately delineate the mass. Tumors are usually hypointense on T1-weighted images and hyperintense on T2-weighted images (Zhang et al. 2008). Mass effect and edema are typically minimal. Contrast enhancement varies in intensity and may be nodular, rim-like, or entirely solid (Figs. 8.1a and 8.2). Syringobulbia or syringomyelia can be seen with spinal cord gangliogliomas (Park et al. 1993; Hamburger et al. 1997; Jallo et al. 2004). MR spectroscopy is usually of limited value due to the indolent nature of these tumors.

Complete surgical resection is the treatment of choice and when achievable is usually curative (Sutton et al. 1987; Compton et al. 2012). The neoplasm itself contains no functioning nervous tissue, but potential eloquence of surrounding parenchyma must be considered in surgical planning (Southwell et al. 2012). A postoperative MRI is useful to assess the extent of resection. Subsequent surveillance imaging should be done to evaluate for recurrence, which can occur in a small percentage of patients. Radiation therapy should be considered for tumor recurrence when further resection is not feasible. Tumors with malignant features (anaplastic features, high MIB-1 labeling index) may require radiation therapy as an adjuvant therapy, regardless of the extent of resection (Hakim et al. 1997; DeMarchi et al. 2011). Radiation therapy for benign lesions may delay time to progression in patients with unresectable disease, but the impact of radiotherapy on progression-free survival for incompletely resected benign tumors remains uncertain (Lang et al. 1993; Compton et al. 2012). Because the role of radiation therapy for subtotally resected, low-grade tumors is unclear, the risks and benefits of radiotherapy should be carefully weighed.

The prognosis following GTR is excellent (Khajavi et al. 1995; Compton et al. 2012; Englot et al. 2012a; Southwell et al. 2012). In one surgical series of 88 patients with a median follow-up of nearly 12 years, 15-year overall survival was 94 % and 10-year progression-free survival was 37 %, with progression being dramatically affected by the extent of the initial resection (Compton et al. 2012). Tumor location is a significant predictor of outcome, most likely because it predicts resectability. Patients who undergo a subtotal excision (STR), most commonly seen in patients with midline tumors, are at higher risk of tumor progression or recurrence (Haddad et al. 1992; Compton et al. 2012). The importance of anaplasia as a prognostic feature is unclear, with different series demonstrating conflicting results (Kalyan-Raman and Olivero 1987; Lang et al. 1993; DeMarchi et al. 2011; Selvanathan et al. 2011). A retrospective analysis in one series of 34 patients, however, did demonstrate a correlation between improved survival and degree of resection as well as tumor grade (Selch et al. 1998). In a large series reported by Luyken et al., the rate of 7.5-year, progression-free survival was 97 % (Luyken et al. 2003). Risk factors for recurrence or malignant progression were residual tumor, frontal tumor location, and a higher-grade lesion. The survival outcomes are also acceptable for gangliogliomas involving the posterior fossa (Baussard et al. 2007) or spinal cord (Jallo et al. 2004).

For patients with tumor-associated epilepsy, seizure control improves dramatically after tumor resection (Englot et al. 2012a; Wallace et al. 2013). GTR appears to be the most important treatment-related factor (Benifla et al. 2006; Giulioni et al. 2006; Park et al. 2008; Southwell et al. 2012). In a recent series of 66 patients with ganglioglioma, 49 of which presented with epilepsy, and 85 % of patients with a seizure history were free of seizures after surgery (Southwell et al. 2012). Ninety-six percent of individuals in whom GTR was achieved were seizure-free, but only 54 % of the group had STR. A recent systematic review and meta-analysis found that postoperative seizure freedom in glioneuronal tumor resection was predicted by GTR, the absence of generalized seizures, and a shorter history of epilepsy (Englot et al. 2012a). Because seizure outcomes are improved in those patients with a shorter duration of epilepsy, early surgical intervention should be considered, particularly in patients with medically refractory epilepsy.

Dysembryoplastic neuroepithelial tumor (DNET) is a benign glioneuronal neoplasm that most commonly occurs in the supratentorial compartment. It was first described by Daumas-Duport and Scheithauer in 1988 (Daumas-Duport et al. 1988). The initial report described 39 children with a morphologically distinct brain tumor and intractable partial seizures.