Final Diagnosis -- Glioblastoma Multiforme



Brain tumors are the leading cause of cancer deaths in children accounting for 20% of pediatric cancers. Gliomas contributing as much as 60% of this group (6). High grade gliomas include glioblastoma and anaplastic astrocytoma. Glioblastomas can occur in any area of the brain. In children, the brainstem is the most common site and the cerebellum the least. Clinical presentation is notable for an evolution of symptoms over weeks to months, consistent with a high grade lesion. The presenting symptoms are variable and related to the location of the tumor. Most commonly they include seizures, visual field defects, hemiparesis, headaches, and increased intracranial pressure. Very young children may have nonspecific symptoms such as irritability, macrocephaly, failure to thrive, and lethargy (6). Glioblastomas are generally de novo lesions in children, especially in young children as in this case. One of the few definitive environmental risk factors associated with increased risk of secondary brain tumor in children is therapeutic radiation exposure. The resulting secondary brain tumors may present as early as 7-9 years after exposure or up to decades later (4).

The genetics of de novo glioblastoma in children are notably different from adult tumors. These tumors tend to have p53 inactivation and no significant epidermal growth factor receptor (EGFR) amplification. Although multiple genetic alterations have been implicated, these also involve the p53 pathway. One study of deep-seated glioblastomas showed all 9 cases with p53 mutations (3). A larger study of 54 patients with tumors from different anatomical sites had p53 immunoreactivity in 54% of cases, with EGFR overexpression in only 26%. Tumors showing p53 mutations were most frequently thalamic, less often cerebral lobar, and least often brainstem in location (2).

Similar to tumors in adults, glioblastomas in children have a highly varied morphology. The tumor cells themselves are highly varied in morphology. Their appearance ranges from pleomorphic cells with or without glassy cytoplasm to small, uniform cells to giant cell forms with lobed or multiple nuclei (1). Interestingly, this tumor had a prominent myxoid component which has been reported (1). Immunohistochemical staining is essential to confirm the tumor cells are GFAP-positive and to determine proliferation rate with Ki-67.

Prognosis for glioblastoma is poor despite advances in diagnosis and treatment, with less than 20% survival at 5 years. The Ki-67 rate has been shown to be a strong predictor of outcomes. A study of 98 patients showed a worse prognosis with increasing Ki-67 labeling. The poorest 5-year survival, approximately 11%, was seen in tumors with more than 36% Ki-67 labeling (5). Unfortunately, the present case had at least 50% Ki-67 labeling and followed the predicted aggressive course of a glioblastoma.

Multiple studies comparing treatment with surgery, radiation, or chemotherapy, have shown children undergoing radical tumor resection have had a statistically significant improvement in progression-free survival. A major complicating factor in radical resection, however, is that glioblastomas often have invaded the critical areas of the cerebrum, limiting patient candidacy. Radiation offers modest prolongation of survival, but is not indicated for children less than 3 years of age due to unacceptable effects on neurological development. As surgery and radiation are rarely curative, experimental chemotherapeutic trials with a variety of combination regimens are being studied. Currently, no therapeutic regimen has yet shown to improve outcomes (6).


  1. Burger PC, Scheithauer BW, Vogel FS (2002). Surgical pathology of the central nervous system and its coverings. Fourth ed: 184-187.
  2. Ganigi PM, Santosh V, Anandh B, Chandramouli BA, Sastry Kolluri VR (2005). Expression of p53, EGFR, pRb, and bcl-2 proteins in pediatric glioblastoma multiforme: a study of 54 patients. Pediatr Neurosurg 41:292-299.
  3. Hayashi Y, Yamashita J, Watanabe T (2004). Molecular genetic analysis of deep-seated gliobalstomas. Cancer Genet Cytogenet 153(1):64-68.
  4. Ohgaki H and Kleihues P (2005). Epidemiology and etiology of gliomas. Acta Neuropathol 109:93-108.
  5. Pollack IF, Hamilton RL, Burnham J, Holmes EJ, Finkelstein SD, Sposto R, Yates AJ, Boyett JM, Finlay JL (2002). Impact of proliferation index on outcome in childhood malignant gliomas: results in a multi-institutional cohort. Neurosurgery 50:1238-1245.
  6. Reddy AT, Wellons JC (2003). Pediatric high grade gliomas. Cancer J 9(2):107-12.

Contributed by: Priya Banerjee, MD and Barbara Crain, MD PhD

Case IndexCME Case StudiesFeedbackHome