Peds Brain Tumors

Epidemiology and Basic Background

  • 1500 children diagnosed yearly in U.S.
  • 80% occur within first 10 years of life
  • 60% occur in the posterior fossa, the “back vault” of the skull below the tentorium (i.e., cerebellum and brainstem). This is very different from adults, who typically get supratentorial tumors.
  • 5-year survival now 50%, though this varies based on age, tumor type, tumor location
  • Low-grade astrocytomas, which can often be treated by surgery alone, have excellent long-term survival if entire visible tumor is removed (i.e., “gross total resection” – 90% survival at as long as 10-20 year follow-ups). Can be deadly if they are intrinsic in the brainstem and thus not amenable to surgical removal.
  • High grade gliomas (e.g., glioblastoma muliforme, are often deadly and require aggressive treatments). Also, even Primitive Neuroectodermal Tumors (PNETs) are often more malignant, and require more aggressive treatments. For example, “standard risk” medulloblastomas have a 5-year survival rate of 50-80% after surgery, chemotherapy and radiotherapy.

Treatment Options and Neuropsych Effects


  • Biopsy, shunt/stent placement (in the event of acute obstructive hydrocephalus), or resection (removal of the tumor or part of the tumor)
  • Surgery is fairly ubiquitous in brain tumors
  • Neuropsych effects
    • No consistent relationship between percent of tumor resected and cognitive outcome
    • Surgical approach may affect outcome (techniques getting better and better every year)
    • Surgical complications are associated with poor outcome
    • Surgery may increase risk for cognitive deficits, but only slightly. Many kids with posterior fossa low-grade astrocytomas do great after only surgical removal.


  • Antineoplastic, steroid, others (e.g., AEDs). Note that many people think only of the antineoplastic drugs as “chemotherapy”, but in fact these kids are often put on many meds with potential cognitive effects.
  • Most administered into bloodstream, but Methotrexate is intrathecally administered (i.e., directly into CSF).
  • Neuropsych effects
    • Possible short-term effects include drops in fine motor speed/coordination and performance on drawing/copying tasks. This may be due to an at least partially reversible peripheral neuropathy.
    • Long-term cognitive effects are clearly less marked than radiation, although the variety of chemotherapy agents, combinations, and delivery systems makes generalizations difficult. There is some evidence of long-term effects on reaction time for several chemo agents.
  • Intrathecal methotrexate accentuates the long-term effects of radiation on cognition.
    • Though not typically considered “chemotherapy”, steroids are often administered. There is some evidence that long-term steroid use can affect reading & math performance, immediate attention span, and visual-constructional/drawing skills.


  • Craniospinal (hitting the whole CNS axis). Helps reduce metastases (including those in high-risk, non-CNS cancers such as leukemia) while also hitting tumor site.
  • Focal (hitting the tumor site from multiple angles to minimize exposure to healthy tissue)
  • Craniospinal with local boost (combine lowered-dose craniospinal with a focal hit to the tumor site). In this case, the craniospinal aspect is meant to head off metastasis.
  • Stereotactic radiosurgery (aka “gamma knife”; very focal with precise boundary measurements). Not as useful if tumor margins are not distinct on imaging or if there is a risk for metastasis.

Neuropsych effects

  • Long-term declines in IQ can result from whole-brain radiation, at a rate of 5-7 FSIQ points/year. This is different from adults.
    • Decline has delayed onset (1-2 years?), and its not clear when decline stops (2 yrs post? 10?)
    • Whole-brain radiation is more deleterious to younger kids’ IQ than to older kids’ IQ (see graph below from Radcliffe et al., 1994). This is a uniquely developmental phenomenon.
  • Relationship between radiation dose and IQ decline is unclear, though there is evidence that reducing the dose to the whole brain and using a focal boost to the tumor has a better outcome
  • No single profile of neuropsychological effects, though the following are often affected:
    • Memory
    • Attention
    • Fine motor speed and coordination
    • Mathematics skills
    • Visual-spatial & Visual construction skills
    • Mechanisms for cognitive morbidity include vasculo-occlusive changes, white matter degeneration/poor myelination, and subsequent skill stagnation that leads to IQ decline.

Other Medical Factors Affecting Neuropsych Outcome

  • Supratentorial tumors in the cerebral hemispheres may be associated with greater long-term impairment than midline and infratentorial tumors
  • Original tumor size and residual tumor volume not consistently associated with neuropsych outcome
  • Recency of symptom onset may be predictive of cognitive outcome, as recent symptom onset and rapid decline may be proxies for more aggressive tumors that will require more aggressive treatments.
  • Cranial nerve and motor impairments can affect functional status.
  • Treated acute obstructive hydrocephalus is not related to eventual cognitive outcome (differs from chronic hydrocephalus)
  • No research yet on the cognitive impact of other sequelae of increased intracranial pressure
  • Though common (esp. in hemispheric tumors), the effect of seizures themselves on inter-ictal functioning is unclear. However, AEDs can carry cognitive side effects
  • Also common, treated endocrine dysfunction seems not to have a measurable cognitive effect.