Proton Radiotherapy For High-risk Pediatric Medulloblastoma: Outcomes For Very Young Children Treated With Upfront Chemotherapy

Reporter: Gita Suneja, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: October 30, 2012

Presenter: Rachel Jimenez, MD
Harvard Radiation Oncology Program


  • Medulloblastoma is the most common brain tumor of childhood
  • Appropriate treatment includes maximal surgical resection, followed by adjuvant chemotherapy and radiation therapy.
  • 5-year survival from medulloblastoma approaches 80%, however declines to 50% in young children or those with high risk disease.
  • Radiation therapy is often delayed in young children to minimized long-term toxicity.
  • The purpose of this study is to report early outcomes for very young children with medulloblastoma treated with upfront chemotherapy followed by proton therapy.

Materials and Methods

  • A retrospective review of pediatric patients who received upfront chemotherapy or proton beam radiation for medulloblastoma or supratentorial PNET at Massachusetts General Hospital from 2002 to 2010 was performed.
  • All patients were age < 60 months
  • All patients had maximal surgical resection followed by chemotherapy then radiation.
  • Primary outcomes were local control, overall survival, and treatment toxicity.
  • Local control was assessed clinically at follow-up visits and with MRI every 6 months.
  • Toxicity was assessed clinical at follow-up visits and with bi-annual endocrine evaluation and annual audiology testing.
  • Neuropsychological testing was performed when available.


  • 15 patients were included in the study cohort. 12 had medulloblastoma and 3 had supratentorial PNET.
  • Median age was 36.8 months.
  • 10 of the 15 patients had unfavorable characteristics, such as large cell/anaplastic histology, M2-3 disease, or subtotal resection.
  • The majority had a gross total resection.
  • 14 were treated with chemotherapy as per a COG or HEADSTART regimen.
  • 12 had craniospinal radiation with an involved field boost, while 3 had involved field radiation therapy alone.
  • The median time from surgery to proton therapy was 219 days.
  • The median craniospinal radiation dose was 23.4 GyE, and the median involved field dose was 50.4 GyE.
  • At median follow-up of 3.1 years, 1 non-disease related death and 1 death after local failure was observed. The remaining 13 patients were alive and without evidence of disease.
  • Mean dose to normal structures included: cochlea 20.9-21.4 GyE, hypothalamus 21.2 GyE, pituitary 20.3 GyE.
  • Ototoxicity and endocrinopathy were the most common long-term toxicities.
  • 10 patients developed mild to moderate high frequency sensorineural hearing loss and 2 required hearing aids.
  • 5 patients had height/weight percentiles below normal, and 3 required exogenous growth hormone +/- levothyroxine.
  • Neuropsychiatric testing was available on a subset of 7 patients, and no significant differences were observed between the initial or follow-up testing.

Author's conclusions

  • Maximal safe resection followed by upfront chemotherapy and proton radiation results in excellent local control rates and minimal long-term toxicity for young children with medulloblastoma or supratentorial PNET.

Clinical Implications

  • The neurotoxic effects of radiation therapy are higher among younger children, and are likely due to the exquisite sensitivity of the developing brain.
  • Numerous studies have demonstrated the neurocognitive deficits in patients treated with cranial and craniospinal radiation at a young age.
  • At present, radiation therapy is omitted in children less than 3 years of age, and multi-agent chemotherapy is often used as a substitute until radiotherapy can be administered (Duffner, Neuro-oncology 1999).
  • The physical properties of proton therapy provide an opportunity for reduction in integral dose to the brain.
  • Numerous treatment planning studies demonstrate a clear difference in normal tissue sparing for protons vs. photon therapy.
  • However, limited clinical data has examined whether this difference in integral dose translates to a clinical benefit in young patients.
  • A study of 3 young children treated at Loma Linda demonstrated successful outcomes with proton radiation (Yuh, Cancer J, 2004)
  • This study demonstrates excellent local control and toxicity outcomes in a sample of young patients with medulloblastoma and PNET.
  • Nonetheless, the analysis was done retrospectively and toxicities were not compared with toxicities seen with conventional radiation or those seen with chemotherapy alone.
  • The median follow-up was short, and it is expected that additional late effects of radiation therapy will not manifest until several years following treatment.
  • Finally, neurocognitive deficit represents one of the most debilitating side effects of cranial radiation in young children, and neuropsychological testing was performed in less than half the patients at irregular intervals. More rigorous testing would have enhanced the results.
  • This important paper demonstrates the feasibility of proton radiation therapy for young children with medulloblastoma and supratentorial PNET, and further investigation in phase III studies is warranted.