Outcome of infants and young children with newly diagnosed medulloblastoma treated on Head Start III protocol

Reporter: Gita Suneja, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 5, 2011

Presenter: Girish Dhall, MD
Presenter's Affiliation: Children’s Hospital Los Angeles


  • The standard of care for newly diagnosed medulloblastoma is tri-modality therapy with surgery, chemotherapy, and radiation all playing a role in both standard-risk and high-risk disease.
    • Craniospinal radiation is indicated for all patients, with dose varying according to Chang staging and ranging from 2340 cGy to 3600 cGy:
      • M0: Tumor limited to posterior fossa
      • M1: Tumor cells present in CSF without radiologic evidence of dissemination
      • M2: Tumor seeding within the brain
      • M3: Tumor seeding within the spinal column
      • M4: Metastatic disease outside the craniospinal axis
  • Treatment of very young patients is particularly challenging due to concern over late effects of radiation therapy, and therefore radiation is typically deferred in children less than 3 years of age.
  • The Head Start trials are a group of studies that explore the use of high dose chemotherapy with autologous hematopoietic cell rescue (AuHCR) as a substitute for craniospinal radiation (CSI) in children under the age of 6 years.
  • Head Start I accrued patients from 1991-1997 and Head Start II from 1997-2003. After initial surgery, patients received five cycles of induction chemotherapy consisting of vincristine, cisplatin, cyclophosphamide and etoposide. Head Start II included methotrexate with each cycle. Following induction, all patients underwent myeloablative chemotherapy using carboplatin, thiotepa and etoposide with AuHCR. Irradiation was used only at relapse. This treatment paradigm eliminated the need for CSI in about 50% of young children with non-metastatic medulloblastoma however, the mortality due to acute toxicity was 20%.
  • Head Start III was developed with the goal of determining whether high dose chemotherapy with AuHCR delivered over a short period of time (< 6 months) without radiation therapy is a feasible approach in the management of children less than 6 years of age with medulloblastoma who had a complete response to initial treatment.


Head Start III was a multi-institiutional, multi-national prospective clinical trial involving 39 institutions.

  • 92 patients with newly diagnosed medulloblastoma were enrolled between April 2003 and December 2009.
  • Schema:
    • All children underwent maximal safe resection of the primary tumor.
    • Induction
      • Cycles 1, 3, and 5 were the same as in Head Start II: vincristine, cisplatin, cyclophosphamide, etoposide, and high-dose methotrexate
      • Cycles 2 and 4 were vincristine, cyclophosphamide, oral etoposide, and oral temozolomide.
    • Consolidation: 1 cycle of myeloablative chemotherapy (thiotepa, carboplatin, etoposide) and AuHCR
    • Only children between 6 -10 years old or with residual tumor pre-consolidation were eligible to received radiation after consolidation
  • Radiation guidelines:
    • Children < 6 yrs, initially M0, but with localized disease remaining after chemotherapy à CSI to 1800 cGy
    • Children < 6 yrs, initially M1, but with localized or disseminated disease after chemotherapy à CSI to 2340 cGy.
    • All children > 6 yrs à CSI to 2340 cGy.
  • The primary objective was to determine 2 yr event-free survival (EFS) and overall survival (OS) for children with standard risk disease under 6 years of age, and high-risk disease under 10 years of age.
  • Standard risk was defined as non-disseminated (M0) disease with gross total resection (R0)
  • High risk disease was defined either by disseminated disease (M1) or incomplete resection (R1)


2-yr EFS (%)

2-yr OS (%)

All patients



M0 (n=39)



M1 (n=51)



3-yr EFS (%)

3-yr OS (%)


All desmoplastic



All classical



All anaplastic




M0 desmoplastic



M0 classical



M0 anaplastic




M1 desmoplastic



M1 classical



M1 anaplastic



  • 2 toxic deaths were observed due to high dose chemotherapy
  • 13 patients received radiation per protocol guidelines, and no protocol violations occurred
  • 3-year radiation-free EFS was 49% for all patients
  • In multivariate Cox regression analyses, histology was the only significant independent predictor of EFS after adjusting for metastatic status, initial extent of resection of the primary tumor, regimen, age and gender. Patients with desmoplastic histology had significantly better EFS than patients with classic tumors (HR = 0.11, 95% CI = 0.033, 0.35). Outcomes for patients with anaplastic and classical tumors were not significantly different from one another.

Authors’ Conclusions

  • 52% of young patients under 6 years of age avoided CSI when treated with the Head Start III regimen.
  • The outcomes presented are the best survival data published on young children with medulloblastoma not receiving radiation therapy.


  • This study establishes the feasibility of eliminating radiation therapy in a cohort of patients under 6 years of age with standard risk medulloblastoma and complete response to initial therapy.
  • Event-free survival and overall survival data suggests that this technique is feasible and safe.
  • Avoidance of CSI may reduce late effects such as decreased height, neurocognitive deficits, and late solid organ dysfunction; however, these potential risks must be balanced against cancer-related outcomes. Outcomes for M0 patients with classical and anaplastic histologies in this study appear to be poorer than would be expected for M0 patients treated with CSI.
  • The number of high-dose chemotherapy treatment-related toxicities was far lower than seen in Head Start I and II.
  • The data presented represents the best survival data published to date for children with medulloblastoma for whom radiation is deferred. This is particularly impressive considering that 60% of the patients in this cohort had disseminated disease at presentation. This is in contrast to other similar studies where approximately 30% of patients had M1 disease at presentation.
  • Although these results are promising, prospective randomized controlled trials are needed to validate the results of Head Start I, II, and III. Direct comparison of survival and other outcomes for children treated with up front CSI versus deferred or delayed RT would be of great interest.
  • The potential of limited field radiotherapy to the posterior fossa only may also be of interest for very young children with medulloblastoma. This type of treatment has potential to decrease local failures while avoiding many of the sequelae associated with CSI.