Subsequent neoplasms (SNs) of the central nervous system (CNS) in survivors of childhood cancer: a case/control study from the Children's Cancer Survivor Study (CCSS)

Reviewer: Walter Sall, MD
Last Modified: June 1, 2003

Presenter: J.P. Neglia
Presenter's Affiliation: University of Minnesota
Type of Session: Scientific


  • Childhood cancer patients survivors are known to be at increased risk of subsequent CNS malignancies.
  • An association between subsequent meningiomas and previous irradiation has been shown previously.
  • A 23-fold increased risk of CNS cancer among survivors of childhood ALL has been seen.
  • It has been postulated that a deficiency of thiopurine metabolism may lead to increased subsequent CNS malignancy risk.
  • This study intended to help define risk factors for SNs of the CNS among childhod cancer survivors.

Materials and Methods

  • The CCSS is a multi-institutional cohort study of childhood cancer survivors.
  • Eligible patients were diagnosed between 1970 and 1986, were under age 21 at diagnosis and survived at least 5 years.
  • For this study, 70 patients with SNs of the CNS were identified. These were compared to 280 age, sex and survival matched controls.
  • Chemotherapy doses and RT doses to the site of origin of secondary CNS malignancies were extracted from medical records.
  • Cases constituted 35 meningiomas, 28 gliomas, 4 PNETs and 1 lymphoma with 2 of unidentified histology.


  • Cases and controls were well matched except for type of initial diagnosis: 54% of cases vs. 37% of controls were ALL survivors.
  • 96% of cases vs. 70% of controls received radiation as part of their initial therapy.
  • No association between SNs of the CNS and chemotherapy dose was seen. A trend towards inreased risk of meningioma with platinum was seen.

Author's Conclusions

  • There appears to be an association between radiation dose and risk of SN of the CNS. The odds ration for SN of the CNS peaks at 54.6 for radiation doses of 3000-4499cGy
  • No association between chemotherapy and SNs was seen
  • Re-analysis with an additional 55 patients is currently underway to clarify these issues.

Clinical/Scientific Implications

    The possible association between original diagnosis and risk of SN of the CNS was not investigated in this study. The arms were not balanced with respect to original diagnosis which could have skewed the results. This imbalance could have led to the association between radiation dose and SNs that was seen. The small patient numbers and heterogeneity of chemotherapy regimens made analysis of chemotherapy differences between cases and controls difficult to analyze. This could have hidden a true association between chemotherapy and risk of SNs of the CNS. The addition of greater numbers of patients in this study will hopefully clarify these issues. A constellation of factors, including radiation, chemotherapy and genetic abnormalities likely leads to the greateer risk of SNs of the CNS in childhood cancer survivors.

Oncolink's ASCO Coverage made possible by an unrestricted Educational Grant from Bristol-Myers Squibb Oncology.