Impact of site of tumor recurrence upon survival for children with recurrent or progressive medulloblastoma.

Object: The object of this study was to identify prognostic factors for survival among children with recurrent medulloblastoma.

Methods: Postprogression survival and patient, tumor, and treatment factors were examined in 46 cases of recurrent medulloblastoma (mean age of patients at diagnosis 6.5 years, mean age at progression 8.4 years). Differences were calculated by Kaplan-Meier log-rank analysis. Multivariate analysis was performed using the Cox proportional hazard model.

Results: The probability of 5-year survival was 26.3%. Forty-one patients received salvage therapy and five patients received hospice care only. Log-rank analysis showed an association between prolonged patient survival and recurrence limited to the primary site (p = 0.008), initial therapy including the Pediatric Oncology Group (POG) regimen for the treatment of brain tumors in infants ("Baby POG;" p = 0.037), and treatment with radiation therapy (RT) following initial progression (p = 0.015). Cox regression analysis showed a significant association between prolonged survival and only one variable--tumor recurrence restricted to the primary site (p = 0.037). There was no significant association between prolonged survival and any other variables, including patient sex, age at progression, interval from tumor diagnosis to progression, initial tumor stage, and salvage treatment with chemotherapy. Subgroup analysis revealed that site of tumor progression was also prognostic for survival among the subgroup of patients older than 3 years of age at diagnosis who were initially treated with RT and chemotherapy (p = 0.017, log-rank test).

Conclusions: Some children with recurrent medulloblastoma will be long-term survivors, and certain features are associated with likelihood of survival. Patients whose tumors recur at only the primary tumor site have an increased chance of prolonged survival.