Do children and adults differ in survival from medulloblastoma? A study from the SEER registry

Studies investigating whether adults have diminished survival from medulloblastoma (MB) compared with children have yielded conflicting results. We sought to determine in a population-based registry whether adults and children with MB differ in survival, and to examine whether dissimilar use of chemotherapy might contribute to any disparity. 1,226 MB subjects were identified using the Surveillance Epidemiology and End Results (SEER-9) registry (1973-2002) and survival analysis performed. MB was defined strictly to exclude non-cerebellar primitive neuro-ectodermal tumors. Patients were stratified by age at diagnosis: < 3 years (infants), 3-17 years (children) and a parts per thousand yen18 years (adults). Because the SEER-9 registry lacks treatment data, a subset of 142 patients were identified using the San Francisco-Oakland SEER registry (1988-2003) and additional analyses performed. There was no significant difference in survival between children and adults with MB in either the SEER-9 (P = 0.17) or SFO (P = 0.89) cohorts but infants fared worse compared to both children (P < 0.01) and adults (P < 0.01). In the SFO sample, children and adults who received chemotherapy plus radiation therapy (XRT) did not differ in survival. Among patients treated with XRT alone, children showed increased survival (P = 0.04) compared with adults. Children and adults with MB do not differ with respect to overall survival, yet infants fare significantly worse. For children and adults with MB treated with both XRT and chemotherapy, we could not demonstrate a survival difference. Similar outcomes between adult and childhood MB may justify inclusion of adults in pediatric cooperative trials for MB.