The prognostic value of histological grading of posterior fossa ependymomas in children: a Children's Oncology Group study and a review of prognostic factors

Abstract

We performed a retrospective analysis of 96 pediatric posterior fossa ependymomas in order to determine the prognostic value of histological grade based on the current WHO grading scheme. The patients were selected among Children's Oncology Group (previously Pediatric Oncology Group-POG) patients enrolled in clinical trials, and on the basis of central pathology review, location, and age. We excluded entities such as sub-ependymoma, myxopapillary, or clear-cell ependymoma, after a consensus diagnosis by three neuropathologists. A total of 66 males and 30 females with a median age of 48 months were identified. The group was analyzed to determine the effects of histological grade, age, gender, and extent of resection on event-free and overall survival. Our results showed that extent of resection, age, and histological grade were independent prognostic variables for event-free survival. The relative risk for extent of resection and histological grade was calculated as 3.59 (P<0.001) and 3.58 (P<0.001), respectively. Overall survival significantly correlated with extent of resection and age, but not with histological grade. We compared our results with peer-reviewed publications on pediatric intracranial ependymomas in the English language between 1990 and 2005. Selection criteria identified 32 manuscripts involving 1444 patients. Extent of resection was a significant factor in 21, age in 12, and histological grading in nine of these studies. Other factors reported to be significant by more than one study included tumor location and radiation treatment. Our findings suggest that histological grade (WHO Grade II vs III) is an independent prognostic indicator for event-free survival, but may not be so for overall survival in pediatric posterior fossa ependymomas. We believe that an accurate assessment of the prognostic value of histological grade depends on the selection of a well-characterized clinical cohort of sufficient size, and the inclusion of relevant histological criteria as outlined in the WHO classification scheme.