Reflection and Reaction

compared with 73% in the SFOP study. It is noteworthy that, in both studies, the proportion of patients with favourable desmoplastic histology was high—GPOH, 47%, SFOP, 33%, with classic non-desmoplastic medulloblastoma accounting for the remainder. Whether this high proportion of low-risk histological subtypes is characteristic of the age-group is unclear. It was a strong predictor of survival in the GPOH, but not the SFOP, study. These two published studies report different brainsparing approaches to therapy with similar outcomes for the completely resected, non-metastatic group, in small national cohorts; other similarly-sized studies are in progress in the USA. These results require confirmation and testing in a randomised comparison. Because quality of survival is the main rationale for testing the selected treatment approaches, it should be proposed as a primary outcome measure along with overall survival. Its measurement in an international study would require the use of a feasible, standardised, questionnaire methodology, applicable in multiple languages. Such techniques have already been proposed and piloted for current International Society of Paediatric Oncology (SIOP) brain-tumour trials in Europe. 8–11 Such a trial would also test brain-sparing treatment approaches in older children with medulloblastoma. An international randomised phase III trial is an essential next step and is under discussion in Europe and USA.