Radiation therapy quality in CCG/POG intergroup 9961: implications for craniospinal irradiation and the posterior fossa boost in future medulloblastoma trials

Bernadine Donahue, Mary A. H. Marymont, Sandra Kessel, Matthew K. Iandoli, Thomas FitzGerald, Emiko Holmes, Mehmet Kocak, James M. Boyett, Amar Gajjar, Roger J. Packer

Purpose: Associations of radiation therapy (RT) deviations and outcomes in medulloblastoma have not been defined well, particularly in the era of reduced-dose craniospinal irradiation and chemotherapy. The aim of this study is to evaluate the quality of RT on Children's Cancer Group/Pediatric Oncology Group 9961 and analyze associations of RT deviations with outcome. Materials and Methods: Major volume deviations were assessed based on the distance from specified anatomical region to field edge. We investigated associations of RT deviations with progression-free survival (PFS), overall survival (OS), and explored associations with demographics and clinical variables. Results: Of the 308 patients who were evaluable for volume deviations, 101 patients (33%) did not have any. Of the remaining 207 patients, 50% had only minor deviations, 29% had only major deviations, and 21% had both minor and major deviations. Of the patients with major deviations, 73% had a single major deviation. The most common major deviation was in the cribriform plate region, followed by the posterior fossa (PF); PF deviations resulted from treating less than whole PF. There were no significant differences in PFS or OS between patients with deviations and those without. There was no evidence of associations of deviations with patient age. Conclusions: Approximately one-third of patients had major volume deviations. There was no evidence of a significant association between these and outcome. This lack of correlation likely reflects the current high quality of RT delivered in Children's Oncology Group institutions, our strict definition of volume deviations, and the relatively few instances of multiple major deviations in individual patients. In is noteworthy that the types of PF volume deviations observed in this study were not adversely associated with outcome. As we move forward, quality assurance will continue to play an important role to ensure that deviations on study do not influence study outcome.