Proton Therapy for Pediatric Ependymoma: Mature Results from a Multi-Center Study

Abstract

Report long-term efficacy and toxicity of proton therapy for pediatric ependymoma. Between 2000 and 2017, 318 children with nonmetastatic grade II/III intracranial ependymoma received proton therapy at two academic institutions. Median age was 3.5 years (range, 0.7-21.3 years); 56% were male. Most (69%) tumors were in the posterior fossa and classified as WHO grade III (64%). Eighty-four percent had a gross total or near total tumor resection before radiotherapy and 30% received chemotherapy. Median radiation dose was 55.8 CGE (range, 50.4-59.4 CGE). The Kaplan-Meier product limit method estimated freedom from time-dependent endpoints, and the log-rank test statistic estimated the level of statistical significance between strata of selected prognostic factors. Median follow-up was 6 years (range, 0.6-19.2 years). Seven-year local control, progression-free survival, and overall survival rates were 77.1% (95% CI 71.7-81.7%), 64.4% (95% CI 58.6-69.8%), and 82% (95% CI 76.9-86.2%), respectively. Subtotal resection was associated with inferior local control (60% vs 80%; p<0.01), progression-free survival (49% vs 67%; p<0.01), and overall survival (69% vs 84%; p<0.05). Male gender was associated with inferior progression-free (59% vs 71%; p<0.01) and overall survival (77% vs 89%; p<0.05). Twenty patients (6.2%) required hearing aids; of these, 12/20 received cisplatin, including all three cases with bilateral hearing loss. 44 patients (13.8%) required hormone replacement, most commonly growth hormone (37/44). Grade 3+ brainstem toxicity rate was 1.6% and more often in patients who received >54 CGE. Two patients (0.6%) died of brainstem necrosis. The rate of second malignancy was 0.9%. Proton therapy offers commensurate disease control to modern photon therapy without unexpected toxicity. The high rate of long-term survival justifies efforts to reduce radiation exposure in this young population with brain tumors. Independent of radiation modality, this large series confirms extent of resection as the most important modifiable factor for survival. The pronounced gender disparity in survival deserves greater attention.