PURPOSE: Proton therapy is the preferred treatment modality for most pediatric central nervous system tumors. The risk of radiation necrosis may be increased at the distal end of the beam because of an increase in linear energy transfer (LET) and relative biological effectiveness (RBE) dose. We report on the association of LET and dose with radiation necrosis after pencil beam scanning proton therapy in pediatric posterior fossa tumors using a case-control framework. MATERIALS AND METHODS: From 2019 to 2022, 33 patients less than or equal to 18 years of age treated with first-line proton therapy for primary tumors in the posterior fossa and with 6 or more months of follow-up magnetic resonance imaging were retrospectively identified. Nine patients with imaging changes consistent with necrosis were matched with controls in a 1:2 fashion based on age, sex, dose, and follow-up time from proton therapy. Dose (Gy [RBE]) and dose-averaged LET (LETd) values for target structures and organs at risk were computed and compared between cases and controls. RESULTS: Within the whole cohort, the mean age was 6.6 years (SD, 4.77) with a median follow-up time of 24.1 months. Within the case-control matched cohort (18 controls and 9 cases), there were no significant differences in age, sex, time to follow-up, tumor location, dose, and use of concurrent chemotherapy. The mean time to necrotic imaging finding was 4.47 months (SD, 2.03). Cases demonstrated significantly higher brainstem D50 (P = .02). LETd was not different between cases and controls. However, when using a combined metric of higher brainstem dose {>
47.5 (Gy [RBE])} and higher LETd (>
3.5 keV/µm), a greater proportion of cases compared with controls met this metric (89% vs 39%, P = .02). CONCLUSIONS: Combined effects of intermediate-to-high dose and LETd in the brainstem may contribute to greater necrosis risk after pencil beam scanning proton therapy in children with posterior fossa tumors.