Variability in Proton Craniospinal Irradiation Practices in the United States: A Pediatric Proton Consortium Registry (PPCR) Study

Abstract

Craniospinal irradiation (CSI) is commonly used for pediatric brain tumors with propensity for spread in the CSF, principally medulloblastoma. Evolving technology has led to the use of highly conformal radiotherapy (RT) techniques for CSI, including proton therapy. Target delineation and plan coverage are therefore critical for CSI, but there is ongoing controversy and variability in these realms, with little available data on practice patterns. We sought to characterize proton CSI practice patterns in the US by examining CSI plans in the Pediatric Proton Consortium Registry (PPCR).We queried the PPCR for all proton CSI data from 2015-2020. Each plan was manually reviewed for: prescription dose; coverage of optic nerves, vertebral bodies, lateral spinal nerve roots, and sacral nerves; esophageal mean and max doses; and lens max doses. Coverage was assessed at the 95% prescription isodose line. We determined the rate of vertebral body sparing (VBS), as defined by SIOP guidelines, in skeletally immature patients (boys < 16y, girls < 14y).PPCR supplied data for 451 patients; 317 had complete RT data (208 males and 109 females). Medulloblastoma was the most common diagnosis (217), followed by germ cell (32). The median age was 8.8y (IQR 5.8 - 12.4). The median CSI dose was 23.4 Gy (range 10.8 - 45 Gy). Most patients were treated prone (95.6%). Optic nerves were covered completely in 152 cases (47.9%). The average max dose to lenses was 79.1% of prescription (22.6 Gy) vs. 38.2% (10.6 Gy) depending on optic coverage (P < 0.01). Four plans (1.3%) failed to cover the lateral spinal nerve roots. Thirty-nine cases (12.3%) did not completely cover sacral nerves. VBs were spared in 87 cases (27.4%), and covered fully in 138 (43.5%). The remaining 92 plans (29.0%) covered the anterior VBs to a lower dose or utilized anterior avoidance (intermediate sparing). Of 277 plans for skeletally immature patients, 52 (18.8%) were VBS. Both VBS and intermediate sparing significantly reduced mean and max esophageal doses compared to full coverage (P < 0.01). VBS rates by institution (n = 13) varied significantly, from 0 to 71.4%. Rates also increased over time (5.8% in 2015-16; 36.8% in 2017-20).There is significant variability in proton CSI practices nationally, with regard to optic nerve, sacral nerve, and VB coverage. VBS was associated with reduced esophageal doses, and this CSI technique has been correlated in clinical studies to reduced GI and hematologic toxicity. There is growing evidence that VBS may not result in clinically significant growth abnormality, and an ongoing phase II trial is studying this. As evidence evolves, the promulgation of clear guidelines is recommended to optimize and standardize target delineation and planning of CSI.