Clinical Outcomes with Varying Proton Craniospinal Irradiation Practices: A Pediatric Proton Consortium Registry (PPCR) Study

Committee/Faculty

Presenting Author(s)

MC

Michael Connor, n/a

MD
UCSD
La Jolla, California, United States

Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.

Co-Author(s)

Ad

Arnold dela Cruz Paulino

MD
Department of Radiation Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX
Houston, Texas, United States

Disclosure: Disclosure information not submitted.
RE

Ralph Ermoian, n/a

MD
Department of Radiation Oncology, University of Washington, Seattle, WA
Seattle, Washington, United States

Disclosure: Disclosure information not submitted.
WH

William Hartsell, n/a

MD
Northwestern Medicine Chicago Proton Center, Warrenville, IL
Warrenville, Illinois, United States

Disclosure: Disclosure information not submitted.
DI

Daniel Indelicao, n/a

MD
Department of Radiation Oncology, University of Florida, Jacksonville, FL
Gainesville, Florida, United States

Disclosure: Disclosure information not submitted.
SP

Stephanie Perkins, n/a

MD
Washington University School of Medicine, Department of Radiation Oncology, St. Louis, MO
St. Louis, Missouri, United States

Disclosure: Disclosure information not submitted.
VM

Victor Mangona, n/a

MD
Texas Center for Proton Therapy, Texas Oncology, Irving, TX
Irving, Texas, United States

Disclosure: Disclosure information not submitted.
ND

Nicholas DeNunzio, n/a

MD
John Theurer Cancer Center at Hackensack University Medical Center, Hackensack, NJ
Hackensack, New Jersey, United States

Disclosure: Disclosure information not submitted.
NL

Nadia Laack, n/a

MD, MS
Mayo Clinic, Rochester, MN
Rochester, Minnesota, United States

Disclosure: Disclosure information not submitted.
CH

Christine Hill-Kayser, n/a

MD, MS
Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA
Philadelphia, Pennsylvania, United States

Disclosure: Disclosure information not submitted.
YK

Young Kwok, n/a

MD, MS
Department of Radiation Oncology, University of Maryland Medical System, Baltimore, MD
Baltimore, Maryland, United States

Disclosure: Disclosure information not submitted.
JH

John Han-Chih Chang, n/a

MD
The Oklahoma Proton Center and University of Oklahoma Health Science Center, Oklahoma City, OK
Nashville, Tennessee, United States

Disclosure: Disclosure information not submitted.
TY

Torunn Yock, n/a

MD, MCH
Massachusetts General Hospital
Boston, Massachusetts, United States

Disclosure: Disclosure information not submitted.
IM

Iain MacEwan, n/a

MD, MCH
Department of Radiation Medicine and Applied Sciences, University of California, San Diego, La Jolla, CA
La Jolla, California, United States

Disclosure: Disclosure information not submitted.

Purpose: Craniospinal irradiation (CSI) is an essential technique in the treatment of pediatric brain tumors with propensity for spread in the CSF. Most investigations have compared proton vs. photon treatment outcomes for CSI. However, there can be substantial variation in treatment techniques or volumes within proton therapy. We sought to characterize the dependence of clinical outcomes on proton CSI practice patterns and plan characteristics using the Pediatric Proton Consortium Registry (PPCR) database.

Methodology: We queried PPCR for CSI data from 2015-2020. Each plan was manually reviewed for: prescription dose; 95% isodose line coverage of optic nerves, vertebral bodies (with vertebral body sparing defined per SIOPE guidelines), and sacral nerves; and esophageal mean/max doses. Clinical outcomes of interest included growth (height-for-age z-scores per WHO Child Growth Standards), acute hematologic toxicity, acute gastrointestinal toxicity, and relapse free survival (RFS). Growth was assessed using multivariate linear mixed models, toxicity using multivariate logistic regression, and RFS using Fine-Gray regression modeling of competing risks.

Results: PPCR supplied data for 451 CSI patients; 317 had complete RT data. Medulloblastoma was the most common diagnosis (217). The median age was 8.8y and the median CSI dose was 23.4 Gy.

On multivariate analysis, height z-score change after RT was affected significantly by time since RT (β=-0.38 per year, p< 0.001), age (β=0.05 per year, p=0.008), and the interaction of time since RT and use of vertebral body sparing (β=0.18 per year, p=0.003). However, the interaction term was not significant on subset analysis of younger skeletally immature patients, suggesting this finding was driven by the preferential use of VB sparing in older children.

Acute hematologic toxicity was associated with increasing CSI dose (OR 1.05, p=0.014), but not with VB sparing or concurrent chemotherapy. Acute GI toxicity was associated with lack of VB sparing (OR 5.29, p< 0.001), as well as increasing age, esophageal Dmax and Dmean, and the use of concurrent chemotherapy.

On multivariate analysis, full vs. incomplete optic nerve coverage was associated with a reduced hazard of relapse (HR 0.49, CI 0.26 – 0.93, p=0.029). Sacral nerve coverage was not associate with relapse.

Conclusions: The use of VB sparing proton therapy for CSI was associated with improved growth outcomes overall, but was driven by the preferential use of this technique in older children. This technique was also associated with reduced acute GI toxicity, but did not impact acute hematologic toxicity. Complete coverage of the optic nerves was associated with a reduced incidence of a relapse, whereas coverage of the sacral nerves did not have an effect. With the increased conformality of proton therapy, further evidence is needed to understand the effects of varying treatment target delineation practices and develop standard recommendations.

References: