Development of the first prediction model for radiation-induced contrast enhancement after proton therapy for posterior fossa tumours in paediatric patients

Abel Bregman ¹, Jikke J Rutgers ², Truls Andersen ¹

⁰Department of Radiation Oncology, University of Groningen, University Medical Centre Groningen, Groningen, the Netherlands.

Radiotherapy and Oncology : Journal of the European Society for Therapeutic Radiology and Oncology +

|

December 22, 2025

View abstract on PubMed

Summary

This study developed a normal tissue complication probability (NTCP) model for radiation-induced contrast enhancement (RICE) in pediatric patients receiving proton therapy. Younger age, higher dose, and brainstem location increase RICE risk.

Area Of Science

Radiation oncology
Pediatric oncology
Medical physics

Background

Radiation-induced contrast enhancement (RICE) serves as a subclinical toxicity marker.
Understanding RICE is crucial for managing long-term effects in pediatric cancer survivors.

Purpose Of The Study

To develop a normal tissue complication probability (NTCP) model for RICE.
To predict RICE risk in pediatric patients undergoing proton therapy for posterior fossa tumors.

Main Methods

A cohort of 75 pediatric patients treated with proton therapy was analyzed.
Multivariable logistic regression models were used to predict RICE based on dose (D), dose-averaged linear energy transfer (D⋅LETd), age, and tumor location.
NTCP was calculated using univariable logistic regression.

Main Results

RICE lesions were identified in 30.7% of patients, predominantly in the brainstem pons.
Increased RICE voxel density correlated with high D and medium-to-high LETd.
Younger age and pons location were significant independent risk factors for RICE.

Conclusions

Developed predictive models for RICE in pediatric proton therapy patients.
RICE probability increases with dose, D⋅LETd, younger age, and brainstem pons location.
The NTCP model demonstrated good predictive performance (AUC 0.79).