Mixed Effect Modeling of Dose and Linear Energy Transfer Correlations With Brain Image Changes After Intensity...

Area of Science:

• Radiation Oncology
• Medical Physics
• Neuroimaging

Background:

• Intensity modulated proton therapy (IMPT) can deliver high doses of radiation.
• High linear energy transfer (LET) from IMPT may increase biological effects.
• Radiation-associated brain image change (RAIC) is a potential complication of IMPT for head and neck cancers.

Purpose of the Study:

• To investigate the voxel-wise correlation between radiation dose and LET.
• To determine the association between dose, LET, and RAIC after IMPT.
• To evaluate the predictive value of dose and LET for RAIC.

Main Methods:

• Analyzed data from 15 patients who developed RAIC after IMPT.
• Contoured RAIC on MRI and coregistered to CT scans.
• Used Monte Carlo simulations to calculate dose and dose-averaged LET (LETd) distributions.
• Applied uni- and multivariate logistic regression to model RAIC associations.

Main Results:

• Found a significant association between RAIC and both dose and LETd.
• Patient heterogeneity in response was observed.
• Models achieved high performance (AUC 0.93-0.95).
• Increased LETd correlated with a higher risk of RAIC, with specific dose-response estimates provided.

Conclusions:

• The effect of LETd is clinically significant for some patients.
• LETd assessment should be considered in clinical IMPT treatment planning.
• Further investigation into LETd's role in RAIC is warranted.