Modeling Normal Tissue Complication Probability of Radiation-Induced Alopecia Following Intensity-Modulated Radiation...

Area of Science:

• Oncology
• Radiation Oncology
• Medical Physics

Background:

• Glioblastoma multiforme (GBM) is an aggressive brain tumor treated with radiotherapy (RT).
• Radiotherapy often causes radiation-induced alopecia (RIA), a significant toxicity affecting patient quality of life.
• Existing models for normal tissue complication probability (NTCP) lack specific application to RIA in GBM patients.

Purpose of the Study:

• To establish the first normal tissue complication probability (NTCP) framework for radiation-induced alopecia (RIA) in glioblastoma multiforme (GBM) patients.
• To compare the predictive performance of the Lyman-Kutcher-Burman (LKB) model and multivariate logistic regression for RIA.
• To identify key dosimetric and clinical factors influencing RIA development.

Main Methods:

• A prospective cohort of 41 GBM patients undergoing intensity-modulated RT (IMRT) was analyzed.
• Normal tissue complication probability (NTCP) was modeled using the LKB framework and multivariate logistic regression.
• Model performance was evaluated using AUC-ROC, Brier score, and Hosmer-Lemeshow tests, assessing dosimetric parameters and clinical variables.

Main Results:

• Grade 2 RIA incidence was 46.3% at 3 months and 31.7% at 6 months post-RT.
• Maximum dose (Dmax) to the scalp and concurrent chemotherapy were significant predictors of RIA (P < 0.05).
• Multivariate logistic regression demonstrated superior predictive accuracy (AUC: 0.91) compared to the LKB model (AUC: 0.89).

Conclusions:

• Multivariate logistic regression provides enhanced predictive accuracy for RIA in GBM patients compared to the LKB model.
• Optimizing IMRT plans to minimize scalp dose and incorporating chemotherapy data are crucial for risk stratification.
• Implementing these predictive models can improve patient-centered care by balancing tumor control and toxicity reduction.