Evaluation of dose prediction errors and optimization convergence errors of deliverable-based head-and-neck IMRT...

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Imaging

Background:

• Intensity-modulated radiation therapy (IMRT) is a standard treatment for head-and-neck (HN) cancers.
• Accurate dose calculation is crucial for effective IMRT optimization.
• Superposition/convolution (SC) algorithms are widely used but may have limitations in complex geometries.

Purpose of the Study:

• To evaluate dose prediction errors (DPEs) and optimization convergence errors (OCEs) in HN IMRT using a superposition/convolution (SC) algorithm.
• To compare a mixed Monte Carlo (MC)/SC optimization method with a full MC optimization method.
• To assess the clinical significance of DPEs and OCEs and their impact on normal tissue dose.

Main Methods:

• Retrospective reoptimization of 13 HN IMRT patient plans.
• Three optimization steps: fast optimization, mixed MC/SC optimization, and full MC optimization.
• Analysis of dose-volume indices and statistical equivalence tests for DPEs and OCEs.
• Correlation analysis between DPEs and OCEs.

Main Results:

• DPEs were within +/- 2.8%, while OCEs were within 5.5%.
• OCEs can be clinically significant even when DPEs are insignificant.
• Full MC optimization reduced normal tissue dose by up to 8.5% compared to the mixed method.
• Strong correlation between DPEs and OCEs in targets (r > 0.71), no correlation for organs at risk.

Conclusions:

• Full MC-based optimization is advantageous for HN IMRT due to lower normal tissue doses.
• Careful consideration of OCEs is necessary for clinical implementation of IMRT.
• The findings highlight the importance of advanced dose calculation algorithms in optimizing radiation therapy.