Fast Monte Carlo log-based framework for robust 4D dose evaluation in proton therapy.

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Imaging

Background:

• Proton therapy for moving targets, especially in the mediastinum, requires advanced dosimetric evaluation.
• Current methods may not fully capture the complexities of organ motion during free-breathing treatments.

Purpose of the Study:

• To develop and validate a robust 4D Monte Carlo (MC) dose evaluation (4DDE) methodology for proton therapy.
• To integrate temporal information of both pencil beam delivery and organ motion for accurate dose assessment.
• To evaluate the impact of respiratory motion on treatment plan robustness.

Main Methods:

• Developed a specialized tool combining GPU-accelerated FRED MC code with temporal patient and delivery data.
• Implemented standard 3D dose evaluation, plan recalculation across respiratory phases, and robust 4DDE.
• Validated the approach using 4D CT scans from a Hodgkin lymphoma patient, employing deformable image registration.

Main Results:

• Robust 4DDE revealed wider dose ranges compared to 3D analysis, indicating the importance of temporal data.
• Voxel-wise worst-case dose distribution showed significant reductions in target coverage (47.4% for 95% isodose, 20.8% for 90% isodose) due to respiratory motion.
• Gamma index analysis demonstrated differences between 3D and 4D evaluations.

Conclusions:

• The developed tool provides a fast and versatile method for 4DDE in proton radiotherapy.
• Accounting for dynamic anatomical changes is crucial for ensuring robust treatment plan evaluations.
• This methodology enhances the accuracy of proton therapy for moving targets.