On the use of a convolution-superposition algorithm for plan checking in lung stereotactic body radiation therapy

Area of Science:

• Medical Physics
• Radiation Oncology
• Radiotherapy Dosimetry

Background:

• Stereotactic body radiation therapy (SBRT) for lung tumors requires precise dose delivery.
• Dosimetric verification of lung SBRT is challenging due to anatomical complexities and tumor motion.
• Advanced algorithms are needed for accurate dose calculation and verification in lung SBRT.

Purpose of the Study:

• To evaluate the accuracy of a new software, M3D, for dosimetric verification of lung SBRT treatment plans.
• To compare M3D's performance against established methods like Monte Carlo (MC), AAA, and XiO.
• To assess the clinical applicability of M3D in lung SBRT verification.

Main Methods:

• Ten lung SBRT patient plans were analyzed.
• Dose calculations were performed using M3D (collapsed-cone-convolution), Eclipse AAA, XiO (superposition), and in-house Monte Carlo (MC).
• Target doses, lung doses (V20 Gy, V5 Gy), and gamma passing rates were compared using MC as the reference.

Main Results:

• M3D and RadCalc generally calculated higher point doses than MC, with M3D showing statistical significance.
• AAA and XiO calculated lower point doses than MC, with AAA being statistically significant.
• M3D calculated higher near-maximum and median target doses and lung doses compared to MC.
• M3D achieved high gamma passing rates (>95% for 5%/1 mm, >80% for 3%/1 mm).

Conclusions:

• M3D demonstrates closer agreement with Monte Carlo simulations for lung SBRT dosimetric verification compared to AAA and XiO.
• Discrepancies between treatment planning systems and M3D are expected for lung SBRT.
• M3D serves as a valuable supplement to direct measurements for lung SBRT verification, accounting for patient-specific heterogeneities.