Monte Carlo vs. pencil beam based optimization of stereotactic lung IMRT

Area of Science:

• Medical Physics
• Radiation Oncology
• Computational Biology

Background:

• Intensity-modulated stereotactic radiotherapy (IMSRT) for lung cancer requires precise dose calculation.
• Finite size pencil beam (fsPB) and Monte Carlo (MC) methods are used for dose optimization.
• Comparing these methods is crucial for improving treatment planning.

Purpose of the Study:

• To compare the efficacy of fsPB and MC-based optimization for lung IMSRT.
• To evaluate dose calculation accuracy and treatment plan quality using different patient models.
• To assess the impact of 4D CT data on treatment plan outcomes.

Main Methods:

• Validated fsPB and MC algorithms using film measurements in a static lung phantom.
• Applied algorithms for lung IMSRT planning using three static patient CT models (static, density overwrite, average CT).
• Calculated dose distributions on static models and recalculated accumulated dose using MC on 4D CT data.

Main Results:

• MC dose engine showed <2% discrepancy, while fsPB showed up to 8% in phantom studies.
• fsPB-optimized plans exhibited organ at risk constraint violations and unpredictable target doses.
• MC-optimized plans, though underestimating target doses, showed comparable organ at risk doses compared to 4D MC recalculations.

Conclusions:

• MC dose engine is feasible and produces superior plans for lung IMSRT compared to fsPB.
• Static patient models introduce predictable bias in MC dose distribution but are considered safe for optimization.
• MC-based optimization on static models is a safe and effective approach for lung IMSRT.