Assessment of Proton Pencil Beam Scanning (PBS) Treatment Quality Using 4D Monte Carlo Simulation in Lung Cancer

Purpose: To assess proton pencil beam scanning (PBS) treatment quality for lung cancer patients who exhibit both tumor motion and tissue heterogeneity using 4D Monte Carlo simulation based on an open-source fast Monte Carlo (MC) algorithm, MCsquare. Methods: Twelve lung patients with different motion amplitudes previously treated with double scattering proton therapy were replanned with PBS. The motion magnitude is obtained from the deformable registration field between the end exhalation and end inhalation CTs. The inaccuracies of analytical dose calculation (ADC) due to tissue heterogeneity was assessed by analyzing the discrepancy between recalculated dose (MC static dose) using MCsquare and TPS (ADC static dose). The interplay effect was investigated by calculating the 4D dose by accumulating the spot dose on the corresponding respiration phase to the reference phase by deformable image registration based on spot timing and patient breathing phase (MC Interplay dose). DVH indices such as D95 were compiled from the three computed dose distributions and analyzed for each patient. Results: The twelve patients had a range of motion from 3.1mm to 12.3mm. The difference in D95 between ADC and MC static dose ranged from 2.1% to 8.2%; between MC interplay average and MC static dose: -0.2% to 2.3%; between maximum and minimum interplay dose: 0.3% to 4.1%. The motion amplitude correlates with the range of the interplay dose, and the D95 difference between ADC and MC can be significantly higher than the difference between MC and interplay dose. The other indices follow the same trend as D95. Conclusion: We demonstrated a 4D fast Monte Carlo simulation to assess both the effect of heterogeneity and motion of lung patients treated with PBS. Implementation of this fast MC in clinics can give a better picture of the combined impact of a patient’s motion and heterogeneity on the dose coverage