Assessment of Clinical Impact of Analytical Dose Calculation in TPS for Proton PBS Treatment Using Fast Monte Carlo Simulation

Purpose/Objective(s): The limited accuracy of analytical dose calculation algorithms (ADC) currently used in commercial TPS will lead to dose degradation for proton treatment plan. Monte Carlo dose calculation (MDC) is generally superior to ADC to model the dose distribution especially for lung and head/neck patients where high heterogeneity involves. The purpose of this study is to provide a validated and fast Monte Carlo (FMC) code to assess the impact of approximations in ADC on clinical pencil beam scanning (PBS) plans covering various sites. Materials/Methods: First, FMC was commissioned and validated using water and tissue-mimicking lung phantom measurements as well as benchmarked with the general purpose Monte Carlo TOPAS. Then a comparative analysis between FMC and ADC were performed for a total of 50 patients with 10 patients per site (including liver, pelvis, brain, head and neck, lung). Differences between FMC and ADC were evaluated using dosimetric indices (target Dmean, D95, homogeneity index, V90) based on dose-volume histogram analysis, a 3D gamma-index (3%/3mm) analysis, and estimations of tumor control probability (TCP). Results: The FMC significantly reduced the calculation time from tens of hours for TOPAS to less than 15mins based on our workstation resource. Both FMC and TOPAS dramatically improved the gamma-index (7%/5mm) passing rate between simulation and 2D film measurements using the lung phantom from below the passing threshold of 85% for ADC to over 93%. Comparison between FMC and TOPAS for selected 20 patients covering 5 sites showed less than 1.7% difference for all the dosimetric indices/TCP value and larger than 99% for the 3D gamma-index passing rate of the target. Generally, through investigations of the difference between ADC and FMC for 50 patients, we found that the impact of approximations in ADC on the plan quality increases from site liver, pelvis, brain, head and neck, to lung, as the degree of tissue heterogeneity increases. The D95s were predicted within 6.5% of the corresponding FMC statistic, while the V90s were within 2.9%. The median gamma-index passing rate for target volumes decreased from 99.3% for liver to 75.8% for long. Dose differences can result in large TCP differences for lung (<10.5%) and head and neck (<7.5%), with smaller differences seen for brain (<2.5%), pelvis and liver (