Purpose: To develop a flexible Monte Carlo (MC) based robustness calculation and evaluation framework to assess treatment plan robustness of dynamic trajectory radiotherapy (DTRT, extending VMAT by dynamic table and collimator rotation) treatment plans, by exploring the impact of uncertainties in patient setup and mechanical accuracy of machine components on the dose distribution. Additionally, a novel robustness index (RI) is defined, with the aim to condense the dosimetric robustness against multiple error scenarios (ESs) into one number.
Methods: An in-house MC based dose calculation framework is extended to assess robustness of treatment plans. This extension allows to automatically start dose calculations for user-defined ESs. Each ES can be a combination of patient setup uncertainties and of the following machine components: Angles of gantry, collimator, table-yaw, table-pitch and table-roll and positions of jaws, MLC banks and single MLC leaves. The framework allows the simultaneous evaluation of the ESs on a GUI according to user-specified dosimetric criteria such as D2%, mean dose, gamma analysis or dose difference to the nominal dose distribution (no error). The RI is defined by the fraction of ESs passing the previously defined DVH/Gamma criteria. The application of the framework is demonstrated on a DTRT plan for a nasopharynx case.
Results: The DTRT plan is evaluated for combinations of systematic errors in gantry, collimator and table angle between -4° and +4°. Robustness criteria of >90% Gamma-passing-rate (2%/2mm evaluated on CTV) lead to a RI 64.1%; and <1 Gy deterioration of D2% and D98% (of CTV) to RI 99.7%.
Conclusion: The extended framework was successfully implemented and demonstrated on a DTRT plan. Effects of uncertainties in patient setup and mechanical accuracy of machine components on the dose distribution can be evaluated in a user-friendly GUI to assess treatment plan robustness. This work was partially supported by Varian Medical Systems.
Funding Support, Disclosures, and Conflict of Interest: This work was partially supported by Varian Medical Systems.