Purpose: Investigations on the dosimetric outcome of energy optimized VMAT is missing in the literature. This work presents: a) an approach for simultaneous optimization of dual energy for a VMAT plan(DE-VMAT), and b) presents the dosimetric comparison between DE-VMAT and conventional single-energy-VMAT (SE-VMAT) approach.
Methods: The heuristic DE-VMAT approach is a two-step process. First, a set of dual energy intensity maps (DEIMs) are simultaneously optimized at a sequence of equally spaced beam angles (5Â°-10Â°) over a full 180Â° angular space using linear programming. In the second step, each of these DEIMs (one for each energy at a given angle) are consecutively segmented into k number of MLC apertures dispersed over a given gantry range by using mixed integer linear programming (MILP). The MILP model constrained the apertures with maximum leaf motion(â‰¤3cm/s) and maximum dose rates (â‰¤600 MU/min). The DE-VMAT was computed on a prostate case and compared with a SE-VMAT plan optimized under the same optimization criteria.
Results: The dose volume histogram(DVH) for DE-VMAT was in good agreement with the one obtained by using 36 discrete intensity maps obtained in the first step, indicating an appropriate convergence by MILP dispersion model. Both, DE-VMAT and conventional SE-VMAT, approaches provided similar target dose homogeneity and conformity with the HI of 0.07 and CN of 0.9 at 1deg/s gantry speed; however the HI and CN were superior for DE-VMAT at 2deg/s gantry speed. In addition, DE-VMAT was able to reduce the dose to the femoral heads, bladder and spare a large volume of rectum at 1 deg/s and 2 deg/s gantry speed.
Conclusion: Overall, simultaneously optimized DE-VMAT is a feasible approach. DE-VMAT can provide better OAR sparing, while providing similar target coverage. This preliminary study demonstrates potential viability of DE-VMAT approach, providing an impetus for further testing for a larger cohort clinical evaluation.