Incorporate organ motion into MLC leaf sequencing for intensity modulated radiation therapy

Abstract Purpose: Intensity modulated radiation therapy (IMRT) has been shown to provide highly conformal dose to the target while significantly sparing the normal tissues. IMRT plans have been routinely delivered via multileaf collimators (MLC) using either dynamic or step-and-shoot mode. The effect of organ motion is usually considered by applying a 1-2 cm margin around the target. The actually delivered target dose distribution is different from the ideal dose distribution and the normal tissues in the margin area receive much higher doses, which may yield higher complications later on. The purpose of this work is to incorporate organ motion into MLC leaf sequencing for IMRT treatments. Our objective is to study the feasibility of this approach and the clinical advantages in terms of dose coverage and normal tissue sparing. Materials and Methods: An MLC leaf sequencer which compensates for organ motion has been created. In this feasibility study, the breast motion was mimicked by a typical breathing pattern of breast cancer patients obtained by measurement using external and internal markers. The leaf sequencer would read in the breathing pattern as a function of leaf pair locations as if the breathing pattern were fed into the MLC controller. The MLC leaf trajectories were adjusted based on breathing motion and Monte Carlo dose calculations were performed using MCDOSE on patient CT phantoms at different phases of the breathing cycle. An in-house program was used to interpolate the intermediate CT slices between the two extreme CT slices of breathing motion: inspiration and expiration. Accurate correlation between the voxels at any point of the breathing cycle was established so that the doses could be accumulated accurately. The calculation results of motion adaptive IMRT plan were compared with those of ideal plan and actually delivered plan without organ motion correction. The effect of time delay of leaf movement was also studied on the IMRT doses for the cases of 0.1, 0.2 and 0.5 second delay. Results: In all the cases studied, target dose conformity and uniformity were improved for the motion adaptive IMRT plan, and the doses to the critical structures, especially those surrounding the target, were greatly reduced. A motion adaptive IMRT plan was superior to a plan without organ motion correction, and it could fully recover the ideal IMRT plan assuming a perfect tracking of MLC leaf movement on breast motion. Our current optical/radiographical tracking system can provide real time organ motion measurement and correction with a time delay of about 1 second. The discrepancy caused by a 0.1 second delay was only 2% in the target coverage, which was considered clinically acceptable. The dose differences caused by the 0.2 and 0.5 second delays were found to be 4% and 10%, respectively. Conclusion: This feasibility study has shown that ignorance of the organ motion in the breast cancer treatments will result in lower dose to the target than prescribed and 50% higher dose to the surrounding normal tissues than allowed. By including organ motion into the MLC leaf sequencing, the target dose uniformity and conformity can be significantly improved and the dose to the adjacent normal tissues and critical structures can be greatly reduced by applying smaller treatment margins. In addition, the leaf motion delay was found to be insignificant in the treatment delivery. It is shown that this new approach can be introduced to the clinical routine to improve the current IMRT technique.