Purpose/Objective: IMRT has been shown to improve radiotherapy dose distributions when applied to lung treatments, especially if the delivery can be registered to the patient′s breathing cycle. However, many technical issues associated with breathing-registered IMRT treatment planning have not yet been addressed. In this study, we explore the management of breathing within radiotherapy planning by incorporating the motion pattern directly into treatment planning process. The movement of the voxels from one CT timeframe to another is “tracked” and modeled. In particular, we provide a timestamp for each voxel that specifies the position of this voxel from one phase to another. In the formulation of the treatment models, instead of using constraints for the voxel for just one phase, as in the static case, we incorporate voxel constraints throughout multiple time periods. Robustness of the algorithm, plan quality, and potential clinical significance are evaluated.
Materials/Methods: 4DCT scans of lung cancer patients were acquired with different breathing phases (phases 0–9, 0:full-inhale, 5:full-exhale). 3 treatment planning strategies are performed and compared. 1) Standard planning with a static PTV based on a single selected phase of the breathing cycle (as control). 2) The Internal Target Volume (ITV) approach, where ITV is defined as the union of CTVs in all breathing phases. 3) CT scans of all phases are used for multi-stage optimization and planning. In this case, the movement of the voxels from one CT timeframe to another is “tracked” and modeled. In the formulation of the treatment models, instead of using constraints for the voxel for one phase as in the static PTV or ITV approach, we incorporate voxel constraints throughout the multiple-phase period. Thus, the resulting optimization model is many times larger than when using either PTV or ITV planning.
Results: The standard PTV plan provides 92–95% coverage over all phases, whereas the ITV and the multistage plans offer 95% coverage, consistently achieved in all phases of the breathing cycle.
The PTV plan results in drastic underdose (over 50%) to the tumor over the breathing cycle. The underdose value for ITV remains rather constant (3%). And the multistage approach offers significant improvement in underdose over the PTV plan, and has less than 7% dose difference from the ITV plan.
Comparing multistage plans to ITV plans, multistage plans reduce the dose to left lung normal tissue, heart, and esophagus. Specifically, for the entire breathing cycle, the maxi-dose to the left lung normal tissue remains at 67Gy (vs 70.8Gy), and the mean dose ranges from 26.7–28.1Gy (vs 33.3–34.4Gy). Hence, the multistage plan significantly reduces the mean dose by 20%. Furthermore, this tumor is close to the heart, and we observe significant dose reduction (20%) to the heart in the multistage plan. Similarly, esophagus dose was reduced by 20%.
Conclusions: For intrafractional organ motion, multistage plan optimization can provide good tumor-coverage plans, improve tumor-underdose, and significantly reduce dose to organs-at-risk (OARs), especially those organs in the proximity of the tumor. Evidence of morbidity reduction to OARs are observed. The challenge involves the ability to solve a large-scale treatment planning problem. With sophisticated mathematical optimization modeling and computational strategies, such planning is possible and can be made available for clinical use. Further investigation of ITV will be conducted to understand if improved plan quality can be achieved via improved construction of the ITV structure. Clinical studies are needed to validate the importance of our approach to treatment outcome.
