Realtime compensation of breathing motion during radiotherapy

A strategy to minimize errors from differential intrafraction organ motion using a single configuration for a ‘breathing’ multileaf collimator

Intensity-modulated radiation therapy (IMRT) can be delivered by the 'sliding-leaves' dynamic multileaf collimator (DMLC) technique. Intrafraction organ motion can be accommodated by arranging an identical tracking motion for 'breathing leaves'. However, this is only possible for very specific circumstances such as regular, mathematically parameterizable, rigid-body, density-conserving, one-dimensional translations. In this paper, we investigate what happens when planes of tissue in the line of sight of the MLC have differential motion with respect to the moving leaves. In this situation, there is no solution to the problem and a perfect tracking motion cannot be arranged. However, an iterative minimization-of-errors 'solution' (or strategy) can be found and the technique is presented for this. From this, under certain mathematically simple differential motions it is possible to obtain some elegant algebraic solutions which are presented. In general, however, a lengthy computational minimization is required and results of examples of these are presented.

Does elastic tissue intrafraction motion with density changes forbid motion-compensated radiotherapy?

Intrafraction organ motion disturbs the otherwise highly conformal dose distributions planned for conformal radiotherapy and intensity-modulated radiotherapy. If the organ motions are rigid-body translations and preserve voxel density, techniques to make the beam follow the target by synchronous breathing can be arranged. Several potential difficulties disturb this oversimple statement and one of these is addressed in this paper. The effects of general elastic motion without preserving density are investigated. The effects are complex. Dose-space voxels disconnect from the planned irradiation bixel; voxels in a set aligned with a particular bixel cease to be so aligned even with a different bixel on elastic motion. Furthermore the density changes to conserve mass and this further perturbs the dose distribution. These effects are demonstrated here via simple examples but it may be impossible to make 'beam-breathing' strategies to compensate for them. Consequently understanding (and possibly ignoring) these second-order effects may be all that is possible.