Purpose: Virtual couch shift (VCS) technique shifts MLC apertures rather than treatment couch to correct patient setup errors. When adapted to flattening-filter-free (FFF) linacs such as the MR-linac, VCS may lead to very different fluence pattern due to the non-flattened radiation field. This study aims to evaluate this effect for various anatomic sites.
Methods: Four most commonly treated sites on MR-linacs (metastatic node, pancreatic, prostate and rectal cancers) were selected for this simulation study. Reference IMRT plan (P0) was generated in Pinnacle using 6MV FFF beam on an Elekta Versa linac with Agility MLC head. Translational shifts of 5, 10, 15 and 20-mm were introduced by shifting isocenter in the opposite direction. A Pinnacle script was developed to generate shifted MLC apertures taking beam divergence into account. For shift-only plan, dose distribution was computed with the original MUs. Shift-only plan was subsequently copied and re-optimized with segment weight optimization (SWO) using the same objectives from P0. Various dosimetric indices for planning target volume (PTV) and organs-at-risk (OARs) were collected and compared between P0, shift-only and SWO.
Results: Shift-only resulted in decreased dose to PTV in the shifted direction and mostly affected Vᵣₓ (PTV received prescription dose). Dose reduction increased with shift distance and target size, and appeared more pronounced for shifts in the superior-inferior (SI) directions. For prostate and rectal cases, Vᵣₓ could drop by ~20% for 10-mm shift and ~50-60% for 20-mm shift. The impact on OAR indices (D0.5cc and D5cc) was relatively small and didn’t show a clear trend. SWO quickly converged within ~20 optimization iterations. All SWO plans achieved comparable plan qualities as P0 except for one extreme situation (rectal case with 20-mm SI shift).
Conclusion: Shift-only method mostly affects target coverage. Simple SWO is effective in restoring plan qualities even with relatively large shift.
Not Applicable / None Entered.