Impact of respiratory motion on dose distribution in SIB-SBRT for lung cancer

Respiratory motion is a major source of dose uncertainty in lung cancer radiotherapy. The dose distribution of simultaneous integrated boost-stereotactic body radiotherapy (SIB-SBRT) is inhomogeneous and is significantly impacted by respiratory motion for lung cancer. The effect of respiratory motion on SIB-SBRT was investigated with a four-dimensional (4D) dose calculation method.

Nineteen previously treated lung cancer patients were selected for this planning study. All patients underwent four-dimensional CT (4D-CT) scanning, and volumetric modulated arc therapy (VMAT) treatments were planned with internal target volume (ITV) and planning target volume (PTV). Dose distributions (3D-plan) were calculated on the average reconstruction of the 4D-CT. 4D dose distributions (4D-plan) were calculated to evaluate respiratory motion effects. These calculations were performed on the CT images of related respiratory phase with a respiration-correlated assignment of the 3D plan's monitor units to the respiratory phases of the 4D-CT. Subsequently, the accumulative 4D dose based on deformable registrations of the CT series was generated and compared to the 3D dose distribution. Dosimetric deviations in targets and organs at risk (OARs) were analyzed with dosimetric parameters, and correlations between dose deviations (ΔV₁₀₀ (ITV, PTV)) and patient characteristics (left-right, SI, anterior-posterior, S, L, Volume (ITV, PTV)) were explored.

With deformable registrations, the median values of relative differences between 3D-plan and 4D-plan 0 were found to be from -6.6% to 12.1% for all targets dosimetric parameters, and from -4.2% to 1.4% for OAR parameters. It was also shown that PTV coverage dropped more significantly than that of ITV with respiratory motion. Strong correlations were observed between the ΔV₁₀₀ (ITV, PTV) and patient characteristic (SI, S, L).

Respiratory motion effects during SIB-SBRT treatment resulted in non-negligible dose variability. Furthermore, with the correlation relationship and respiratory motion parameters, the dose coverage reduction of targets could be predicted.