Session Item

Online MR guided radiotherapy of abdominal and thoracic targets could potentially benefit from the use of respiratory beam gating. The purpose of the current clinical study was to investigate the dosimetric consequences of target motion during adaptive SBRT of adrenal gland metastases treated on an MR linac without beam gating options.

The study is comprised of 21 patients with adrenal gland metastases treated to 45 Gy in 3 fractions (n=8), 50 Gy in 5 fractions (n=11), or 60 Gy in 8 fractions (n=2) on a 1.5 T MR linac. Dose planning aimed to achieve a conformal and inhomogeneous dose distribution, which maximises the GTV V_95% and mean dose, covers the PTV with 67% of the prescribed dose while respecting hard OAR constraints for stomach, intestines, great vessels, and kidneys. The size of the PTV margins was based on the peak-to-peak respiration motion of the target assessed on pre-treatment 4DCT. Each treatment fraction included a 3D T2w MR image, used for plan adaptation, and alternating sagittal and coronal 2D BFFE cine MR images at 5 Hz for motion monitoring during beam-on. Patients were treated while breathing freely, however, for 7 patients is was attempted to reduce the respiration induced target motion by use of an abdominal compression belt. The time-resolved position of the target during radiotherapy was determined using rigid registration between the online 2D cine images and the daily 3D planning MR images. The delivered GTV dose was estimated from the daily planned dose that was rigidly shifted and accumulated according to the time-resolved target position changes due to drift and respiration motion.

The 4DCT peak-to-peak respiration motion assessed prior to treatment correlated with the target peak-to-peak respiration motion measured during treatment (R=0.80, p<0.01 – see fig. 1). The respiration motion was significantly lower in patients treated with a compression belt (Mann–Whitney U test, p=0.03). The population mean DVH curves for the planned and delivered GTV dose is shown in figure 2A. The curves differ in regions above V_85%, with a slightly lower delivered dose than the planned dose. The difference in V_95% is shown in figure 2B as a function of the respiration motion amplitude. Reductions in V_95% due to target motion was below 2% for the majority of the patients, while larger reductions were seen in patients with larger respiration motion.

Dose coverage degradation due to target motion during beam delivery is generally small in patients with adrenal gland metastases treated on an MR linac without gating options, but more noticeable degradation may be observed in patients with larger respiratory motion. The pre-treatment target respiration motion observed using 4DCT is conserved during treatment and is therefore valid for PTV margin estimation. The respiratory motion is reduced in patients treated with a compression belt; thus, the belt is a valid alternative to avoid the prolonged treatment time needed for gated treatments.