Background and Purpose: The precision of positron emission tomography (PET)-guided radiotherapy may be compromised by motion.To address this, we demonstrated the feasibility of a motionrobust PET-to-magnetic resonance linear accelerator (MR-linac) pipeline. Materials and Methods:For comparison, PET images of a phantom with a target in a movable insert were acquired under a patient-derived breathing waveform and static conditions.Motion correction using the PET Maximum-Likelihood Motion and Activity (MLMA) reconstruction was compared with static reference and motion-blurred images.Targets were delineated using thresholding based on maximum standardized uptake value, and intensity-modulated radiation therapy (IMRT) plans were generated.Plans were delivered on an MR-linac with and without gating or multi leaf collimator (MLC) tracking.Dosimetry was assessed using film measurements, evaluating dose differences and 3%/3 mm gamma pass rates.The pipeline was further evaluated using clinical PET data, delivered with and without gating and tracking after motion correction.Dosimetric evaluation was performed using a cylindrical multi-diode phantom.Results: Motion-blurred PET produced larger gross target volumes (GTV) (GTV:58 cm 3 , GTV boost :14 cm 3 ) than motion-corrected PET (GTV:52 cm 3 , GTV boost :8 cm 3 ), the latter matching the static reference (GTV:52 cm 3 , GTV boost :9 cm 3 ).Motion correction reduced dose discrepancies and improved gamma pass rates (72%), with further gains from gating (90.3%) and MLC tracking (98.3%).The measurements of patient simulations showed higher gamma pass rates for the motion-corrected plan (static-delivery:81.8%, gating:99.2%,MLC-tracked:100%) compared to motion-blurred plan (static-delivery:49.4%
Everard et al. (Wed,) studied this question.