Motion management optimization in radiotherapy: From the most common to the most uncommon patient

In radiotherapy (RT), the treatment is thoroughly planned and optimized to fulfil the goal of delivering a high dose to the target, while sparing as much normal tissue as possible. This implies that the patient position and anatomy should be the same as they were during the planning image acquisition. To achieve this, it is important to have motion management methods for predicting, monitoring, and mitigating patient motion. These methods involve for instance adding extra margins around the target accounting for motion uncertainties, treatment in breath-hold, and imaging before and during beam-on. Ultimately, these techniques enhance treatment accuracy by reducing discrepancies between the planned and delivered dose. The work in this thesis aimed to evaluate and optimize motion management for patients treated with RT.

Various motion management techniques were analysed for different treatments and patient groups. Surface imaging (SI) and its potential to improve the workflow was assessed for both our most common patients (breast and prostate cancer patients) treated with conventional RT and for some of our most uncommon patients (canine patients) treated with the emerging RT technique FLASH-RT. The potential dosimetric effect of prostate motion and inadequate motion management in the magnetic resonance linear accelerator (MR-linac) workflow was investigated for three different planning target volume (PTV) margins. The dosimetric effects of different motion management methods were also evaluated for our most uncommon patients, ventricular tachycardia (VT) patients receiving stereotactic body RT (SBRT), by simulating treatment in breath-hold (BH) and treatment in free breathing with and without abdominal compression (AC). The potential of AC to decrease respiratory induced heart motion was also investigated.

This thesis demonstrated that SI can improve the initial patient setup accuracy and efficiency and is currently the only feasible motion management option for our FLASH-RT treatments. Further, it was shown that there is a risk of underdosage of the prostate in the MR-linac workflow if position correction is not carried out just before beam-on. Finally, we demonstrated that for most patients heart motion was reduced with AC, however it also increased motion for a few patients. It was shown that AC can reduce motion but also push the stomach closer towards the target, making AC dosimetrically unfavourable. Treatment in BH appeared to be dosimetrically preferable, however an induvial assessment of both BH and AC should be carried out for all VT patients.

In conclusion, this thesis has improved the initial patient setup accuracy and efficiency, implemented motion management for a new treatment technique, raised awareness of risks if proper motion management is left out and demonstrated dosimetric effects of different motion management techniques. This thesis has contributed to increased knowledge for future margin reductions, breathing adapted RT, and new motion management implementations and highlights the importance of continuous motion management optimization in conventional, novel, and emerging treatments.