Dealing with rectum motion during radiotherapy: How can we anticipate it?

Abstract

Introduction

Intra- and inter-fraction rectum motion is important for pelvic radiotherapy (RT). This study assesses how RT session duration, the presence or the absence of an intra-rectal tumour, and the distance from the anorectal junction (ARJd) impact rectal motion.

Materials and methods

Analyses used cone-beam computed tomographies (CBCTs) from RT patients treated for rectal and prostate cancer. Three structures were evaluated: (1) the entire rectum in patients without a rectal tumour (Rectum_Prostate); (2) the non-invaded portion (Rectum_Rectum) and (3) the tumour-invaded portion (Rectum_Tumour) in rectal cancer patients.

Intrafraction motion was assessed using the Hausdorff distance 95% and the Mean distance-to-agreement between structures delineated on the first CBCT and the 2 subsequent CBCTs within a same RT session. Interfraction motion was quantified by comparing structures delineated on the planning-CT and the first CBCT of each session.

Linear mixed model evaluated rectum motion in relation to time, tumour presence, and ARJd, respectively.

Results

We included 10 patients with and 10 without rectal cancer, collecting 385 CBCTs. A significant correlation (p < 0.05) between rectum motion and RT session duration was found. Intrafraction motion was significantly higher in prostate cancer patients (Rectum_Prostate motion > Rectum_Rectum and Rectum_Tumour, p < 0.01). For interfraction motion, only the mean distance to agreement was significantly higher for Rectum_Prostate (p < 0.05). Motion increased significantly with ARJd for all three structures (p < 0.001).

Conclusions

Session duration, absence of a tumour, and ARJd are associated with larger intra- and interfraction rectal motion. This highlights the need for tailored RT treatment, including online-adaptive RT, to manage intra- and interfraction variations. Rectal motion should be handled differently for patients with prostate cancer and those with rectal cancer.