Quantification of Organ Motion in Male and Female Patients Undergoing Long-Course Radiation Therapy for Rectal Cancer in the Supine Position

Purpose: Previous studies have reported data on the internal rectal motion of patients with rectal cancer treated in the prone position. With the introduction of intensity modulated techniques, more patients are treated in the more reproducible supine position. Data informing specific margins for this treatment position are sparse, as are data comparing rectal motion characteristics and factors in male and female patients. The purpose of this retrospective study was to quantify and compare the interfractional rectal movement characteristics of male and female patients with rectal cancer treated with long-course chemoradiation therapy in the supine position. The data will aid the generation of internal target volume margins accounting for this organ's internal physiological movements.

Methods and materials: Cone beam computed tomography (CBCT) images were acquired from 19 male and 16 female patients with rectal cancer on the first 3 days of treatment and weekly thereafter. The rectum, bladder, and femoral heads were delineated on the planning CT (PCT) and 6 CBCT for each patient. Overall, 245 images were analyzed. All patients were treated with a full bladder. The rectum was divided into three 5-cm segments (upper, mid, and lower). The motion of the rectum was quantified by documenting the anteroposterior and lateral distances as measured using fixed anatomic landmarks, namely from the anterior aspect of the sacrum and mid-left femoral head, respectively. These measurements were taken at 1-cm intervals from the inferior border of L5 vertebrae. The sigmoid was excluded from these measurements. Estimations of systematic and random physiological movement error were determined and margins were calculated.

Results: Two hundred forty-five image sets (19 PCT + 114 CBCT for male, 16 PCT + 96 CBCT for female) on patients who had undergone long-course radiation therapy were analyzed. Rectal tumor location was 31% in the inferior rectum, 46% in the mid rectum, and 23% in the superior rectum. Random rectal motion (mean of the per-patient standard deviation [σ]) was largest for the upper and mid rectum in the anterior direction. There were statistically significant differences in σ between male and female patients in the left lateral motion of the mid and inferior rectum as well as the anterior, posterior, and right motion of the inferior rectum (mid left: P < .0005; lower left: P < .0005; lower posterior: P = .001; lower anterior: P = .032; lower right: P = .001). Suggested internal target volume margin guidelines are therefore nonisotropic and vary per segment of rectum and sex.

Conclusions: In our present study, interfractional rectal motion is shown to be significantly different between male and female patients. Our data suggest that the use of asymmetrical sex-specific margins in patients with rectal cancer treated in the supine position should be considered.