Effect of endorectal balloon positioning errors on target deformation and dosimetric quality during prostate SBRT

• Published in: Physics in medicine & biology Vol. 58; no. 22; pp. 7995 - 8006
• Main Authors: Jones, Bernard L, Gan, Gregory, Kavanagh, Brian, Miften, Moyed
• Format: Journal Article
• Language: English
• Published: England IOP Publishing 21.11.2013
• Subjects: Aged; Algorithms; Cone-Beam Computed Tomography; cone-beam CT; dose verification; endorectal balloon; Humans; Image Processing, Computer-Assisted; Male; Middle Aged; prostate; Prostatic Neoplasms - diagnostic imaging; Prostatic Neoplasms - surgery; Quality Control; Radiometry; Radiosurgery - instrumentation; Radiosurgery - methods; Radiotherapy Setup Errors; Rectum; SBRT
• ISSN: 0031-9155; 1361-6560
• DOI: 10.1088/0031-9155/58/22/7995

An inflatable endorectal balloon (ERB) is often used during stereotactic body radiation therapy (SBRT) for treatment of prostate cancer in order to reduce both intrafraction motion of the target and risk of rectal toxicity. However, the ERB can exert significant force on the prostate, and this work assessed the impact of ERB position errors on deformation of the prostate and treatment dose metrics. Seventy-one cone-beam computed tomography (CBCT) image datasets of nine patients with clinical stage T1cN0M0 prostate cancer were studied. An ERB (Flexi-Cuff, EZ-EM, Westbury, NY) inflated with 60 cm3 of air was used during simulation and treatment, and daily kilovoltage (kV) CBCT imaging was performed to localize the prostate. The shape of the ERB in each CBCT was analyzed to determine errors in position, size, and shape. A deformable registration algorithm was used to track the dose received by (and deformation of) the prostate, and dosimetric values such as D95, PTV coverage, and Dice coefficient for the prostate were calculated. The average balloon position error was 0.5 cm in the inferior direction, with errors ranging from 2 cm inferiorly to 1 cm superiorly. The prostate was deformed primarily in the AP direction, and tilted primarily in the anterior-posterior superior-inferior plane. A significant correlation was seen between errors in depth of ERB insertion (DOI) and mean voxel-wise deformation, prostate tilt, Dice coefficient, and planning-to-treatment prostate inter-surface distance (p < 0.001). Dosimetrically, DOI is negatively correlated with prostate D95 and PTV coverage (p < 0.001). For the model of ERB studied, error in ERB position can cause deformations in the prostate that negatively affect treatment, and this additional aspect of setup error should be considered when ERBs are used for prostate SBRT. Before treatment, the ERB position should be verified, and the ERB should be adjusted if the error is observed to exceed tolerable values.