Getting in Line: Impact of Indexing on Treatment Setup Accuracy for Prostate Radiotherapy

Abstract

Purpose/Aim

Patient positioning is vital to the accurate delivery of radiation therapy treatments, especially in the era of ultra-hypofractionation and stereotactic body radiotherapy (SBRT). Indexing immobilization devices improves positional consistency and may minimize setup errors. With daily onboard imaging, changes in positioning can be visualized and rectified translationally, however, rotations may not always be corrected without 6 degrees-of-freedom compliant linear accelerator beds. The purpose of this retrospective analysis is to determine whether utilizing indexing in patient positioning for prostate patients improves treatment setup in terms of rotations.

Methods/Process

Images from prostate patients treated between August 2023 to November 2023 with indexed and unindexed immobilization were reviewed. All patients were positioned using two standard pillows under head, a thin mattress, and a black leg immobilizer. Indexing was achieved using an in-house fabricated bar mounted at specific coordinates on each tabletop. The mounting position acted as an anchor to fix the leg immobilizer, allowing a reference for the distance from the inferior aspect of the leg immobilization (bottom of the foot) to the set-up mark. Patient planning CTs and cone beam CTs (CBCTs) acquired daily before treatment were restored in the radiation therapy electronic medical records and re-matched to determine the extent of rotations in all three directions from set-up. Student t-test was used to determine differences between the cohorts.

Results or Benefits/Challenges

A total of 988 CBCTs were successfully restored and rematched. The study consisted of four cohorts: Patients treated to their primary disease and pelvic nodal volumes with indexing (Indexed_Pelvis, n=330) and without indexing (Unindexed_Pelvis, n=306), and patients treated to their primary disease only with indexing (Indexed_Prostate n=200), and without indexing (Unindexed_Prostate, n=152). The maximum rotation exhibited was 4 degrees in the pitch for patients with indexed immobilization, whereas the maximum rotation for the unindexed cohorts was 5.3 degrees. The average yaw from Index_Pelvis was statistically higher than Unindexed_Pelvis (meanIndexed_Pelvis = 0.57, meanUnindexed_Pelvis = 0.38, p<0.001). The average pitch for Indexed_Pelvis was statistically lower than the average pitch for Unindexed_Pelvis (meanIndexed_Pelvis = 0.68, meanUnindexed_Pelvis = 1.29, p<0.001). No statistical differences were found for the roll exhibited by the Indexed_Pelvis and Unindexed_Pelvis cohorts. All rotations for the Indexed_Prostate group were statistically lower than the Unindexed_Prostate group (yaw: meanIndexed_Prostate = 0.32, meanUnindexed_Prostate = 0.46, p<0.001; pitch: meanIndexed_Prostate = 0.64, meanUnindexed_Prostate = 1.17, p<0.001; roll meanIndexed_Prostate = 0.37, meanUnindexed_Prostate = 1.0, p<0.001).

Conclusions/Impact

The use of a simple patient positioning device such as the black-leg immobilizer can result in setup rotations that on average were consistently <2 degrees in all directions. Use of indexing resulted in an average rotation of <1 degree in all three directions. Thus, indexing is a valuable technique that should be implemented to improve prostate patient setup especially in the era of ultra-hypofractionation, and stereotactic body radiotherapy to improve treatment accuracy.