Session Item

Accurate patient setup is essential for delivering radiation therapy (RT) in accordance with the treatment plan. Interfraction translational and rotational errors have been determined from daily kVCBCT to evaluate the accuracy of two different immobilization devices in both moderate and extreme hypofractionated prostate RT.

Daily CBCTs from 40 and 33 prostate cancer patients who underwent at a single institution moderate hypofractionated RT in 20 fractions with bended knees (BK) and extreme hypofractionated RT in 4 or 5 fractions with straight knees (SK), respectively, were retrospectively analysed and compared, amounting to 1180 analysed datasets. The two immobilization devices are shown in Figures 1 and 2. Systematic and random shift components were evaluated using a clipbox matching based on a grey-value algorithm and validated by experienced radiation oncologists and RTTs. For both patient populations, the mean value of the average deviations M, the group systematic error ∑ and the group random error σ were computed for translations and rotations on all axes, respectively. Additionally, a Wilcoxon Mann Whitney test was performed to assess statistical significance between results (p=0.05).

Translation and rotation M values were generally less than 1 mm and less than 0.6°, respectively. Statistical analysis between M values of the two populations led to comparable results, except for a significant improvement in X translational axis (p=0.006) and a significant degradation in Z rotational axes (p=0.003) in SK setup respect to BK. Both systematic and random translation errors were negligible and corrected by the Linac couch movement. The systematic and random rotational errors, which could not be corrected by standard treatment tables, resulted in less than 0.6° on the Y and Z axes for both devices. On the X axis, ∑ component was 1.38° for SK positioning and 1.44° for BK, while σ component was 0.91° and 1.05° for SK and BK, respectively. The overall number of problematic setup sessions was very low and limited to a small subset of patients.

Both positioning with straight knees and bended knees showed extremely high accuracy with no significant differences for moderated and extreme hypofractionated prostate radiotherapy. Patients experienced optimal comfort for the whole treatment duration with both devices. Notably, RTTs involved in patient positioning reported an overall perceived improvement in patient setup with straight knees and a substantial reduction in the procedure time.