Repositioning accuracy of a commercially available double-vacuum whole body immobilization system for stereotactic body radiation therapy
Martin Fuss,Bill J Salter,Premavathy Rassiah,D Cheek,Sean X Cavanaugh,Terence S Herman
DOI: https://doi.org/10.1177/153303460400300107
Abstract:We evaluated the repositioning accuracy of a commercially available stereotactic whole body immobilization system (BodyFIX, Medical Intelligence, Schwabmuenchen, Germany) in 36 patients treated by hypofractionated stereotactic body radiation therapy. CT data were acquired for positional control of patient and tumor before each fraction of the treatment course. Those control CT datasets were compared with the original treatment planning CT simulation and analyzed with respect to positional misalignment of bony patient anatomy, and the respective position of the treated small lung or liver lesions. We assessed the stereotactic coordinates of distinct bony anatomical landmarks in the original CT and each control dataset. In addition, the target isocenter was recorded in the planning CT simulation dataset. An iterative optimization algorithm was implemented, utilizing a root mean square scoring function to determine the best-fit orientation of subsequent sets of anatomical landmark measurements relative to the original treatment planning CT data set. This allowed for the calculation of the x, y and z-components of translation of the patient's body and the target's center-of-mass for each control CT study, as well as rotation about the principal room axes in the respective CT data sets. In addition to absolute patient/target translation, the total magnitude vector of patient and target misalignment was calculated. A clinical assessment determined whether or not the assigned planning target volume safety margins would have provided the desired target coverage. To this end, each control CT study was co-registered with the original treatment planning study using immobilization system related fiducial markers, and the computed isodose calculation was superimposed. In 109 control setup CT scans available for comparison with their respective treatment planning CT simulation study (2-5 per patient, median 3), anatomical landmark analysis revealed a mean bony landmark translation of -0.4 +/- 3.9 (mean +/- SD), -0.1 +/- 1.6 and 0.3 +/- 3.6 mm in x, y and z-directions, respectively. Bony landmark setup deviations along one or more principal axis larger than 5 mm were observed in 32 control CT studies (29.4%). Body rotations about the x-, y- and z-axis were 0.9 +/- 0.7, 0.8 +/- 0.7 and 1.8 +/- 1.6 degrees, respectively. Assuming a rigid body relationship of target and bony anatomy, the mean computed absolute target translation was 2.9 +/- 3.3, 2.3 +/- 2.5 and 3.2 +/- 2.7 mm in x, y and z-directions, respectively. The median and mean magnitude vector of target isocenter displacement was computed to be 4.9 mm, and 5.7 +/- 3.7 mm. Clinical assessment of PTV/target volume coverage revealed 72 (66.1%), 23 (21.1%), and 14 (12.8%), of excellent (100% isodose coverage), good (>90% isodose coverage), and poor GTV/isodose alignment quality (less than 90% isodose coverage to some aspect of the GTV), respectively. Loss of target volume dose coverage was correlated with translations >5 mm along one or more axes (p<0.0001), rotations >3 degrees about the z-axis (p=0.0007) and body mass index >30 (p<0.0001). The analyzed BodyFIX whole body immobilization system performed favorably compared with other stereotactic body immobilization systems for which peer-reviewed repositioning data exist. While the measured variability in patient and target setup provided clinically acceptable setup accuracy in the vast majority of cases, larger setup deviations were occasional observed. Such deviations constitute a potential for partial target underdosing warranting, in our opinion, a pre-delivery positional assessment procedure (e.g., pre-treatment control CT scan).