Precision of Image-Guided Radiotherapy (IGRT) in Six Degrees of Freedom and Limitations in Clinical Practice

Abstract

To evaluate the precision of image-guided radiotherapy (IGRT) using cone-beam computed tomography (CB-CT) for volume imaging and a robotic couch for correcting setup errors in six degrees of freedom. 47 consecutive patients with 372 fractions were classified according to whether a patient fixation device was used (pat(fix): n = 28) or not (pat(non-fix): n = 19). Prior to treatment a CB-CT was acquired and translational and rotational setup errors were corrected online without an action level using a robotic couch (HexaPOD). A second CB-CT was acquired after the correction process and after treatment in 134 and 238 fractions, respectively. In 17 fractions (4.6%) rotational errors > 3 degrees exceeded the motion range of the HexaPOD. Errors (3D vector) after the correction process were significantly smaller for pat(fix) compared to pat(non-fix) (p < 0.001): 0.9 mm +/- 0.5 mm and 1.6 mm +/- 0.8 mm, respectively. For pat(non-fix) the correction of rotational errors resulted in displacements of the patients on the angled couch of 0.6 mm/1 degree. Intrafractional motion further decreased precision in pat(non-fix) but not in pat(fix). Very high precision in cranial and extracranial treatment of immobilized patients was demonstrated. Without application of adequate immobilization the correction of rotational errors and intrafractional patient motion significantly decreased the accuracy of the online correction protocol.