Frameless radiosurgery: Less invasive, more accurate

Abstract

Purpose Stereotactic radiosurgery using frame-based positioning is a well-established technique for treatment of benign and malignant lesions. By contrast, a new trend towards frameless systems using image-guided techniques is gaining acceptance. This study was designed to compare the overall accuracy of frameless with framebased radiosurgery, to evaluate immobilization characteristics (i.e. intrafraction motion) of a frameless mask device to invasive fixation (frame-based technique) and to evaluate if verification images during radiosurgery treatment can correct for mechanical inaccuracy of table rotations and intrafraction motion. Materials and methods Multiple hidden target tests (HTT) were performed to measure the overall accuracy of the two positioning techniques for radiosurgery (i.e. frameless using stereoscopic X-ray imaging and 6-degree-of-freedom (DOF) registration/positioning and frame-based using invasive ring and localizer box) and to measure the overall accuracy at different table rotations with and without 6 DOF registration/positioning. Forty patients with 66 brain metastases were enrolled for frameless radiosurgery using 6 DOF registration/positioning. Positioning results were collected before and after treatment to assess intrafraction motion. Obtained data was benchmarked to literature for comparison with frame-based technique. Sixteen trigeminal-neuralgia patients were enrolled to assess positioning errors after every table position. Stereoscopic verification images and 6 DOF registration/positioning were performed and analyzed after each table rotation. Results HTT showed overall three-dimensional (3D) accuracy of 0.76 mm (SD0.46 mm) for the frameless technique, 0.87 mm (SD0.44 mm) for the frame-based technique for table positions at 0°. Rotating the couch to 90° and 270°, showed deviations of 0.83 mm (SD0.34 mm) and 0.62 mm (SD0.16 mm), respectively, if uncorrected. These deviations were reduced to 0.04 mm (SD0.03 mm) and 0.06 mm (SD0.04 mm) by image-guided corrections at 90° and 270°. The mean 3D intrafraction shift was 0.58 (SD0.42 mm) and comparable to frame-based techniques. By applying verification images during treatment (every 4.6 min SD 2.3 min), the mean 3D intrafraction shift decreased to 0.51 mm (SD0.13 mm). Conclusions With proper immobilization and X-ray verification images during treatment, frameless radiosurgery can be delivered as accurate as the frame-based approach with comparable immobilization characteristics to the invasive head ring and can be considered to be a reliable alternative for radiosurgery treatments. Stereotactic radiosurgery using frame-based positioning is a well-established technique for treatment of benign and malignant lesions. By contrast, a new trend towards frameless systems using image-guided techniques is gaining acceptance. This study was designed to compare the overall accuracy of frameless with framebased radiosurgery, to evaluate immobilization characteristics (i.e. intrafraction motion) of a frameless mask device to invasive fixation (frame-based technique) and to evaluate if verification images during radiosurgery treatment can correct for mechanical inaccuracy of table rotations and intrafraction motion. Multiple hidden target tests (HTT) were performed to measure the overall accuracy of the two positioning techniques for radiosurgery (i.e. frameless using stereoscopic X-ray imaging and 6-degree-of-freedom (DOF) registration/positioning and frame-based using invasive ring and localizer box) and to measure the overall accuracy at different table rotations with and without 6 DOF registration/positioning. Forty patients with 66 brain metastases were enrolled for frameless radiosurgery using 6 DOF registration/positioning. Positioning results were collected before and after treatment to assess intrafraction motion. Obtained data was benchmarked to literature for comparison with frame-based technique. Sixteen trigeminal-neuralgia patients were enrolled to assess positioning errors after every table position. Stereoscopic verification images and 6 DOF registration/positioning were performed and analyzed after each table rotation. HTT showed overall three-dimensional (3D) accuracy of 0.76 mm (SD0.46 mm) for the frameless technique, 0.87 mm (SD0.44 mm) for the frame-based technique for table positions at 0°. Rotating the couch to 90° and 270°, showed deviations of 0.83 mm (SD0.34 mm) and 0.62 mm (SD0.16 mm), respectively, if uncorrected. These deviations were reduced to 0.04 mm (SD0.03 mm) and 0.06 mm (SD0.04 mm) by image-guided corrections at 90° and 270°. The mean 3D intrafraction shift was 0.58 (SD0.42 mm) and comparable to frame-based techniques. By applying verification images during treatment (every 4.6 min SD 2.3 min), the mean 3D intrafraction shift decreased to 0.51 mm (SD0.13 mm). With proper immobilization and X-ray verification images during treatment, frameless radiosurgery can be delivered as accurate as the frame-based approach with comparable immobilization characteristics to the invasive head ring and can be considered to be a reliable alternative for radiosurgery treatments.