Patient immobilization for stereotactic radiosurgery in the treatment of malignancies of the cervical spine: 9-point mask versus non-9-point mask immobilization system.

Purpose/Objective(s): Stereotactic radiosurgery with image guidance is becoming widely adopted for the treatment of spinal metastases. The rapid dose fall off achieved with radiosurgery can be exploited for therapeutic benefit (e.g. to respect the spinal cord tolerance). However, even small deviations in position may result in clinically meaningful outcomes. We sought to investigate the initial setup accuracy and intrafractional motion between a 9-point mask and a non-9-point mask immobilization system for stereotactic radiosurgery in the treatment of malignancies of the cervical spine. Purpose/Objective(s): We identified 15 patients with known or presumed malignant disease involving the cervical spine at our institution. Radiosurgery was delivered on a linear accelerator with cone beam CT (CBCT) image guidance. Patient immobilization was achieved with either a 9-point mask affixed to the treatment table or a stereotactic head mask with the patient lying in a blue bag (non-9-point mask). All patients were treated on a six degree of freedom couch. An initial localization CBCT was obtained for each patient and matched to the bony anatomy. Positional corrections were made followed by an initial verification CBCT. Further positional corrections with image verification were made for deviations greater than 1 mm (vertical, longitudinal, lateral) or 1° (rotation, pitch, roll). Radiosurgery was otherwise initiated. A mid-verification CBCT was obtained midway through treatment to assess for intrafractional motion. Positional corrections were again made if the 1 mm/1° tolerance was exceeded followed by a mid-shift verification CBCT as needed. The second half of the treatment was then delivered. Data on all positional deviations at the time of initial localization, initial verification, mid-verification, and mid-shift verification were recorded and analyzed. Results: All 15 patients underwent radiosurgery without incident. The largest positional deviations were noted at the time of initial localization. Initial localization and verification were more accurate with the 9-point mask immobilization system. Only with respect to roll was the non-9-point mask immobilization system more accurate for initial localization. At the time of mid-verification, the mean positional deviations with both immobilization systems were similar and within 1 mm/1° across all variables. Mid-shift verifications were more accurate with the 9-point mask immobilization system. Conclusion: A 9-point mask immobilization system achieves high positional accuracy and reproducibility for stereotactic radiosurgery in the treatment of malignancies of the cervical spine. Positional deviations midway through treatment were similar between the 2 immobilization systems suggesting that intrafractional motion may be a function of positional accuracy achieved prior to treatment delivery. Long-term data will be needed to determine whether these results translate into clinically meaningful outcomes