Introduction: With the advancement and successful treatment of metastatic spinal cord disease, newer treatments are needed for the long-term survivors of recurrent disease. The lack of a standardized re-treatment regimen and the difficulty in delineating the tumor margins among patients who have received the treatment with metallic spinal fixation and conventional radiation are two of the challenges to be faced in recurrent metastatic spinal cord disease. In these patients, we applied hypofractionated stereotactic radiosurgery by defining the tumor margin with (18)F-fluorodeoxyglucose (FDG) positron emission tomography (PET).
Patients and methods: Three consecutive recurrent spinal metastasis patients underwent the CyberKnife treatment (Accuray, Inc., Sunnyvale, CA) from March 2004 to July 2004. A three-fraction schedule was applied at approximately 24 hour intervals. One patient had sarcoma and the other two patients had breast cancer. All patients had received previous conventional radiotherapy after operation ranging from 30 Gy to 45 Gy. CT-based planning was corrected by the FDG-PET hyperuptake area with the help of nuclear medicine. The mass responses were followed not only by MRI but also by FDG-PET, which was taken prior to treatment, and at one and six months after the treatment. The changes in standard uptake value (SUV) of serial PET were taken as a measure of response. To evaluate the relative SUV changes from different pretreatment values, we set a reduction index (RI), which represents the ratio of SUV change to pretreatment SUV.
Results: No significant complications were noted during treatment with a mean follow-up of 13.3 months. The tumor volume on CT-based planning was 2.2 times larger than that of the CT-PET combined planning in case 1 of paraspinal muscle invasion. But the tumor volumes showed minimal changes in the other cases, in which the metastatic tumors were confined to the vertebral bodies. The SUV one month after treatment showed variable decreases and the RI ranged from 0.07 to 0.7. However, the SUVs at 6 months were well correlated with the clinical results. One patient showed marginal failure and the other two patients showed local control of the tumor, as their RI values were 0.65 and 0.87, respectively.
Conclusion: To our knowledge, this is the first report using FDG-PET with radiosurgery in patients with recurrent spinal metastases hidden under metallic artifacts. The mass responses measured by SUV changes in FDG-PET correlated with the clinical results.