2194 - Impact of Carbon Fiber vs. Titanium Spinal Instrumentation on Post-Operative Stereotactic Body Radiation Therapy Dose Delivery to Target Volumes and Critical Structures
Presenter(s)
D. Ross1, S. Enyew1, A. Qin1, M. C. LeCompte1, M. Khan2, A. Bydon3, K. Kebaish4, S. H. Lee4, L. R. Kleinberg1, D. Lubelski3, and K. J. Redmond5; 1Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 2Department of Radiology and Radiological Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 3Department of Neurosurgery, Johns Hopkins University School of Medicine, Baltimore, MD, 4Department of Orthopedic Surgery, Johns Hopkins University School of Medicine, Baltimore, MD, 5Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University, Baltimore, MD
Purpose/Objective(s): Post-operative spinal stereotactic body radiation therapy (SBRT) is being increasingly utilized in patients with spinal metastases for durable local control. Spinal instrumentation with titanium (T) results in significant artifact on MRI and heterogeneity of dose in treatment planning. Carbon fiber (CF) implants have become available. The purpose of this study is to compare imaging artifact and treatment plan quality in patients with T and CF spinal instrumentation.
Materials/Methods: Patients receiving post-operative spine SBRT at a single institution between 2017-2025 were retrospectively reviewed. Hardware composition was recorded. Patients were matched amongst the CF and T groups by location in the spine and number of vertebral levels. Dosimetric data including PTV coverage, dose to critical structures, minimum and maximum doses were collected. Artifact was quantified by contouring the volumes of hardware and surrounding distortion on MRI T2-weighted sequences and CT. The ratios of volume on MRI to CT were compared. Biologically effective dose (BED) was calculated using alpha/beta=10. A Wilcoxon rank-sum statistical analysis was performed.
Results: Thirty-three patients with 34 sites were included. Twenty-two had T and 11 patients (12 sites) had CF. The number of patients with 1, 2, 3, 4, and 5 vertebral levels involved for CF and T were: 1) 3/12 (25.0%) and 7/22 (31.8%), 2) 2 (16.7%) and 5 (22.7%), 3) 5 (41.7%) and 7 (31.2%), 4) 1 (8.3%) and 2 (9.0%), 5) 1 (8.3%) and 1 (4.5%). The median volumes of CF and T artifact on CT were 26.1cc (IQR 15.6-36.5) and 38.2cc (25.7-50.6) and on MRI were 49.8cc (21.4-78.2) and 113.3cc (69.3-157.4). Median ratios of MRI to CT volume for CF and T groups were 2.3 (1.9-2.6) and 2.8 (2.3-3.2) [p=0.031]. Median CTV volume for CF and T were 182.2cc (95.4-240.5) and 182.5cc (56.7-308.3). Median PTV volumes were 168.0cc (91.0-244.9) and 197.4cc (60.5-334.4). Median BED maximum PTV doses were 168.4 Gy (146.0-190.8) and 114.9 Gy (88.3-141.5) which were significantly higher in the CF group (p=0.02). BED PTV minimums were also significantly higher in the CF group with medians 23.1 Gy (21.1-25.1) and 16.1 Gy (11.8-20.4) [p=0.03]. Spinal cord (median 20.6 Gy vs. 17.7 Gy) and cord PRV (median 35.3 Gy and 31.6 Gy) maximum BED doses were not significantly different (p=0.40 and p=0.053). PTV coverage was not significantly different (82.6% and 88.0%, p=0.54).
Conclusion: These data demonstrate a significant artifact reduction with CF implants compared to T in addition to improved dosimetry on plan evaluation. Reduced artifacts allow superior target delineation and avoidance of invasive CT myelogram. Furthermore, prior studies suggest that minimum dose contributes to local control. Future studies are needed to evaluate if these planning benefits translate to improved local control.