PV QA 3 - Poster Viewing Q&A 3
TU_24_3108 - Application of Deformable Image Registration to Quantify the Effect of Spinal Radiation During Childhood
Tuesday, October 23
1:00 PM - 2:30 PM
Location: Innovation Hub, Exhibit Hall 3
Application of Deformable Image Registration to Quantify the Effect of Spinal Radiation During Childhood
E. D. Scher1, R. E. Vatner2, and K. T. Huang1; 1University of Cincinnati, Cincinnati, OH, 2University of Cincinnati and Cincinnati Children's Hospital Medical Center, Cincinnati, OH
Purpose/Objective(s): Spinal growth is complex, and the effect of spinal radiation during childhood is dependent upon patient age, gender, and radiation dose. We investigate the application of feature-based deformable registration as a tool for quantitative longitudinal analysis of vertebral body growth in pediatric patients after abdominal radiotherapy. Materials/Methods: A retrospective study of all pediatric patients treated with abdominal radiotherapy in our department from 2012-2016 was performed (n=16). Of these, three had pre- and post-treatment CT imaging available for analysis. Automated feature-based deformable registration was performed on the CT images from simulation and follow-up for each patient, and checked manually to confirm accurate registration of the spine. Vertebral body growth was quantified using displacement vectors derived from maps of relative volumetric voxel-based axial growth using in-house software. This was compared using an unpaired t-test to published expected normal growth of thoracic and vertebral bodies values for age and gender. Results: The three analyzed patients were treated at 1 month, 6 years, and 13 years of age, for a neuroblastoma, rhabdomyosarcoma, and desmoplastic round cell tumor, respectively. Median thoracic radiation doses were 4.34, 7.61, and 26.27 Gy, respectively. The observed in-field thoracic relative growth rate per year was 0.020±0.013, 0.007±0.011, 0.009±0.006 mm, compared with the age matched expected growth rates of 0.183±0.005, 0.028±0.001, and 0.046±0.001 mm, respectively (p < 0.0001). Median lumbar radiation doses were 4.38, 26.71, and 31.91 Gy, respectively. The observed in-field lumbar relative growth rate per year was 0.025±0.008, 0.014±0.000017, 0.013±0.006 mm, and the expected growth rates were 0.430±0.156, 0.023±0.003, and 0.042±0.004 mm, respectively (p < 0.0001). Conclusion: We present a novel application of deformable image registration for measurement of axial vertebral body growth in pediatric patients exposed to low dose spinal radiation during abdominal radiotherapy. This technique allows for precise growth analysis of individual vertebral bodies and correlation of radiation dose to magnitude of impaired growth. Impaired spinal growth was observed even after low doses of radiotherapy (< 4 Gy), and was more pronounced in younger patients. These results suggest a benefit from radiotherapy plans with even marginal sparing of vertebral body dose.
Author Disclosure: E.D. Scher: resident physician; University of Cincinnati. R.E. Vatner: None. K.T. Huang: None.