Dose Sculpting Intensity Modulated Radiation Therapy for Vertebral Body Sparing in Children with Neuroblastoma
Menée auprès de 34 patients pédiatriques atteints d'un neuroblastome paravertébral (âge médian : 4,3 ans ; 20 garçons et 14 filles), cette étude évalue l'effet, sur la croissance vertébrale, d'une technique de radiothérapie avec modulation d'intensité visant à épargner les vertèbres
Purpose : Long-term survivors of high-risk neuroblastoma have an increased incidence of musculoskeletal anomalies. Historically, the radiation field to the primary site was designed to cover adjacent vertebral bodies to minimize growth deformities. With intensity modulated radiotherapy (IMRT), one can shape dose distributions to spare vertebral bodies without sacrificing tumor coverage. We reviewed our institution’s experience with dose sculpting IMRT to assess its effect on vertebral body growth. Methods and Materials : From 2000 to 2011, 88 children with neuroblastoma underwent radiation at our institution. We included children with paravertebral tumors with at least 3 years of evaluable post-treatment imaging and excluded children who underwent spine re-irradiation prior to follow-up. If vertebral bodies could not be spared, we gave these "target" vertebral bodies at least 18 Gy. Thoracic and lumbar vertebral bodies were assessed separately. Dose data for target, spared, and internal control vertebral bodies were extracted. We used multivariate generalized estimating equation (GEE) modeling to assess the effect of dose and other clinical factors on vertebral body growth. Results : 34 patients (20 boys, 14 girls) met study criteria. Median age at start of radiation was 4.3 years; all but one had prior high-dose chemotherapy with stem cell rescue. Mean growth rates of target, spared, and control vertebral bodies (cm/body/year) were 0.027, 0.032, and 0.044 in thoracic spine and 0.033, 0.055, and 0.083 in lumbar spine. On multivariate GEE analysis, higher dose, older treatment age, male gender, and thoracic spine location were significantly associated with decreased vertebral body growth (p<0.0001, p<0.0001, p=0.007, and p<0.0001, respectively). Dose and spine location were significant in three-way interaction model (p<0.0001). Conclusion : Vertebral bodies spared by IMRT grew faster than target vertebrae. Regardless of intent to spare or target, multivariate analysis confirms that lower dose results in significantly increased growth rate. This technique should be investigated prospectively.