O3243: Pathologic Fractures in Patients with Neuroblastoma Impacts Overall Survival

Background: Neuroblastoma is the third most common solid tumor in children. It is estimated that 45-50% of new Neuroblastoma patients are characterized as high-risk with a 5-year survival rate of 61% based on Children’s Oncology Group’s Risk Grouping. The presence of bone metastases in high-risk neuroblastoma may lead to pathologic fractures, however, the risk factors and outcomes are not well described.

Methods: Retrospective review of a prospectively collected database of children with high-risk neuroblastoma. High-risk patients without metastatic bone disease were excluded. Patient demographics, data related to pathologic fracture, and disease burden were collected. The disease burden was measured based on the number of osseous metastatic sites including the skull, shoulders, ribs/chest, thoracic/lumbar spine, humerus, radius/ulna, pelvis, femur, and tibia/fibula. Metastatic bone disease was recorded as unilateral or bilateral. Sites noted as ‘Unilateral’ counted as one area of involvement, while bilateral involvement was counted as two. These numbers were then summed to measure the number of areas of osseous metastases. Patients with a pathologic fracture were matched to patients without a fracture based on age, gender, and metastasis to the site of pathologic fracture in a 2 to 1 ratio. Areas of osseous metastases, and if the patient was deceased was recorded for both control and pathologic fracture cases. We computed significance using a conditional logistic regression and the Mantel-Haenszel method.

Results: 362 patients with neuroblastoma had metastases to the bone. 11 patients (3.03%) suffered 16 pathologic fractures after osseous metastases, of which 3 patients suffered multiple fractures. 12 of the 16 fractures occurred in long bones (i.e., femur, humerus, tibia/fibula), 7 of which were diaphyseal and 5 were metaphyseal. Operative treatment was necessary for the treatment of 3 of the pathologic fractures. Based on a conditional logistic regression, there was not a significant relationship between an increase in the number of osseous metastases and the risk for pathologic fracture when compared to controls (p = 0.94). 9 of the 11 patients were deceased at the time of the study, and the other 2 patients were in remission at the time of this study. Based on the Mantel-Haenszel Method, there was a significant difference in the proportion of death among patients with a pathologic fracture compared to their matched controls (p < 0.05).

Conclusion: Pathologic fractures among patients with Neuroblastoma are rare. Results indicate that an increase in overall presence of metastatic bone disease does not correlate with an increased risk for pathologic fracture. However, a pathologic fracture is a sign of poor prognosis. Future studies should analyze risk factors related to lesion size and anatomical location.