Pediatric neuroblastoma with intraspinal extension: the role of surgical management

OBJECTIVE Neuroblastoma with spinal involvement accounts for up to 30% of pediatric spinal tumors and can cause profound neurological deficits. Chemotherapy is the preferred treatment option, but in select patients resection may be indicated. The goal of this study was to identify preoperative factors that led to early surgical intervention, with a specific emphasis on identifying differences on long-term neurological function and spinal deformity in the recent treatment era. METHODS A retrospective chart review was performed on all children diagnosed with neuroblastoma at a single institution from 2007 to 2020. Patient demographics, symptoms (motor deficit and sphincter dysfunction), and tumor characteristics (e.g., ¹²³I metaiodobenzylguanidine [MIBG] avidity, MYCN amplification, chromosomal abnormality, pathology, catecholamine secretion, and stage) were recorded. Spine involvement included neural or vertebral extension, spinal cord compression, and/or T2 signal change on MRI. Survival, neurological status (motor deficit, sphincter dysfunction), and spine deformity at last follow-up were compared using univariate and multivariate analyses. The variables that contributed to neurological and deformity outcome were assessed with binomial logistic and linear regression models using R software. RESULTS Seventy-seven of the 160 patients with neuroblastoma had spinal neuroblastoma, meaning either bone metastases alone (n = 43) or intraspinal extension with or without neurological deficit (n= 34). Most patients with spinal neuroblastoma were treated with chemotherapy and/or radiation therapy (97% and 57%, respectively). Resection of the spinal tumor was performed in 14 (18%) patients, all of whom also received chemotherapy. Between the surgical and nonsurgical patients, no baseline demographic differences were found. However, surgical patients were more likely to present with either motor deficits (50% vs 5%, p = 0.0011) or bladder/bowel dysfunction (14% vs 0%, p 0.035), and a shorter median time to onset of neurological symptoms (33 vs 80 days, p = 0.0096). Surgical patients also had a significantly shorter median overall survival (33.0 vs 54 months, p = 0.014). Of the 14 patients who underwent spine surgery, 2 patients underwent surgery at the time of diagnosis while the remaining 12 underwent initial chemotherapy followed later by resection. The 2 patients who underwent initial surgery had excellent outcomes, with neither long-term motor or bowel/bladder deficits nor spinal deformity. CONCLUSIONS Surgical patients had shorter overall survival. However, the 2 patients with radiographic evidence of cord compression and acute neurological symptom onset who underwent initial, immediate surgery within 3 days of diagnosis had fewer long-term neurological deficits than surgical patients who underwent initial trials of chemotherapy. Thus, acute decompression may provide benefit in carefully selected patients with acute neurological deficits and cord compression on imaging.