Impact of extent of resection on survival in high-risk neuroblastoma: A systematic review and meta-analysis.

ABSTRACT

BACKGROUND: Despite the development of new treatment options, the prognosis of high-risk neuroblastoma patients is still poor. Many studies designed to elaborate the association between the extent of resection (EOR) and outcome have reported conflicting results. We performed a meta-analysis to assess whether greater EOR is associated with improved overall survival (OS) and event-free survival (EFS) in patients with high-risk neuroblastoma. METHODS: Embase, PubMed, Cochrane library, and conference proceedings were searched between March 10 and October 1, 2017. Studies of pediatric patients with newly diagnosed high-risk neuroblastoma comparing various EOR and presenting objective overall or event-free survival data were included. Primary outcomes were relative risk (RR) for mortality at 3 and 5 years. Secondary outcomes were 3-year and 5-year EFS rates. RESULTS: 19 retrospective studies including a total of 2358 cases were identified. Compared with subtotal resection (STR), patients who underwent gross total resection (GTR) had significantly decreased mortality at 3 years (RR, 0.69; 95% CI, 0.58-0.82; P < 0.001; I2 = 27%) and 5 years (RR, 0.70; 95% CI, 0.60-0.82; P < 0.001; I2 = 38%). A similar decrease was revealed in the 3-year risk for mortality for STR compared with biopsy (RR, 0.71; 95% CI, 0.53-0.95; P = 0.02; I2 = 0%). When comparing any resection with biopsy, resection group also showed a decreased risk for mortality at 3 years (RR, 0.66; 95% CI, 0.53-0.83; P < 0.001; I2 = 8%) and 5 years (RR, 0.67; 95% CI, 0.50-0.91; P = 0.009; I2 = 61%). With respect to the risk ratio for EFS, there were no significant differences in any comparisons. CONCLUSION: This literature highlights the importance of "extent of resection" in treating high-risk neuroblastoma, and when feasible, the currently available evidences in favor of the use of GTR for high-risk neuroblastoma for reducing 3- and 5-year mortality. LEVEL OF EVIDENCE 3A.