Neuroblastoma is a childhood cancer with remarkably divergent tumour behaviour and the presence of metastatic disease is a powerful predictor of adverse outcome. However, the importance of the involvement of specific metastatic sites or overall metastatic burden in determining outcome has not been fully explored. We analysed data from the International Neuroblastoma Risk Group database for 2250 patients with stage 4 disease treated from 1990 to 2002. Metastatic burden was assessed using a 'metastatic site index' (MSI), a score based on the number of metastatic systems involved. Overall, involvement of bone marrow, bone, lung, central nervous system, or other sites was associated with worse outcome. For patients aged ≥18 months, involvement of liver had the greatest impact on outcome and was associated with tumour MYCN amplification and adrenal primary and lung metastases. Increased MSI was associated with worse outcome and higher baseline ferritin/lactate dehydrogenase. We explored the impact of initial treatment approach on these associations. Limiting the analysis to patients allocated to protocols including stem cell transplant (SCT), there was no longer an association of outcome with metastatic involvement of any individual system or increasing MSI. Thus, treatment escalation with SCT (and the addition of differentiating agents to maintenance therapy) appears to have provided maximal benefit to patients with greatest metastatic disease burden. These findings underscore the importance of examining prognostic factors in the context of specific treatments since the addition of new therapies may change or even negate the predictive impact of a particular variable.
Keywords: Antineoplastic protocols; Liver; Metastasis; Myeloablative chemotherapy; Neuroblastoma; Prognosis; Stem cell transplantation.
Copyright © 2016 Elsevier Ltd. All rights reserved.