Review of image defined risk factors in localized neuroblastoma patients: Results of the GPOH NB97 trial

BACKGROUND: Recently, an international expert group proposed revision of the International Neuroblastoma Staging System (INSS). Localized disease can be classified as L1 without and as L2 with image defined risk factors (IDRF published in JCO 2005; 23:8483-8489). Our aim was to evaluate IDRF for the prediction of resectability, complications, and outcome. PROCEDURE: Records of 520 localized neuroblastoma patients of the NB97 trial were reviewed. Patients were retrospectively classified as having IDRF or not. A total of 366 evaluable patients were then analyzed for extent and complications of surgery and the prognostic value of IDRF. RESULTS: Any IDRF was present in 26/160 of stage 1, 49/113 of stage 2, and 64/93 of stage 3 patients. Complete primary resection was achieved in 156/227 patients without IDRF and 43/139 patients with IDRF (P < 0.001). The frequency of complications was higher if any IDRF was present: 37/139 versus 33/227 (P = 0.006). Lack of IDRF was associated with better event free survival (3-year-EFS 86 +/- 2% vs. 75 +/- 4%, P = 0.010), whereas overall survival was similar (3-year-OS 98 +/- 1% vs. 96 +/- 2%, P = 0.462). EFS clearly depended on INSS stage (3-year-EFS 93 +/- 2% in stage 1, 78 +/- 4% in stage 2, and 69 +/- 5% in stage 3, P < 0.001). OS was not different (3-year-OS 98 +/- 1% vs. 99 +/- 1% vs. 94 +/- 2%, P = 0.056). Multivariate analysis demonstrated an impact of INSS stage on EFS only. IDRF were not shown to be significant for predicting EFS or OS. CONCLUSIONS: IDRF were useful in predicting risk and completeness of operation. IDRF failed as independent risk predictors in localized neuroblastoma. INSS more precisely identified patients with poor prognosis.