Purpose: To assess the relevance of N-Myc gene amplification (NMA) as a prognostic factor in localized neuroblastoma (NB) and to evaluate whether less intensive adjuvant treatment is advisable in infants without NMA.
Patients and methods: Assessment of NBs included clinical and imaging data to allow tumor-node-metastasis (TNM) staging, biologic determinations (N-Myc gene analysis), and standard histology and work-up to eliminate metastatic spread (metaiodobenzylguanidine [MIBG] scintigraphy and extensive bone marrow staging). Resectability was defined according to imaging findings. Chemotherapy was indicated in children older than 1 year at diagnosis who had postoperative residual disease or lymph node (LN) involvement, in infants with NMA, or as primary treatment in children with an unresectable NB, including dumbbell tumors. Radiotherapy was recommended in children older than 1 who presented with persistent gross residual disease at the end of therapy.
Results: Between 1990 and 1994, 316 consecutive children who presented with a localized NB were registered in the NBL 90 study. The median age was 12 months, and 42 patients had dumbbell tumors (13%). NMA was found in 22 of 225 assessable children (10%) and correlated with adverse prognostic indicators such as age older than 1 year, an abdominal primary tumor, a large tumor (T3), and unresectability. Among 186 children who had primary excision, five died of surgery-related complications. Primary chemotherapy was given to 130 patients, which allowed removal of the tumor in all but four. The 5-year overall survival (OS) and event-free survival (EFS) rates were, respectively, 91% and 84% with a median follow-up time of 36 months. The outcome of infants and older children was similar (P = .2). EFS of patients with resectable tumors was slightly better than with unresectable primary tumors (EFS, 89% v 78%; P = .02). In dumbbell NBs, neurologic recovery was achieved in 74% of cases that presented with symptoms, and initial laminectomy was avoided in 75% of children. In a univariate analysis, large tumors, high neuron-specific enolase (NSE) and lactate dehydrogenase (LDH) levels, positive LNs, macroscopic residue, and NMA adversely influenced outcome. In the multivariate analysis, NMA was the most powerful unfavorable predictive indicator: OS and EFS rates for these children were 36% and 32%, compared with 98% and 90% in nonamplified tumors (P < .001).
Conclusion: Our data confirm the overall good prognosis of localized NBs, even when unresectable. NMA is the most relevant adverse prognostic factor in localized NBs, and more intensive treatment should be investigated in these patients. Prospective studies of other biologic factors are warranted to tailor therapy more accurately. The EFS of children who underwent primary surgery was excellent, and further justifies elimination of adjuvant treatment provided they have no NMA. Despite the elimination of postoperative therapy, infants with non-NMA tumors have an excellent outcome, which suggests that initial chemotherapy can be further reduced in case of unresectable NBs.