Neuroblastoma (NB) is the most common extracranial solid tumor of childhood and the third most common pediatric cancer. Although numerous factors including patient age, disease stage and genetic abnormalities have been shown to be predictive of outcome, the mechanisms responsible for the highly variable clinical behavior of this tumor remain largely unknown. In order to gain new insights into the biology of this tumor, we performed microarray analysis and compared the gene expression patterns of NB detected by mass screening, characterized by highly probable spontaneous regression, versus stage 4 NB with poor prognosis. The bioinformatics analysis revealed a set of 19 discriminatory genes that may play a significant role in the natural progression of the disease. Validation of these results and further mechanistic studies would shed new light on the biology of tumor progression, and provide new tools to predict clinical outcome in children with NB.