Neuroblastoma is the third most common childhood cancer after leukemias and cancer of the central nervous system. Long noncoding RNA MEG3 polymorphisms have been shown to confer cancer susceptibility; however, their roles in the genetic predisposition to neuroblastoma remain unclarified. To answer this question, we genotyped two MEG3 polymorphisms, rs7158663 G>A and rs4081134 G>A, in 392 neuroblastoma children and 783 controls by TaqMan method. The results showed that neither single locus nor the combination analysis supported an association between MEG3 polymorphism and neuroblastoma risk. Interestingly, we found that subjects carrying rs4081134 AG/AA genotypes significantly tended to develop neuroblastoma among subgroups with age >18 month (adjusted OR=1.36, 95% CI=1.01-1.84) and clinical stage III+IV disease (adjusted OR=1.47, 95% CI=1.08-1.99), when compared with reference group. In the combined analysis of MEG3 polymorphisms, we found that carriers of 2 risk genotypes were more likely to have higher risk of developing neuroblastoma than those with 0-1 risk genotype among children more than 18 months of age (adjusted OR=1.36, 95% CI=1.01-1.84, P=0.042), and with clinical stages III+IV disease (adjusted OR=1.47, 95% CI=1.08-2.00, P=0.014). Our data suggest MEG3 as a weak-effect neuroblastoma susceptibility gene. Well-designed studies with large sample studies are needed to further validate this finding.