GABAergic system gene expression predicts clinical outcome in patients with neuroblastoma.

Abstract

Neuroblastoma (NB) is a common childhood malignancy characterized by heterogeneous clinical behavior. The purpose of this study was to identify potential NB biomarkers that may improve outcome prediction. The suppression subtractive hybridization (SSH) technique was used to identify the genes differentially expressed between NB and control tissue. RNA isolated from 235 primary NB tumor samples obtained from the Children's Cancer Group was evaluated for expression of the candidate markers using quantitative reverse transcriptase polymerase chain reaction (Taqman assays). The association between the mRNA expression levels in the identified candidate genes and clinical outcome was evaluated. SSH analysis identified differential expression of members of the GABAergic gene family in NB. Lower levels of gamma-aminobutyric acid (GABA) receptor-associated protein (GABARAP) gene expression predict decreased survival among all patients. GABA(A) delta receptor subunit gene expression was predictive of a poor outcome among Evans stage IV-S patients. An index of five coexpressed GABA(A) receptor subunits was identified (GABA(A) profile [GAP score]). Patients with a higher GAP score (> -1) had a survival advantage. Multivariate analysis showed that GABARAP and GABA(A) alpha2 receptor subunit gene expression levels and GAP score remained predictors of clinical outcome after accounting for current prognostic indicators. Dysregulation of the GABAergic system may constitute a fundamental event in the development of NB, and assessment of GABAergic system gene expression could provide improved patient stratification and potential new therapies.