Translating expression profiling into a clinically feasible test to predict neuroblastoma outcome

Area of Science:

• Molecular biology and oncology
• Genomics and gene expression analysis
• Biomarker discovery and validation

Background:

• Neuroblastoma is a pediatric cancer with variable outcomes.
• Accurate prediction of patient prognosis is crucial for treatment stratification.
• Microarray technology provides comprehensive gene expression data but can be complex and costly.

Purpose of the Study:

• To evaluate the feasibility of using highly parallel quantitative real-time PCR (qPCR) for predicting neuroblastoma patient outcomes.
• To compare the performance of qPCR with microarray analysis for gene expression profiling and outcome prediction.
• To assess the utility of qPCR data for clinical risk stratification of neuroblastoma.

Main Methods:

• Gene expression profiles were generated for 63 neuroblastoma patients using both Affymetrix microarrays and qPCR on microfluidic cards (MFC).
• Key neuroblastoma marker genes (MYCN, NTRK1/TrkA) and top discriminating genes were quantified by qPCR.
• Bioinformatics analyses, including Prediction Analysis of Microarray (PAM) and Computerized Affected Sibling Pair Analyzer and Reporter (CASPAR), were applied to the qPCR data.

Main Results:

• The MFC-based qPCR method demonstrated high reproducibility.
• Outcome prediction using PAM on qPCR data achieved 80% accuracy, comparable to microarray results.
• qPCR data, analyzed with CASPAR, proved significant for assessing the risk of neuroblastoma relapse (P = 0.0006).

Conclusions:

• Multiplex real-time PCR is a feasible and promising approach for neuroblastoma outcome prediction.
• This method can simplify complex whole-genome array data, offering a more convenient and cost-effective tool for clinical application.
• qPCR holds potential for routine use in clinical settings for improved neuroblastoma risk assessment and patient management.