In order for gene expression profiling in blood to become a reliable and reproducible tool in large-scale investigations, a better understanding of intra- and interindividual variability comparing used methods is needed. Several studies have shown that the PAXgene system using whole blood samples results in higher variability of gene expression profiles and a decrease in expressed genes compared to PBMC-based methods [6-9]. However, Whitney et al. observed a higher variability of gene expression profiles in individuals with disease than among healthy individuals in blood, indicating the feasibility of using gene expression profiling in blood for disease detection and diagnosis [6].
