Predicting pituitary tumor behavior remains a challenge, since the prognostic value of identified pathological markers has not yet been evaluated. Genome-wide transcriptome analyses allow the identification of molecular markers to highlight global changes in gene expression and enable the discovery of signaling pathways within complex biological networks. While several transcriptome studies comparing normal with tumoral pituitary as a whole or with respect to subtype have given interesting data concerning pituitary pathogenesis, none have considered pituitary tumor prognosis. Only one study to date, focusing on the pathological classification and progression of prolactin tumors, has identified some molecular markers with diagnostic and prognostic value. We reviewed data in the literature on human pituitary tumor transcriptome and conducted a meta-analysis on the expression of genes involved in cell growth, proliferation and the cell cycle. Based on our expertise, we evaluate the interests and the limitations of using this approach with human pituitary tumors.
2010 Elsevier Ireland Ltd. All rights reserved.