Gene expression profiles based on ZAP70 mRNA level.

Stamatopoulos B, Haibe-Kains B, Meuleman N, Bron D, Martiat P and Lagneaux L

Background: Gene expression profile is a powerful tool to better understand the biology, the clinical outcome and the molecular mechanism implicated in chronic lymphocytic leukemia (CLL). This disease presents an extremely variable clinical course with overall survival times ranging from months to decades. Therefore a plethora of prognostic factors which classified patients in poor or good behaviour have been investigated. Zeta-associated protein 70 (ZAP70) is one of the most promising prognostic factors to predict CLL evolution. Furthermore, we previously described a quantitative real-time PCR  (qPCR) method to measure ZAP70 and demonstrated its prognostic power (Stamatopoulos et al, Clin. Chem., 2007).

Aims: In this study, we compared gene expression profile of patients expressing high versus low ZAP70 mRNA level in order to find genes not only associated with prognosis but also with cell biology. We also confirmed some microRNA differentially expressed between these two groups and linked them to treatment-free survival (TFS) and overall survival (OS).

Methods: ZAP70 was evaluated by qPCR in a cohort of 108 patients ;  two groups of 7 patients were chosen in the top-20 of patients expressing high and low level of ZAP70 mRNA and their gene expression profiles were compared using Affymetrix technology. Selected genes were verified by qPCR in an extended patient cohort (n=85) with a median follow-up of 72 months. Adhesion/migratory capacities into a stromal microenvironment or in response to conditioned medium were also evaluated. Finally, we investigated some microRNA differential expression by qPCR in a cohort of 61 patients with a median follow-up of 74 months.

Results: 43 probe sets were differentially expressed with a FDR<10%, 135 with a P<0.001 and 932 with a P<0.05. Several of these genes were TFS and/or OS significant predictors: PDE8A and FCRL family genes were downregulated in ZAP70+ patients and can predict TFS and OS; ITGA4 mRNA was upregulated in ZAP70+ patients and can significantly predict OS. Moreover pathway analysis reveals an over-representation of adhesion/migration genes. CXCR4/SDF1-alpha pathway was one of them. We observed a downregulation of CXCR4 in stromal-adherent cells only in ZAP70+ patients indicating that only these patient cells can respond to microenvironment stimulus. Furthermore, ZAP70+ patient cells can significantly better adhere to fibronectin. MicroRNA comparison confirmed the differential expression of miR-29c and miR-223 and we showed for the first time that these two microRNA had a TFS and OS individual prognostic power.

Conclusions: This study identifies new prognostic factors (genes and microRNA) and shows the better adhesion/migratory capacities of ZAP70+ cells in their microenvironment explaining their better survival and the aggressiveness of the disease.