Abstract
Abstract 4388
Chronic lymphocytic leukemia (CLL) is a heterogeneous disease; ZAP-70 protein expression and IgVH mutational status have shown to be strong associated and to offer important prognostic information.
Our aim was to determine gene expression profiles of 46 CLL patients divided into three classes: group one (n=26) with mutated IgVH and ZAP-70-, group two (n=12) with unmutated IgVH and ZAP-70+, and group three (n=8) included CLL patients with unmutated IgVH and ZAP-70-, or mutated IgVH and ZAP-70+ respectively. Afterwards, in a second phase of the study, 16 other patients were investigated. Finally, the purpose was to define prognostic biomarkers and the biological pathways related to CLL.
We determined gene expression profiles using Affymetrix HG U133 Plus 2.0 in CD19+ leukemic cells. Differentially expressed genes were detected using ANOVA and t-test adapted for microarray data analysis and corrected for multiple testing using false discovery rate p-values. Subjects were clustered in groups with similar expression signature using cluster analysis (K-means, Euclidean distance).
Statistical analysis revealed 154 differentially expressed probe-sets in the first (mutated IgVH and ZAP-70-) vs the second (unmutated IgVH and ZAP-70+) group, corresponding to 88 genes annotated in public databases. Interestingly, six genes were associated to the following biological pathways: MAPKsignaling (heat shock 70kD protein 8 HSPA8), B cell receptor signaling (ZAP-70, CKLF-like MARVEL transmembrane domain containing 3 CMTM3, dual adaptor of phosphotyrosine and 3-phosphoinositides DAPP1), Matrix Metalloproteinasis (transcription factor 20, TCF20), Apoptosis (X-linked inhibitor of apoptosis XIAP) and T cell receptor signaling (ZAP-70). In particular, ZAP-70, HSPA8, CMTM3 were significatively underexpressed while XIAP, TCF20 and DAPP1 were overexpressed in the first group of patients in comparison to the second group, respectively.
Based on the expression of the 88 genes identified in the comparison between the first and the second group of patients, the 8 patients of the third class were divided in two clusters: 5 subjects were more similar to the first class, while 3 subjects appeared to be more similar to the second one. In particular, cluster analysis revealed that the 46 patients were better partitioned in two rather than in three classes, based on their expression profiles. 16 additional subjects were independently analyzed in a second phase of the project. Based only on the expression of the 88 genes previously identified, all of them were correctly classified in group one and group two.
Further analysis was carried on the total of 62 subjects (n=46+16), dividing them in group A (n=17), showing deletion of 17p13 region and group B (n=45) without the deletion. Statistical analysis showed no correlation between groups A and B with respect to the previously defined group one, two and three. Moreover, no genes were identified as significantly differentially expressed in group A vs B.
In conclusion, our preliminary data revealed different gene expression signatures in B-cell chronic lymphocytic leukemia prognostic subgroups of patients, defined by IgVH mutational status and ZAP-70 expression.
The functional pathways related to: MAPK signaling, B cell receptor signaling, apoptosis and T cell receptor signaling may ultimately influence CLL biology. Gene expression profiling studies are in progress on larger series of CLL patients in order to assess the association of the molecular signature, based on the identified genes and their pathways, with respect to prognostic information.
No different gene expression signatures appeared considering CLL patients divided in two groups differing from the presence or the absence of deletion of 17p13 region respectively
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
