Microarray Gene Expression Signature Indicates ARSD As a New Marker Associated with Igvh Mutational Status, ZAP-70 and Disease Progression in CLL,

Abstract 3447

Several studies have focused on the identification of biological markers that influence the clinical heterogeneity of Chronic lymphocytic leukemia (CLL). Until now IGVH and ZAP-70 are the most relevant prognostic markers in CLL suggesting the separation of two patient subgroups: with good MTZAP-70⁻ (mutated and ZAP-70 negative) and poor UMZAP-70⁺ (unmutated and ZAP-70 positive) prognosis.

Gene expression profiling in CLL has been demonstrated to be useful for the development of new biomarkers as prognostic factors able to predict the clinical course, the response and resistance to therapy.

The aim of this study has focused on the identification of different molecular signatures capable of revealing new potential molecular predictors for prognostic assessment or genetic risk and intriguing biological pathways.

We have determined gene expression profiling of B cells in 112 CLL patients divided into three classes: the first with MT and ZAP-70⁻, the second with UM and ZAP-70⁺, and the third included both UM ZAP-70⁻ and MT ZAP-70⁺ using Affymetrix HG-U133 Plus 2.0.

Among the set of 65 genes identified by microarray analysis we have found: AGPAT2, APP, ARSD, CHPT1, CRY1, DCLK2, LPL, MBOAT1, P2RX1, RIMKLB, ZAP-70, ZNF66 whose expressions were higher in second class (UMZAP-70⁺) while ADAM29, EGR3, FGL2, FUT8, NRIP1, TGFBR3, YPEL1 showed higher expression levels in first class (MTZAP-70⁻). Both analysis of differential expression and cluster analysis revealed that CLL patients were better partitioned in two rather than in three classes, based on their expression profiles.

We have also noted that the expression signature differentiating between MTZAP-70⁻ and UMZAP-70⁺ B cells from CLL patients showed that gene coding for enzymes which regulate lipid metabolism (sphingolipid/glycerolipid/glycerophospholipid) ARSD, LPL, MBOAT1, CHPT1, AGPAT4, AGPAT2, PLD1 were overexpressed in UMZAP-70⁺ compared to MTZAP-70⁻.

In particular we focused on ARSD as a new gene in CLL whose expression was consistently higher in UMZAP-70⁺ compared to MTZAP-70⁻ CLL B cells.

Western blotting experiments showed that the expression level of ARSD protein was 1.8 times higher in UMZAP-70⁺ compared to MTZAP-70⁻ B cells (p-value=0.0002), whereas ARSD protein levels in B cells from healthy donors (CTRL) were 2.03 (p-value=0.0006) and 3.73 (p-value=0.001) times lower compared to MTZAP-70⁻ and UMZAP-70⁺, respectively. The marked increase in ARSD expression in UMZAP-70⁺ compared to MTZAP-70⁻ and CTRLs was statistically significant as shown in Figure 1.

View largeDownload slide

View largeDownload slide

Close modal

At present our attention has been focused on ARSD role in predicting the entry in therapy. By means of a ROC analysis carried out with the Youden's method to determine the cut-off value we have found it at 99, so that we have pragmatically set it at 100. Even if the patient number is limited the difference in treatment-free survivorship between the two groups (above and below 100) was statistically significant (p = 0.0205) and the cut-off value used as prognostic index showed 66.7% sensitivity and positive predictive value and 75.0% specificity and negative predictive value. Differences in treatment-free survival are showed in figure 2.

Figure 2

View largeDownload slide

Treatment-free survival

Figure 2

View largeDownload slide

Treatment-free survival

Close modal

In conclusion 65 genes were differentially expressed in MTZAP-70⁻ vs UMZAP-70⁺ CLL patients; among the 65 genes 7 were involved in lipid metabolism, in particular ARSD being identified in sphingolipid pathway. Western blotting experiments detected significantly higher ARSD levels in UMZAP-70⁺ compared to MTZAP-70⁻ and CTRLs B cells.

Future studies will assess the role of ARSD as a prognostic factor in clinical trials combined with IGVH and ZAP-70 as well as the sphingolipid metabolism as a putative new biological mechanism in CLL.