Background

ZAP-70 expression has been used as a negative prognostic indicator for patients with chronic lymphocytic leukemia (CLL) (N Engl J Med 2003;348:1764; N Engl J Med 2004;351:893); however, its usefulness as a clinical test has been undermined by technical difficulties relating to the subjective cut-off threshold for positivity. In fact, analyses of samples from CLL patients enrolled in the US Intergroup Phase III trial, E2997, by standard procedures did not result in ZAP-70 (or any other analyte) providing significant prognostic information. We have previously developed a powerful flow cytometric amplification system to assess many important intracytoplasmic molecules including signaling and apoptotic pathway components. We hypothesized that the technology could reveal previously hidden relationships between outcome and molecular expression.

Methods

We analyzed 79 blood and marrow samples from CLL patients enrolled in the E2997 trial for the expression of ZAP-70 and 28 other analytes. The 28 other analytes included mostly signaling pathway molecules and apoptotic pathway components. Molecular expression levels determined by flow cytometry were analyzed for their relationship to the clinical data. CD19+ B lymphocytes from 42 healthy volunteers were also assessed as a comparison group.

Results

We found that CLL samples could be clearly stratified by ZAP-70 expression with a distribution having a peak-trough-peak configuration (Figure). In this way, a definitive threshold was obtained by using the trough as the cut-off. Interestingly, ZAP-70 expression in B cells from healthy persons was unimodal and it peaked in the trough of the distribution of the CLL samples (Figure). Using the ZAP-70 expression trough as a cutoff, unlike the E2997 study, we could delineate patient populations with distinct progression-free survival (p < 0.001) and overall survival (p = 0.004). Moreover, by multivariate analysis ZAP-70 expression was found to be an independent variable (p = 0.01) relating to progression-free survival along with 17p deletion and Rai staging.

By assessing the expression levels of the other 28 analytes stratified by ZAP-70 expression, we found statistically significant differences in 10 molecules which included phosphoantigens representing activation levels of various signaling pathways and apoptotic pathway constituents. These findings suggest that ZAP-70 stratification defines 2 distinct types of CLL cells.

Besides expression, we found highly significant levels of bivariate correlations between expression levels of the molecules examined including over 30 with correlation coefficients of 0.85 or greater. Interestingly, the correlations varied according to ZAP-70 expression. In samples with high levels of ZAP-70 expression, ZAP-70 was correlated with 13 other analytes with r > 0.5 whereas phospho-ZAP-70 was only correlated with 1 other analyte with the 0.5 threshold. Conversely, in samples with low levels of ZAP-70, phospho-ZAP-70 was correlated with 12 other molecules with r > 0.5 whereas ZAP-70 was correlated with no other analytes. An example showing that ZAP-70 was strongly correlated with Mcl-1 in samples with high levels of ZAP-70 but not in samples with low levels is presented below in the figure. The differential intermolecular relationships seen by ZAP-70 stratification support the possibility that ZAP-70 expression defines 2 different types of CLL cells with clinical outcome implications.

Conclusions

Our study has uncovered highly significant differential expression levels, intermolecular associations, and organization in CLL samples stratified by ZAP-70 expression. Since stratification by ZAP-70 gave significant prognostic information, we propose that our flow cytometric amplification technology permits characterization of the distinct molecular constitution of the 2 types of CLL cells and provides us with a unique opportunity to understand how the composition of leukemic cells translates into clinical outcome. Additionally, these data provide clues for the development of novel therapies.

Disclosures:

Kaplan:Pathfinder Biotech: Equity Ownership, Membership on an entity’s Board of Directors or advisory committees, Patents & Royalties, Research Funding. Kaye:Pathfinder Biotech: Employment.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution