Subnetwork-Based Analysis of Chronic Lymphocytic Leukemia Identifies Pathways That Associate with Disease Progression,

The clinical course of patients with chronic lymphocytic leukemia (CLL) is heterogeneous. Whereas some patients develop aggressive disease requiring early treatment, others can have highly indolent disease and not require therapy for many years. Several prognostic factors have been identified that can stratify patients into groups that differ in their relative tendency for disease progression and/or survival. Microarray studies have highlighted differences in mRNA levels found between such CLL subgroups. Here, we pursued a subnetwork-based analysis of gene expression profiles to discriminate between groups of patients with disparate risks for disease progression. The clinical characterization of patients, blood-sample preparation, and microarray processing all follow the unified protocol implemented by the Microarray Innovations in LEukemia (MILE) program, which proposed standards for microarray-based assays in the diagnosis and sub-classification of leukemia. From an initial cohort of 130 patients, we identified 38 prognostic subnetworks that could predict the relative risk for disease progression requiring therapy from the time of sample collection (Fig. 1A). The prognostic power of these subnetworks then was validated on a second cohort of patients in the MILE study and on another set of CLL patients evaluated outside the MILE program (Fig. 1B). The identified subnetworks could assess the risk for requiring therapy at the time of tissue collection more accurately than established markers (Fig. 1C). Statistical analyses of these and the microarray data collected in prior studies revealed the greatest divergence in gene expression was observed using samples collected within 1 year of diagnosis. Thereafter there was increasing congruence in the expression levels of some subnetworks between patients over time. Moreover, the expression levels of such predictive subnetworks could evolve in patients with otherwise indolent disease characteristics to resemble those associated with patients found to have aggressive disease at diagnosis. These analyses suggest that degenerate pathways apparently converge into common pathways that are associated with disease progression. We conclude that, in addition to having predictive power, these identified subnetworks represent an array of pathways associated with disease progression. As such, these results have implications for understanding cancer evolution and for the development of novel treatment strategies for patients with CLL.