Identification of immunophenotypic signatures by clustering analysis in pediatric patients with philadelphia chromosome-positive acute lymphoblastic leukemia

Abstract

Detection of Philadelphia chromosome t(9;22) (Ph) in children with precursor-B-ALL (pB-ALL) is an adverse prognostic factor, thus leading to a high-risk protocol for treatment. RT-PCR is the gold-standard for the detection of this abnormality. Specific gene and protein expression signatures have recently been identified for genetic subclasses in childhood and adult ALL using gene profiling and flow cytometric analyses, respectively. Our aim is the characterization of Ph+ pB-ALL for a fast and cheap screening approach in routine immunophenotyping applied at diagnosis. Forty-one children with Ph+ and 99 Ph- newly diagnosed pB-ALL (AIEOP cohort) were analyzed. The expression level of 16 marker proteins was monitored by five color flow cytometry (FC) and quantified in terms of Geometric Mean Fluorescence (GMF). Computational analyses were applied to the patient cohort: we identified a Cluster A, including the majority of Ph+ patients (35/41), associated with upregulation of CD52, TdT, CD45, CD34, HLA-DR, CD33, and downregulation of CD38, CD24, CD58, CD22, CD19; a Cluster B+C gathers most of the Ph- patients (86/99) showing the opposite tendency for listed markers. The immunophenotypic method identifies Ph+ cases with a comprehensive accuracy of 86% providing a rapid and effectual screening method for the identification of Ph+ pB-ALL.