Microarray Gene Expression Studies Reveal Potential Classifier Genes That Allow Prediction of Response to Interferon in Chronic Myeloid Leukemia Patients.

Interferon has been extensively used to treat CML patients over the past years. However, not all patients respond and the reasons for this heterogeneity are largely unknown. Microarray-based gene expression analysis enables the study of thousands of genes in one single experiment and has been shown to be a useful tool to assign samples to specific categories. The goal of the study was to identify gene/s whose expression will permit prediction of response to interferon. Blood samples (n=9) or leukapheresis products (n=11) were collected from 20 CML chronic phase patients at diagnosis and all patients were subsequently treated with a-interferon. Responders were defined as patients achieving a complete or major cytogenetic response at 12 months or earlier after initiation of interferon (n=9) and patients with no cytogenetic response 5 to 18 months after interferon start were assigned to the non-responder group (n=11).

Mononuclear cells from two healthy blood donors were pooled and the RNA used as a reference sample. RNA was extracted from mononuclear or leukapheresis cells and gene expression profiles were analyzed using cDNA microarrays with approximately 7500 genes. Genes were organized according to the absolute difference in the median expression levels between the two groups and the top 20 genes were then used in an “all pair” selection procedure to identify the gene/s with the highest prediction accuracy. Six genes were selected by this procedure: RNAse2, PRG2, NRGN, LTF, JARID1A and DEFA4. These genes had an estimated error rate in predicting interferon response of 0,13 with a Fisher linear discriminant test.

Our results show that gene expression analysis may permit the identification of genes whose expression levels may be useful in the prediction of response to interferon in CML diagnostic samples.