GENE EXPRESSION PROFILING IN ISCHEMIC AND NONISCHEMIC CARDIOMYOPATHY
Johns Hopkins University
Background Despite our growing understanding of the pathophysiology and management of heart failure, there exist no strategies to individualize therapy using predictors of long-term prognosis and response to therapy. Gene expression analysis using microarray technology provides a phenotypic resolution not possible with standard clinical criteria and could offer insights into disease mechanisms and also identify markers useful for diagnostic, prognostic, and therapeutic purposes. Thus, the two major applications of this technology are gene discovery and molecular signature analysis. These two applications were explored in studies involving the two major forms of cardiomyopathy, ischemic and nonischemic (ICM and NICM, respectively). Methods For a gene discovery analysis, we compared the gene expression of 21 NICM and 10 ICM samples with that of 6 nonfailing (NF) hearts using Affymetrix U133A microarrays and Significance Analysis of Microarrays software. For molecular signature analysis, we identified and validated an etiology signature with Prediction Analysis of Microarrays software using 48 ICM and NICM myocardial samples obtained from different institutions and at different clinical stages. Results The gene discovery analysis demonstrated that compared to NF hearts, 257 genes were differentially expressed in NICM and 72 genes in ICM. Only 41 genes were shared between the two comparisons and an analysis of the gene subsets revealed shared and unique disease-specific gene expression between end-stage cardiomyopathy of different etiologies. The molecular signature analysis demonstrated that an etiology prediction profile accurately distinguished between ICM and NICM, was generalizable to iii samples from separate institutions, specific to disease stage, and unaffected by differences in clinical characteristics. Conclusions We have demonstrated that there are shared and distinct genes involved in the development of heart failure of different etiologies, and that a molecular signature can accurately identify etiology in cardiomyopathy. These findings highlight the utility of the two distinct applications of gene expression analysis, and support ongoing efforts to develop cause-specific therapies and expression profiling-based biomarkers in heart failure. The ultimate goal is individualized therapy, whereby a heart failure patient could, through gene expression analysis, be offered an accurate assessment of prognosis, and how individualized medical therapy could affect his or her outcome.
Gene expression, Microarray, Cardiomyopathy, Heart failure