Biomarkers of Diastolic and Systolic Heart Failure

Abstract

s S139 Canadian Cardiovascular Society (CCS) Highlighted Poster HEART FAILURE HIGHLIGHTED POSTER SESSION Thursday, October 17, 2013 126 BIOMARKERS OF DIASTOLIC AND SYSTOLIC HEART FAILURE Z Hollander, KK Lam, V Chen, JE Wilson-McManus, RT Ng, SJ Tebbutt, BM McManus, A Ignaszewski, T Anderson, J Ezekowitz, GY Oudit, JR Dyck Vancouver, British Columbia BACKGROUND: Heart failure (HF) is a modern day epidemic with a rising prevalence and affects over 23 million global citizens. Methods to identify or diagnose a patient with HF include clinical history, physical exam, imaging and blood tests. The most commonly used blood tests are markers of natriuretic peptides, however, there is a large diagnostic “grey zone” in natriuretic peptides that renders them less useful as a stand-alone HF diagnostic test. In addition, they have not been shown to distinguish between diastolic HF (DHF) and systolic HF (SHF). METHODS: A discovery cohort comprised of 38 DHF patients, 54 SHF patients, and 22 healthy controls (HC) were enrolled in the Alberta HEART initiative. Plasma and PAXgene whole blood samples were analyzed at the PROOF Centre of Excellence with iTRAQ and MRM mass spectrometry and Affymetrix GeneST microarrays, respectively. Samples from a separate replication cohort, again from Alberta Heart (35 DHF, 77 SHF, and 36 HC), were analyzed with MRM mass spectrometry. The PROOF Centre biomarker development pipeline was used to identify novel blood test content by comparing DHF with SHF or comparing HC with HF patients. The computational analysis strategy consisted of prefiltering, univariate and multivariate tests, and classification. The performance of the sets of genes and proteins was assessed in the replication cohort. RESULTS: Proteomic biomarkers of HF were identified in the discovery cohort. When these protein panels were tested in the replication cohort, the area under the receiver operating characteristics curve (AUC) was 0.93. The NT-proBNP cut-off between DHF and HC was determined within the discovery cohort. An analysis combining the cohort-specific NT-proBNP cutoff with a set of novel proteins revealed an increase in performance characteristics (Figure), particularly for the specificity (from 34% to 80%) withNRI of 49%. Panels combining genes and proteins were able to distinguish between DHF and SHF patients. Cross-validation results using the discovery cohort data are promising for differentiating DHF from SHF using a combination of blood-based genes and proteins in women (AUC1⁄40.87) and in men (AUC1⁄4 0.83). CONCLUSION: Novel biomarkers of HF that can improve on the performance characteristics of NT-proBNP have been identified and replicated. Biomarker panels containing genes and proteins have been identified for discriminating between DHF and SHF. All results require further replication and validation in larger cohorts from patients outside of Alberta. 127 DECISION-MAKING ABOUT GOALS OF CARE FOR HOSPITALIZED PATIENTS WITH HEART FAILURE (DECIDE-HF): A PILOT STUDY N Aleksova, C Demers, PH Strachan, J Hoefs, M Yous, J MacIver, J Downar, M Wang, R Fowler, DK Heyland, HJ Ross, J You