Abstract Background Heart failure is a complex clinical syndrome that occurs at the end stage of heart disease. It is a major health problem in western countries with an overall population prevalence of 2–3% which rises sharply to 10–20% at 75 years of age. Despite advances in therapy for heart failure, improvment of clinical outcomes remains a challenge for physicians: half of patients die within 4 years of diagnosis and 40% of patients, who have been admitted to hospital are dead or re-admitted to hospital within 1 year. The aim of this study was to find plasma proteins that predict differences in clinical response from standard therapy in patients with heart failure. Methods Patients with heart failure who met inclusion and exclusion criteria were recruited. Uptitration of angiotensin-converting-enzyme-inhibitors and β blockers was done over 6 months. Patients were followed up for clinical events such as death and admission to hospital for heart failure within the next 24 months. We compared plasma proteins in 50 patients who responded to standard treatment (responders) with 50 patients who died or were re-admitted to hospital (non-responders). Plasma samples were pooled and depleted of 14 high abundance proteins and then reduced and alkylated, before digestion with trypsin to peptides. Peptides were analysed on two-dimensional liquid chromatography coupled to a tandem mass spectrometry in high definition mode which used a high pH reverse phase liquid chromatography in the first step before separation with conventional low pH reverse phase columns in the second step. Findings 434 proteins were identified in the plasma of patients with heart failure. 137 proteins in both responders and non-responders were significantly overexpressed or underexpressed (p<0·05), with 97 upregulated and 40 downregulated. Several of these proteins have the potential to become novel biomarkers for prediction of treatment response. They can be classified according to five pathobiological processes in heart failure: (1) neurohormones—hepatocyte growth factor like protein and insulin like growth factor II; (2) vascular system—intercellular adhesion molecule 2, sex hormone binding globulin, and prostaglandin H2 D isomerase; (3) inflammation—C-reactive protein, and mannose binding protein C; (4) matrix and cellular remodelling—retinol binding protein 4; and (5) cardiorenal system—cystatin C. Interpretation The discovery of novel biomarkers in this study will not only help to understand the pathophysiology of heart failure better, but also lead to the development of a more personalised approach in predicting treatment response to guide medical therapy. For example, in patients with heart failure who have abnormally increased concentrations of retinol-binding protein 4, a marker for matrix and cellular remodelling, a drug for reducing collagen deposition could be useful. In this way, unnecessary treatment in non-responders could be avoided (with implications for cost and adverse effects) and novel therapeutic targets could be identified for design of therapies to improve outcomes. Funding European Union FP7 Project.