<b><i>Introduction:</i></b> Overall survival of patients with end-stage renal disease (ESRD) remains poor. Oxidative stress is one of the major risk factors associated with mortality in this patient group. As glutathione S-transferases (GST) are well-established antioxidants, we hypothesized that a model including GST gene polymorphisms, oxidative damage byproducts and cell adhesion markers has a prognostic role in ESRD patient survival. <b><i>Methods:</i></b> A prospective study of 199 patients with ESRD on haemodialysis was conducted. GST genotype, oxidative stress byproducts and cell adhesion molecules were measured in plasma. Multivariate Cox regression and Kaplan-Meier survival analyses were performed to test the predictive ability of these parameters in the 8-year follow-up period. <b><i>Results:</i></b> GSTM1-null genotype was associated with significantly shorter overall (HR 1.6, <i>p</i> = 0.018) and cardiovascular-specific (HR 2.1, <i>p</i> = 0.010) survival. Oxidative stress byproducts (advanced oxidation protein products [AOPP], prooxidant-antioxidant balance [PAB], malondialdehyde [MDA]) and cell adhesion molecules (soluble vascular cell adhesion molecule-1 [sVCAM-1] and soluble intercellular adhesion molecule-1 [sICAM-1]) demonstrated a significant predictive role in terms of overall and cardiovascular survival. When 6 biomarkers (GSTM1 genotype, high AOPP/PAB/MDA/­sVCAM-1/sICAM-1) were combined into a scoring model, a significantly shorter overall and cardiovascular survival was observed for patients with the highest score (<i>p</i> &#x3c; 0.001). <b><i>Conclusion:</i></b> We identified a novel panel of biomarkers that can be utilized in predicting survival in ESRD patients. This biomarker signature could enable better monitoring of patients and stratification into appropriate treatment groups.