|| Checking for direct PDF access through Ovid
Prognostic biomarkers are useful in allogeneic stem cell transplantation (SCT) to predict survival and relapse outcomes. We sought to derive a prognostic scoring system, which augmented the predictive power of the disease risk index (DRI) by incorporating biomarkers and validating their significance after SCT. The outcomes of overall survival (OS) and relapse were assessed with non-relapse mortality (NRM) treated as a competing risk to relapse. Six hundred and two patients were identified through a retrospective analysis of allogeneic SCT recipients for haematological malignancy between 2000 and 2013 in a single centre. Multivariate analysis confirmed the significant predictors of OS pre-SCT were serum ferritin > 1000 μg/L (hazard ratio (HR) 1.94, 95% comorbidity index (CI): 1.44-2.60), Hb < 100 g/L (HR 1.71, 95% CI: 1.27-2.30) and albumin < 30 g/L (HR 2.65, 95% CI: 1.30-5.40). In combination with DRI, these biomarkers significantly improved the Harrell's C statistic (excluding biomarkers: C = 0.60, 95% CI: 0.57-0.64; with biomarkers: C = 0.65, 95% CI: 0.62-0.69, P < 0.001). Four prognostic groups were derived at the pre-SCT time point: Group 1 (Scores 0-1, n = 180, HR = 1 (ref)), Group 2 (Scores 2-5, n = 298, HR 2.7, 95% CI: 1.8-3.9), Group 3 (Scores 6-7, n = 87, HR 4.5, 95% CI: 3.0-6.9) and Group 4 (Scores 8-10, n = 9, HR 13.4, 95% CI: 5.9-30.2). These prognostic models were also predictive of relapse and NRM and remained valid at day 100, 12 months and 24 months post SCT.