Background: Angiogenin is a potent inducer of angiogenesis. We prospectively evaluated the prognostic significance of serum angiogenin from 204 consecutive non-Hodgkin lymphoma (NHL) patients diagnosed and treated in a single institution.
Methods: Serum angiogenin, VEGF, and bFGF concentrations at diagnosis were determined using a quantitative sandwich enzyme immunoassay technique. Kaplan-Meier survival curves were compared by the log-rank test. Multivariate survival analyses were performed using the parametric model of Weibull and the non-parametric proportional hazards model of Cox.
Results: Patients with a high serum angiogenin at diagnosis (>median; 401 ng/ml) had significantly lower 5-year survival rate than those with a low (≤ median) angiogenin (42% versus 63%, respectively; P = 0.0073). Serum angiogenin provided additional information to the International Prognostic Index (IPI) identifying a subgroup (serum angiogenin >median and IPI>1) with very poor prognosis (5-year survival 19%, P < 0.0001). In receiver operating characteristic (ROC) analyses the accuracy of the IPI to correctly classify patients with favourable or poor survival was improved from fair to good by complementing the IPI with serum angiogenin concentration. With patients who initially achieved complete response (CR) after chemotherapy, a high angiogenin at diagnosis (>median; relative risk (RR) 2.38; P = 0.0077) and an advanced tumour stage (III-IV; RR 2.41; P = 0.0087) were the only independent predictors for patients with unfavourable outcome although first responding well to therapy.
Conclusions: We conclude that elevated serum angiogenin surfaced as an independent predictor for failure in long-term treatment response and for poor overall survival in a series of 204 NHL patients, and might thus also complement the IPI in identifying the patients with particularly aggressive and/or treatment resistant disease.
Copyright © 2011 Elsevier Ltd. All rights reserved.