Phagocyte-related S100 proteins and cytokines in acute lymphoblastic leukemia and their prognostic value

Abstract

In this retrospective cohort study, we evaluated the level of biomarkers of inflammation like phagocyte-related S100 proteins and a panel of cytokines in 128 children with pre-B ALL and 22 with T-ALL. The biomarkers were evaluated at diagnosis and during antileukemic therapy (day 29 and after six months) and we evaluated their correlation with basic laboratory values. Further, for the children with pre-B ALL, we evaluated whether the biomarkers could predict the outcome of ALL expressed as minimal residual disease (MRD), relapse, and death. The levels of S100A9, S100A12, IL-1beta, IL-12p70, IL-13, IL-17, IL-18, and MPO serum levels increased significantly as chemotherapy was initiated. The difference was most pronounced for S100A9 and S100A12, which had strong positive correlations with the neutrophil counts. In contrast, TNF-alpha, IL-6, IL-10, CCL-2, MMP-3, and CD25 serum levels decreased after chemotherapy. Although none of these biomarkers appear to be an independent predictor of outcomes, in predictive models with MRD as the outcome, AUC increased from 76% (95% CI 68–84%) when using initial risk group stratification alone to 83% (95% CI 73–91%) in a multivariate predictive model including initial risk group stratification and the biomarkers S100A12, TNF-alpha, and IL-10.