Objective: To evaluate the predictive value of biomarkers of inflammation like phagocyte-related S100 proteins and a panel of inflammatory cytokines in order to differentiate the child with acute lymphoblastic leukemia (ALL) from juvenile idiopathic arthritis (JIA).
Study design: In this cross-sectional study, we measured S100A9, S100A12, and 14 cytokines in serum from children with ALL (n = 150, including 27 with arthropathy) and JIA (n = 236). We constructed predictive models computing areas under the curve (AUC) as well as predicted probabilities in order to differentiate ALL from JIA. Logistic regression was used for predictions of ALL risk, considering the markers as the respective exposures. We performed internal validation using repeated 10-fold cross-validation and recalibration, adjusted for age.
Results: In ALL, the levels of S100A9, S100A12, interleukin (IL)-1 beta, IL-4, IL-13, IL-17, matrix metalloproteinase-3, and myeloperoxidase were low compared with JIA (P < .001). IL-13 had an AUC of 100% (95% CI 100%-100%) due to no overlap between the serum levels in the 2 groups. Further, IL-4 and S100A9 had high predictive performance with AUCs of 99% (95% CI 97%-100%) and 98% (95% CI 94%-99%), respectively, exceeding both hemoglobin, platelets, C-reactive protein, and erythrocyte sedimentation rate.
Conclusions: The biomarkers S100A9, IL-4, and IL-13 might be valuable markers to differentiate ALL from JIA.
Keywords: S100A12; S100A9; arthropathy; childhood acute lymphoblastic leukemia; cytokines; interleukin-13; interleukin-4; juvenile idiopathic arthritis.
Copyright © 2023 The Author(s). Published by Elsevier Inc. All rights reserved.