Validation of a 2-Gene Blood Test for Kawasaki Disease in Febrile Children

Abstract

Key Points Question Can a 2-gene blood test distinguish Kawasaki disease from other febrile illnesses in children? ab-zoi260441-1 Findings In this diagnostic study of 541 febrile children, expression ofandmeasured by quantitative polymerase chain reaction accurately identified Kawasaki disease. Performance was consistent across 5 independent cohorts, including incomplete Kawasaki disease, diverse febrile-control etiologies, and coronary artery phenotypes. ab-zoi260441-2 Meaning These findings suggest that a simple 2-gene blood test may provide an objective adjunct to clinical criteria for early Kawasaki disease diagnosis and reduce delays in initiating intervention. ab-zoi260441-3 key-points This diagnostic study validates a whole-blood quantitative polymerase chain reaction assay measuringandexpression for distinguishing Kawasaki disease from other pediatric febrile illnesses. teaser electronic Importance Kawasaki disease (KD) remains a clinical diagnosis without an objective molecular test. Early identification is critical to prevent coronary artery complications through timely intravenous immunoglobulin therapy. ab-zoi260441-4 Objective To validate a 2-gene whole-blood quantitative polymerase chain reaction (qPCR) assay measuringandexpression for distinguishing KD from other pediatric febrile illnesses. ab-zoi260441-5 Design, Setting, and Participants This multicenter diagnostic study was conducted in Taiwan and Shanghai, China. Patient blood samples were collected prospectively between 2012 and 2023 in Taiwan and between 2022 and 2023 in Shanghai and analyzed retrospectively from children younger than 8 years with KD and febrile controls (FCs) with viral, bacterial, or mixed infections. Data were analyzed from January 2022 to August 2025. ab-zoi260441-6 Main Outcomes and Measures Diagnostic accuracy of a prespecified 2-gene KD score derived from change in cycle threshold values normalized to glyceraldehyde 3-phosphate dehydrogenase, assessed by area under the receiver operating characteristic curve (AUC), sensitivity, specificity, predictive values, and likelihood ratios. ab-zoi260441-7 Results A total of 541 children (mean [SD] age, 3.7 [1.9] years; 300 [55.5%] male), including 243 children with KD and 298 febrile controls, were analyzed. The KD score achieved an AUC of 0.91 (95% CI, 0.88-0.94), with a sensitivity of 94% (95% CI, 93%-97%) and a specificity of 82% (95% CI, 78%-86%). The positive likelihood ratio was 5.12, and the negative likelihood ratio was 0.05. Performance was consistent across cohorts, including incomplete KD, diverse FC etiologies, and coronary artery phenotypes. The assay was implemented as a laboratory-developed test. Analytical validation demonstrated high linearity (> 0.99), precision (coefficient of variation <5%), and sample stability for up to 6 days at 4 °C or for 24 hours at room temperature. ab-zoi260441-8 Conclusions and Relevance This diagnostic study found that a 2-gene laboratory-developed whole-blood qPCR assay measuringandexpression accurately distinguished KD from other febrile illnesses using standard molecular platforms. Prospective evaluation in broader populations is warranted to determine its clinical utility for reducing diagnostic delay and coronary complications. ab-zoi260441-9