Cytokine-augmented risk stratification for six-month incident nephritis in children with IgA vasculitis

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Background Children with IgA vasculitis (IgAV) may develop nephritis after initially presenting without kidney involvement. We evaluated whether a cytokine augmentation score (CAS) was associated with improved six-month incident nephritis risk stratification beyond clinical variables and routine inflammatory markers. Methods This single-center cohort screened children younger than 18 years with IgAV from January 2023 to June 2024 and used standardized six-month renal outcome ascertainment. Children meeting accepted IgAV classification criteria, with no nephritis during the baseline decision window, complete baseline biomarker data, and six-month renal outcome ascertainment formed the cohort. A routine inflammatory index score (RIIS) summarized z -transformed natural-log C-reactive protein-to-albumin ratio (CAR), neutrophil-to-lymphocyte ratio (NLR), and systemic immune-inflammation index (SII); CAS summarized z -transformed natural-log interleukin (IL)-6, tumor necrosis factor alpha (TNF-α), IL-17A, IL-18, and IL-23. Multivariable logistic regression compared a routine clinical plus RIIS model with a cytokine-augmented model; the area under the receiver operating characteristic curve (AUC), Brier score, calibration, bootstrap optimism correction, decision-curve net benefit, and sensitivity analyses were evaluated. Results Among 447 screened children, 337 were analyzed; 91 (27.0%) developed incident nephritis. CAS was independently associated with nephritis after adjustment for age, sex, symptom-to-blood-draw interval, gastrointestinal involvement, systolic blood-pressure percentile, subthreshold urinary warning signs, and RIIS (adjusted odds ratio per standard deviation (SD), 2.97; 95% confidence interval (CI), 2.01–4.39; p < 0.001). Adding CAS improved the AUC from 0.724 to 0.801 (Δ0.077; 95% CI, 0.035–0.124; p = 0.001), reduced the Brier score from 0.168 to 0.149, and yielded an optimism-corrected AUC of 0.782. Conclusions In this single-center derivation cohort, CAS improved internally evaluated discrimination, Brier score, and risk separation beyond clinical variables and RIIS. The model remains investigational pending external validation, assay standardization, and implementation testing.