Table_1_Serum cytokine and chemokine profiles and disease prognosis in hepatitis B virus-related acute-on-chronic liver failure.docx

BackgroundHepatitis B virus-related acute-on-chronic liver failure (HBV-ACLF) has significant morbidity and mortality and is associated with the induction of cytokines/chemokines, which might contribute to the pathogenesis of liver injury. This study aimed to explore the cytokine/chemokine profiles of patients with HBV-ACLF and develop a composite clinical prognostic model.
MethodsWe prospectively collected blood samples and the clinical data of 107 patients with HBV-ACLF admitted to the Beijing Ditan Hospital. The concentrations of 40-plex cytokines/chemokines were measured in 86 survivors and 21 non-survivors using the Luminex assay. Discrimination between the cytokine/chemokine profiles in different prognosis groups was analyzed using the multivariate statistical techniques of principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). An immune-clinical prognostic model was obtained using multivariate logistic regression analysis.
ResultsThe PCA and PLS-DA indicated that cytokine/chemokine profiling could clearly distinguish patients with different prognoses. A total of 14 cytokines, namely, IL-1β, IL-6, IL-8, IL-10, TNF-α, IFN-γ, CXCL1, CXCL2, CXCL9, CXCL13, CX3CL1, GM-SCF, CCL21, and CCL23, were significantly correlated with disease prognosis. Multivariate analysis identified CXCL2, IL-8, total bilirubin, and age as independent risk factors that constituted the immune-clinical prognostic model, which showed the strongest predictive value of 0.938 compared with those of the Chronic Liver Failure Consortium (CLIF-C) ACLF (0.785), Model for End-Stage Liver Disease (MELD) (0.669), and MELD-Na (0.723) scores (p < 0.05 for all).
ConclusionThe serum cytokine/chemokine profiles correlated with the 90-day prognosis of patients with HBV-ACLF. The proposed composite immune-clinical prognostic model resulted in more accurate prognostic estimates than those of the CLIF-C ACLF, MELD, and MELD-Na scores.