Systemic Inflammatory Mediators Are Effective Biomarkers for Predicting Adverse Outcomes in Clostridioides difficile Infection

Clostridioides difficile infection (CDI) can result in severe disease and death, with no accurate models that allow for early prediction of adverse outcomes. To address this need, we sought to develop serum-based biomarker models to predict CDI outcomes. We prospectively collected sera ≤48 h after diagnosis of CDI in two cohorts. Biomarkers were measured with a custom multiplex bead array assay. Patients were classified using IDSA severity criteria and the development of disease-related complications (DRCs), which were defined as ICU admission, colectomy, and/or death attributed to CDI. Unadjusted and adjusted models were built using logistic and elastic net modeling. The best model for severity included procalcitonin (PCT) and hepatocyte growth factor (HGF) with an area (AUC) under the receiver operating characteristic (ROC) curve of 0.74 (95% confidence interval, 0.67 to 0.81). The best model for 30-day mortality included interleukin-8 (IL-8), PCT, CXCL-5, IP-10, and IL-2Rα with an AUC of 0.89 (0.84 to 0.95). The best model for DRCs included IL-8, procalcitonin, HGF, and IL-2Rα with an AUC of 0.84 (0.73 to 0.94). To validate our models, we employed experimental infection of mice with C. difficile. Antibiotic-treated mice were challenged with C. difficile and a similar panel of serum biomarkers was measured. Applying each model to the mouse cohort of severe and nonsevere CDI revealed AUCs of 0.59 (0.44 to 0.74), 0.96 (0.90 to 1.0), and 0.89 (0.81 to 0.97). In both human and murine CDI, models based on serum biomarkers predicted adverse CDI outcomes. Our results support the use of serum-based biomarker panels to inform Clostridioides difficile infection treatment.