Hector R Wong, Natalie Z Cvijanovich, Nick Anas, Geoffrey L Allen, Neal J Thomas, Michael T Bigham, Scott L Weiss, Julie C Fitzgerald, Paul A Checchia, Keith Meyer, Michael Quasney, Mark Hall, Rainer Gedeit, Robert J Freishtat, Jeffrey Nowak, Shekhar S Raj, Shira Gertz, Jocelyn R Grunwell, Christopher J Lindsell

We previously derived and validated the Pediatric Sepsis Biomarker Risk Model (PERSEVERE) to estimate baseline mortality risk in children with septic shock. The PERSEVERE biomarkers are serum proteins, selected from among the proteins directly related to 80 mortality risk assessment genes. The initial approach to selecting the PERSEVERE biomarkers left 68 genes unconsidered. Determine if the 68 previously unconsidered genes can improve upon the performance of PERSEVERE and provide biological information regarding the pathophysiology of septic shock. We reduced the number of variables by determining the biological linkage of the 68 previously unconsidered genes. The genes identified through variable reduction were combined with the PERSEVERE-based mortality probability to derive a risk stratification model for 28-day mortality using Classification and Regression Tree methodology (n = 307). The derived tree, PERSEVERE-XP, was then tested in a separate cohort (n =77). Variable reduction revealed a network consisting of 18 mortality risk assessment genes related to tumor protein 53 (TP53). In the derivation cohort, PERSEVERE-XP had an area under the receiver operating characteristic curve (AUC) of 0.90 (95% C.I.: 0.85 to 0.95) for differentiating between survivors and non-survivors. In the test cohort, the AUC was 0.96 (95% C.I.: 0.91 to 1.0). The AUC of PERSEVERE-XP was superior to that of PERSEVERE. PERSEVERE-XP combines protein and mRNA biomarkers to provide mortality risk stratification with possible clinical utility. PERSEVERE-XP significantly improves upon PERSEVERE and suggests a role for TP53-related cellular division, repair, and metabolism in the pathophysiology of septic shock.