Early expression of IL-10, IL-12, ARG1, and NOS2 genes in peripheral blood mononuclear cells synergistically correlate with patient outcome after burn injury

Abstract

Background: No methods exist to rapidly and accurately quantify the immune insult created by burn injuries. The development of a rapid, noninvasive clinical biomarker assay that evaluates a burn patient's underlying immune dysfunction and predicts clinical outcomes could transform burn care. We aimed to determine a set of peripheral biomarkers that correlates with clinical outcomes of burn patients.

Methods: This prospective observational study enrolled two patient cohorts within a single burn center into an institutionally approved institutional review board study. Blood draws were performed <48 hours after injury. Initial unbiased immune gene expression analysis compared 23 burn patients and 6 healthy controls using multiplex immune gene expression analysis of RNA from peripheral blood mononuclear cells. We then performed confirmatory outcomes analysis in 109 burn patients and 19 healthy controls using a targeted rapid quantitative polymerase chain reaction. Findings were validated and modeled associations with clinical outcomes using a regression model.

Results: A total of 149 genes with a significant difference in expression from burn patients compared with controls were identified. Pathway analysis identified pathways related to interleukin (IL)-10 and inducible nitric oxide synthase signaling to have significant z scores. quantitative polymerase chain reaction analysis of IL-10, IL-12, arginase 1 (ARG1), and inducible nitric oxide synthase demonstrated that burn injury was associated with increased expression of ARG1 and IL-10, and decreased expression of nitric oxide synthase 2 (NOS2) and IL-12. Burn severity, acute lung injury, development of infection, failure of skin autograft, and mortality significantly correlated with expression of one or more of these genes. Ratios of IL-10/IL-12, ARG1/NOS2, and (ARG1-IL-10)/(NOS2-IL-12) transcript levels further improved the correlation with outcomes. Using a multivariate regression model, adjusting for patient confounders demonstrated that (ARG1-IL-10)/(NOS2-IL-12) significantly correlated with burn severity and development of acute lung injury.

Conclusion: We present a means to predict patient outcomes early after burn injury using peripheral blood, allowing early identification of underlying immune dysfunction.

Level of evidence: Prognostic/Epidemiological; Level II.