Objectives: To find and validate generalizable sepsis subtypes using data-driven clustering. Design: We used advanced informatics techniques to pool data from 14 bacterial sepsis transcriptomic datasets from eight different countries (n = 700). Setting: Retrospective analysis. Subjects: Persons admitted to the hospital with bacterial sepsis. Interventions: None. Measurements and Main Results: A unified clustering analysis across 14 discovery datasets revealed three subtypes, which, based on functional analysis, we termed “Inflammopathic, Adaptive, and Coagulopathic.” We then validated these subtypes in nine independent datasets from five different countries (n = 600). In both discovery and validation data, the Adaptive subtype is associated with a lower clinical severity and lower mortality rate, and the Coagulopathic subtype is associated with higher mortality and clinical coagulopathy. Further, these clusters are statistically associated with clusters derived by others in independent single sepsis cohorts. Conclusions: The three sepsis subtypes may represent a unifying framework for understanding the molecular heterogeneity of the sepsis syndrome. Further study could potentially enable a precision medicine approach of matching novel immunomodulatory therapies with septic patients most likely to benefit.

It is not understood how immune inflammation influences the pathogenesis of severe acute respiratory syndrome (SARS). One area of strong controversy is the role of interferon (IFN) responses in the natural history of SARS. The fact that the majority of SARS patients recover after relatively moderate illness suggests that the prevailing notion of deficient type I IFN-mediated immunity, with hypercytokinemia driving a poor clinical course, is oversimplified. We used proteomic and genomic technology to systematically analyze host innate and adaptive immune responses of 40 clinically well-described patients with SARS during discrete phases of illness from the onset of symptoms to discharge or a fatal outcome. A novel signature of high IFN-α, IFN-γ, and IFN-stimulated chemokine levels, plus robust antiviral IFN-stimulated gene (ISG) expression, accompanied early SARS sequelae. As acute illness progressed, SARS patients entered a crisis phase linked to oxygen saturation profiles. The majority of SARS patients resolved IFN responses at crisis and expressed adaptive immune genes. In contrast, patients with poor outcomes showed deviated ISG and immunoglobulin gene expression levels, persistent chemokine levels, and deficient anti-SARS spike antibody production. We contend that unregulated IFN responses during acute-phase SARS may culminate in a malfunction of the switch from innate immunity to adaptive immunity. The potential for the use of the gene signatures we describe in this study to better assess the immunopathology and clinical management of severe viral infections, such as SARS and avian influenza (H5N1), is therefore worth careful examination.

Introduction: Human host immune response following infection with the new variant of A/H1N1 pandemic influenza virus (nvH1N1) is poorly understood. We utilize here systemic cytokine and antibody levels in evaluating differences in early immune response in both mild and severe patients infected with nvH1N1. Methods: We profiled 29 cytokines and chemokines and evaluated the haemagglutination inhibition activity as quantitative and qualitative measurements of host immune responses in serum obtained during the first five days after symptoms onset, in two cohorts of nvH1N1 infected patients. Severe patients required hospitalization (n = 20), due to respiratory insufficiency (10 of them were admitted to the intensive care unit), while mild patients had exclusively flu-like symptoms (n = 15). A group of healthy donors was included as control (n = 15). Differences in levels of mediators between groups were assessed by using the non parametric U-Mann Whitney test. Association between variables was determined by calculating the Spearman correlation coefficient. Viral load was performed in serum by using real-time PCR targeting the neuraminidase gene. Results: Increased levels of innate-immunity mediators (IP-10, MCP-1, MIP-1β), and the absence of anti-nvH1N1 antibodies, characterized the early response to nvH1N1 infection in both hospitalized and mild patients. High systemic levels of type-II interferon (IFN-γ) and also of a group of mediators involved in the development of T-helper 17 (IL-8, IL-9, IL-17, IL-6) and T-helper 1 (TNF-α, IL-15, IL-12p70) responses were exclusively found in hospitalized patients. IL-15, IL-12p70, IL-6 constituted a hallmark of critical illness in our study. A significant inverse association was found between IL-6, IL-8 and PaO2 in critical patients. Conclusions: While infection with the nvH1N1 induces a typical innate response in both mild and severe patients, severe disease with respiratory involvement is characterized by early secretion of Th17 and Th1 cytokines usually associated with cell mediated immunity but also commonly linked to the pathogenesis of autoimmune/inflammatory diseases. The exact role of Th1 and Th17 mediators in the evolution of nvH1N1 mild and severe disease merits further investigation as to the detrimental or beneficial role these cytokines play in severe illness.

Improved risk stratification and prognosis prediction in sepsis is a critical unmet need. Clinical severity scores and available assays such as blood lactate reflect global illness severity with suboptimal performance, and do not specifically reveal the underlying dysregulation of sepsis. Here, we present prognostic models for 30-day mortality generated independently by three scientific groups by using 12 discovery cohorts containing transcriptomic data collected from primarily community-onset sepsis patients. Predictive performance is validated in five cohorts of community-onset sepsis patients in which the models show summary AUROCs ranging from 0.765–0.89. Similar performance is observed in four cohorts of hospital-acquired sepsis. Combining the new gene-expression-based prognostic models with prior clinical severity scores leads to significant improvement in prediction of 30-day mortality as measured via AUROC and net reclassification improvement index These models provide an opportunity to develop molecular bedside tests that may improve risk stratification and mortality prediction in patients with sepsis.