2565. A Novel Prognostic Gene Set for the Prediction of Severe Dengue. (26th November 2018)

• Record Type: Journal Article
• Authors: Robinson, Makeda L; Sweeney, Timothy E; Barouch-Bentov, Rina; Sahoo, Malaya K; Sanz, Ana Maria; Pu, Szu-Yuan; Ortiz, Eliana; Albornoz, Luis; Suarez, Fernando Rosso; Montoya, Jose G; Pinsky, Benjamin; Khatri, Purvesh; Einav, Shirit
• Is Part Of: Open forum infectious diseases. Volume 5(2018)Supplement 1
• Page Start: S72
• Page End: S72
• Publication Date: 2018-11-26
• Subjects: Communicable diseases -- Periodicals; Medical microbiology -- Periodicals; Infection -- Periodicals; 616.9
• DOI: 10.1093/ofid/ofy209.173
• Languages: English
• ISSNs: 2328-8957

Abstract

Abstract: Background: There is an urgent need for the identification of biomarkers predictive of severe dengue. Single cohort transcriptomic studies have not yielded a parsimonious gene set predictive of severe dengue. We hypothesized that integration of gene expression data from heterogeneous patient populations with dengue infection would yield a set of conserved genes that is predictive of severe dengue and generalizable across cohorts. Methods: Ten dengue gene expression datasets were identified in publicly available microarray repositories. A novel integrated multicohort platform was used to detect differentially expressed gene transcripts between uncomplicated and severe dengue patients and validate the identified putative signature in silico and prospectively in a new cohort of 34 dengue patients in Colombia. Dengue diagnosis was made by NS1 antigen and anti-DENV IgM antibody and confirmed by RT-PCR assays, ELISA, and IgG avidity measurements. The expression level of the signature genes was measured via microfluidic qRT-PCR assays in blood samples collected longitudinally during the course of illness. Results: Using the multicohort analysis to analyze 446 peripheral blood samples of patients with dengue infection from 7 publicly available gene expression datasets, we identified a 20 gene set that predicts the development of severe dengue. We in silico validated the diagnostic power of this gene set to separate severe dengue from dengue with or without warning signs in 3 independent datasets composed of 84 samples with a global area under the ROC curve (AUC) of 0.80 [95% CI 0.68–0.88]. We prospectively validated the gene set in a new cohort composed of 34 dengue patients from Colombia with an AUC of 0.89 [95% CI 0.81–0.97]. The severity scores measured in patients with severe dengue progressively declined in longitudinal samples. Conclusion: Our data indicate that the identified 20 gene signature predicts the development of severe dengue in patients prior to its onset and suggest that dengue infection itself triggers this host response. These findings may provide new insight into the pathogenesis of severe dengue and have implications for the development of a prognostic molecular assay to identify patients at risk to develop severe dengue. Disclosures: T. E. Sweeney, Inflammatix, Inc.: Employee and Shareholder, Salary. P. Khatri, Inflammatix, Inc: Scientific Advisor and Shareholder, Licensing agreement or royalty.