Mortality prediction in sepsis via gene expression analysis: a community approach

Timothy E. Sweeney; Thanneer Malai Perumal; Ricardo Henao; Marshall Nichols; Judith A. Howrylak; Augustine M.K. Choi; Jesús F. Bermejo-Martín; Raquel Almansa; Eduardo Tamayo; Emma E. Davenport; Katie L. Burnham; Charles Hinds; Julian Charles Knight; Christopher W. Woods; Stephen Francis Kingsmore; Geoffrey S. Ginsburg; Hector R. Wong; Grant Peter Parnell; Benjamin M. P. Tang; Lyle L. Moldawer; Frederick Alan Moore; Larsson Omberg; Purvesh Khatri; Ephraim L. Tsalik; Lara M. Mangravite; Raymond J. Langley

doi:10.1038/s41467-018-03078-2

BioinformaticsNature CommunicationsFeb 9, 2018

Peer-reviewed journal article

Mortality prediction in sepsis via gene expression analysis: a community approach

Timothy E. Sweeney iD, Thanneer Malai Perumal iD, Ricardo Henao iD, Marshall Nichols iD, Judith A. Howrylak, Augustine M.K. Choi iD, Jesús F. Bermejo-Martín iD, Raquel Almansa iD, Eduardo Tamayo iD, Emma E. Davenport iD, Katie L. Burnham iD, Charles Hinds iD, Julian Charles Knight iD, Christopher W. Woods iD, Stephen Francis Kingsmore iD, Geoffrey S. Ginsburg iD, Hector R. Wong iD, Grant Peter Parnell iD, Benjamin M. P. Tang iD, Lyle L. Moldawer iD, Frederick Alan Moore, Larsson Omberg iD, Purvesh Khatri iD, Ephraim L. Tsalik iD, Lara M. Mangravite iD, Raymond J. Langley iD

Abstract

Improved risk stratification and prognosis 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 three scientific groups were invited to independently generate prognostic models for 30-day mortality using 12 discovery cohorts (N=650) containing transcriptomic data collected from primarily community-onset sepsis patients. Predictive performance was validated in 5 cohorts of community-onset sepsis patients (N=189) in which the models showed summary AUROCs ranging from 0.765-0.89. Similar performance was observed in 4 cohorts of hospital-acquired sepsis (N=282). Combining the new gene-expression-based prognostic models with prior clinical severity scores led to significant improvement in prediction of 30-day mortality (p<0.01). These models provide an opportunity to develop molecular bedside tests that may improve risk stratification and mortality prediction in patients with sepsis, improving both resource allocation and prognostic enrichment in clinical trials.

PDF DOI