Integrated host-microbe plasma metagenomics for sepsis diagnosis in a prospective cohort of critically ill adults

• DOI: 10.1038/s41564-022-01237-2
• Publication Date: 2022-10-20T16:05:38Z
• Authors: Katrina L. Kalantar; Lucile Neyton; Mazin Abdelghany; Eran Mick; Alejandra Jauregui; Saharai Caldera; Paula Hayakawa Serpa; Rajani Ghale; Jack Albright; Aartik Sarma; Alexandra Tsitsiklis; Aleksandra Leligd...; Stephanie A. Chri...; Kathleen Liu; Kirsten N. Kangel...; Carolyn Hendrickson; Pratik Sinha; Antonio Gomez; Norma Neff; Angela Pisco; Sarah B. Doernberg; Joseph L. Derisi; Michael A. Matthay; Carolyn S. Calfee; Charles R. Langelier
• Keywords: Host response

We carried out integrated host and pathogen metagenomic RNA DNA next generation sequencing (mNGS) of whole blood (n = 221) plasma 138) from critically ill patients following hospital admission. assigned into sepsis groups on the basis clinical microbiological criteria. From whole-blood gene expression data, we distinguished with non-infectious systemic inflammatory conditions using a trained bagged support vector machine (bSVM) classifier (area under receiver operating characteristic curve (AUC) 0.81 in training set; AUC 0.82 held-out validation set). Plasma also yielded transcriptional signature several genes previously reported as biomarkers, bSVM diagnostic (AUC 0.97 0.77 Pathogen detection performance mNGS varied site infection. To improve virus, developed secondary transcriptomic 0.94 0.96 combined microbial features to develop an model that identified 99% microbiologically confirmed cases, predicted 74% suspected 89% indeterminate cases. In summary, suggest integrating profiling broad-range nucleic acid is promising tool for diagnosis.