Validation and utility of ARDS subphenotypes identified by machine-learning models using clinical data: an observational, multicohort, retrospective analysis

Distinct respiratory tract biological pathways characterizing ARDS molecular phenotypes

“…We used a previously validated three-variable classifier model to determine ARDS molecular phenotype based on plasma IL-8, bicarbonate, and Protein C levels 14 . When plasma biomarkers were unavailable from the day of TA collection (n=5), we used a recently described and validated machine learning model to assign phenotype 11,15 . 10 out of 41 ARDS subjects (24%) had hyperinflammatory ARDS, which is consistent with the proportion of hyperinflammatory subjects observed in previous studies [5][6][7][8][9] .…”

“…For these subjects, we used a validated clinical classifier model to assign phenotype. 11,15 RNA sequencing Following enrollment, TA was collected and stored in RNAse free conditions as previously described 60 . Metagenomic next generation sequencing (mNGS) of RNA was performed on TA specimens using an established sequencing pipeline.…”

“…Two clinically distinct molecular phenotypes of ARDS (termed "hyperinflammatory" and "hypoinflammatory") have been identified using latent class analysis of clinical and plasma biomarker data in five clinical trial cohorts and three observational studies 1,[5][6][7][8][9][10][11] . The hyperinflammatory phenotype is characterized by elevated plasma inflammatory cytokines (IL-8, IL-6, TNFr-1), lower plasma Protein C and bicarbonate, and higher 90-day mortality compared to the hypoinflammatory phenotype.…”

Inhaled nitric oxide: role in the pathophysiology of cardio-cerebrovascular and respiratory diseases

“…This makes the clinical evidence available not conclusive so far. Hopefully, the current insights of the ARDS stratification into phenotypes—that are biologically and clinically different features within the same definition of ARDS [ 219 , 220 ]—may help the understanding of the effects of numerous pharmacological treatments in ARDS including iNO [ 221 ].…”

Cutting the Gordian knot of heterogeneity: Can integrated systems biology modelling rescue critical care syndromes?

“…Nevertheless, as a proof of concept, this study is a novel and disruptive contribution to a growing body of literature seeking to identify homogeneous patient subtypes within heterogenous clinical syndromes by leveraging modern data science techniques. 9 , 10 The promise of such precision medicine approaches is two-fold: (1) to advance biological understanding of disease and, more importantly, (2) provide prognostic and therapeutic information to directly impact patient care. Confirming the utility of biological subtypes through prospective trials will be a key step in translating hypothetical frameworks into valuable clinical tools.…”