BACKGROUND:To enhance biological understanding of ARDS, pneumonia, and sepsis and to accelerate therapeutic development in these areas, the National Institutes of Health developed the ARDS, Pneumonia, and Sepsis (APS) Consortium. RESEARCH QUESTION:Is the APS Consortium study rapidly generating data and biospecimens from a large cohort of critically ill adults with ARDS, pneumonia, and sepsis that will facilitate phenotyping of these syndromes? STUDY DESIGN AND METHODS:The APS Consortium Phenotyping Study is a multicenter, longitudinal, prospective, observational cohort study aimed at enrolling 4,000 critically ill adults with ARDS, pneumonia, sepsis, or a combination thereof over 4 years. Data and biospecimens are collected to characterize many aspects of each participant's chronic health, acute illness, and long-term recovery to facilitate phenotyping, that is, subclassifying ARDS, pneumonia, and sepsis into precise biologically based subsets with shared pathophysiologic characteristics. Feasibility of the study was assessed by evaluating the first 1,000 participants in terms of recruitment pace, participant characteristics, biospecimen collection, and proportion with confirmed ARDS, pneumonia, and sepsis based on expert adjudication. RESULTS:The first 1,000 participants were recruited ahead of schedule in < 13 months. Median age was 64 years, 75% received vasopressors, 50% received invasive mechanical ventilation, and 25% died in the hospital within 4 weeks of enrollment. Biospecimen collection rates were high, with 99% of participants with blood samples, 98% with upper respiratory swabs, 37% with lower respiratory samples, 80% with urine samples, and 65% with gastrointestinal samples. Expert adjudication resulted in 40% classified with ARDS, 52% classified with pneumonia, and 89% classified with sepsis. INTERPRETATION:The APS Consortium Phenotyping Study is producing a cohort of critically ill adults with ARDS, pneumonia, and sepsis with high severity of disease and a rich set of data and biospecimens. The study will continue to full enrollment of 4,000 participants. CLINICAL TRIAL REGISTRATION:ClinicalTrials.gov; No.: NCT06521502; URL: www. CLINICALTRIALS:gov.

Due to its nonspecific clinical criteria, sepsis is clinically, microbiologically, pathophysiologically and immunologically highly heterogeneous. Consequently, despite hundreds of clinical trials, no host-targeted therapy has been shown to be ubiquitously efficacious, leading investigators to pursue more precision-based approaches for enriching sepsis populations through the identification of subgroups or phenotypes. Here, we review the myriad domains in which heterogeneity is observed in sepsis and the challenges and opportunities they offer to improve outcomes. We review current strategies used by investigators leveraging novel biological measurements and/or computational algorithms to identify more homogeneous subgroups either based on pathogen or host characteristics or both. Finally, we review some of the most promising recent advances that seek to bring these complex and innovative discoveries to the bedside to facilitate precision medicine in sepsis.

BACKGROUND:Acute respiratory distress syndrome (ARDS) is a clinically defined, biologically heterogeneous condition with no proven disease-modifying therapies. Retrospective analyses have identified two biologically distinct subphenotypes (hyperinflammatory and hypoinflammatory) of ARDS, with differing outcomes and responses to therapy. Rapid identification of these subphenotypes in an actionable timeframe has previously not been possible. The PHIND study aimed to prospectively identify these subphenotypes and to demonstrate differing 60-day mortality. METHODS:The PHIND study was a prospective, multicentre, observational cohort study conducted in intensive care units (ICUs) within the National Health Service in the UK and the Health Service Executive in Ireland. Adult patients aged 18 years and older with ARDS or acute hypoxaemic respiratory failure (AHRF) were enrolled within 72 h of onset of the syndrome. Eligible patients were required to be receiving invasive mechanical ventilation, non-invasive ventilation, or high-flow nasal oxygen. Plasma interleukin (IL-6) and soluble TNF receptor-1 (TNFR1) were quantified at enrolment using a near-patient benchtop immunoanalyser (Randox multiSTAT) with a run time of approximately 1 h. Together with plasma bicarbonate measured from an arterial blood sample, these values were used to prospectively determine subphenotypes on an individual patient basis using a validated parsimonious logistic regression model. The primary outcome was 60-day mortality. The study was registered on ClinicalTrials.gov, NCT04009330. FINDINGS:Between Nov 22, 2019, and Sept 28, 2023, 1853 patients from 30 centres were screened for eligibility. Of these, 1328 were excluded and 525 were recruited into the study, with 512 individuals included. 308 (60%) patients were male, 204 (40%) were female, and mean age was 57·0 years (SD 15·1). 443 (87%) patients were white, 18 (4%) were Black, and 16 (3%) were Asian. 490 were subphenotyped using the near-patient assay: 89 (18%) were classified as hyperinflammatory and 401 (82%) as hypoinflammatory. The primary outcome of 60-day mortality was measured in 486 patients after four patients withdrew consent for confirmation of vital status. 60-day mortality was significantly higher in the hyperinflammatory group (45 [51%] of 88) than in the hypoinflammatory group (111 [28%] of 398; risk ratio 1·8 [95% CI 1·4-2·4], p<0·0001). After adjustment, hyperinflammatory patients had increased odds of 60-day mortality (adjusted odds ratio 2·7 [95% CI 1·6-4·4], p=0·0002). INTERPRETATION:Rapid identification of ARDS inflammatory subphenotypes using a near-patient assay was feasible and associated with many clinical characteristics and outcomes consistent with those described in earlier retrospective studies, including mortality, prevalence of sepsis, and incidence of metabolic acidosis. These findings support the implementation of precision medicine approaches in ARDS and the urgent need for prospective, subphenotype-stratified interventional trials. FUNDING:Innovate UK, Randox Laboratories, and Belfast Health & Social Care Trust.

BACKGROUND:Inflammatory phenotypes of acute respiratory distress syndrome (ARDS) predict outcomes and can respond differently to treatment strategies. We aimed to establish whether these phenotypes differ in respiratory mechanics and in response to lung-protective ventilation strategies. METHODS:In this retrospective cohort study, data from two cohorts were harmonised. Patients with moderate-to-severe ARDS with oesophageal manometry data from the EPVent-2 trial (14 hospitals across the USA and Canada) and a retrospective cohort at Beth Israel Deaconess Medical Center (Boston, MA, USA) were merged and lung mechanics were compared. Patients had to be aged 18 years or older, have moderate to severe ARDS, and be monitored with oesophageal manometry. To analyse the primary outcome of 60-day mortality after ARDS onset, we used multivariable Cox models for each inflammatory phenotype to study the associations between measures of lung-protective ventilation (driving pressure, transpulmonary driving pressure, and end-expiratory transpulmonary pressure) and 60-day mortality in all patients who had complete data for all variables. FINDINGS:Between Jan 1, 2008, and Jan 31, 2024, 5778 patients were assessed for eligibility (200 in the EPVent-2 cohort and 5578 in the BIDMC cohort). Of these patients, 890 were included in this study cohort (200 from the EPVent-2 trial and 690 from the retrospective cohort), of whom 424 (48%) had the hyperinflammatory phenotype and 466 (52%) had the hypoinflammatory phenotype. 232 (55%) patients in the hyperinflammatory group and 136 (29%) patients in the hypoinflammatory group died within 60 days (p<0·0001). The effects on 60-day mortality were more pronounced among patients with the hypoinflammatory phenotype than the hyperinflammatory phenotype for high respiratory system driving pressure (≥15 cm H2O; adjusted hazard ratio 2·01 [95% CI 1·39-2·91] vs 1·46 [1·11-1·94]; pinteraction=0·033) and high transpulmonary driving pressure (≥12 cm H2O; 2·36 [1·64-3·39] vs 1·18 [0·84-1·60]; pinteraction=0·0010). In addition, having an end-expiratory transpulmonary pressure within plus or minus 2 cm H2O was protective among the hypoinflammatory (0·66 [0·46-0·93]) but not the hyperinflammatory phenotype (0·97 [0·73-1·27]). Excess mortality among the hyperinflammatory phenotype was mediated by extrapulmonary organ failure (proportion mediated 46% [+17 to +79]) but not respiratory failure (0% [-3 to +4]). INTERPRETATION:Our findings suggested that in ARDS, the association between lung-protective mechanical ventilation and 60-day mortality is greater in patients with the hypoinflammatory phenotype than the hyperinflammatory phenotype, therefore patients with hypoinflammatory ARDS could be an important target population for enrichment of future clinical trials. However, our findings do not support different ventilation strategies based on phenotype. Although both phenotypes present with similar lung mechanics, extrapulmonary organ failure is the key driver of excess mortality among patients with the hyperinflammatory phenotype. FUNDING:Société Française d'Anesthésie-Réanimation, the University Hospital of Montpellier, Philippe Foundation, the Department of Anesthesia at Beth Israel Deaconess Medical Center, and Jeffrey and Judy Buzen.

Biological heterogeneity in sepsis and related systemic inflammatory syndromes has constrained therapeutic advancement and reduced the efficiency of conventional trial designs. Reproducible host response-derived sepsis subgroups recur across clinical, protein biomarker-based, and transcriptomic dimensions. A practical route to the implementation and evaluation of real-time subgrouping is a 2-stage pipeline that combines assessment of sepsis subgroup eligibility with subsequent classification. Priorities for future work are embedding classifiers and diagnostic platforms with high validity into clinical workflow so that prospective clinical trials can test their clinical utility.

BACKGROUND Sepsis encompasses considerable biological and clinical heterogeneity. Previously, 2 phenotypes (“hyperinflammatory” and “hypoinflammatory”) have been consistently identified within sepsis via latent class analysis. These phenotypes differ in their biological features, clinical outcomes, and therapeutic responses to interventions. Prior studies of sepsis heterogeneity have focused primarily on the host response. Here, we investigate the potential influence of the causative pathogen on sepsis heterogeneity and pathobiology. METHODS We performed a retrospective observational analysis of 8,280 critically ill patients with sepsis to identify associations between pathogen characteristics and the hyperinflammatory and hypoinflammatory patient phenotypes. We also performed controlled murine and swine modeling of sepsis and lung injury and a secondary analysis of 449 patients enrolled in the EUPHRATES randomized controlled trial. RESULTS Pathogen characteristics (pathogen identity, burden, virulence, and anatomic site of infection) were strongly and independently associated with the previously reported phenotypes. In a cohort of critically ill patients with sepsis, infection with gram-negative pathogens, primarily Enterobacterales spp. (e.g., Escherichia coli , Klebsiella pneumoniae ), was strongly associated with the hyperinflammatory phenotype. The hyperinflammatory phenotype was also independently associated with increased pathogen burden, virulence, and initial anatomic site of infection. In controlled murine and swine modeling, both the identity and burden of the pathogen provoked key biological features of the hyperinflammatory phenotype. Among patients with sepsis, the prognostic value of lactate clearance varied substantially by phenotype. In a secondary analysis of a randomized trial of polymyxin B hemoadsorption (which removes circulating endotoxin), hypoinflammatory patients experienced worse survival. CONCLUSIONS Our results demonstrate the central importance of pathogen features in the clinical and biological heterogeneity of sepsis. Future studies of sepsis pathobiology and heterogeneity should expand their scope beyond the host response, as understanding pathogen-host interactions will be crucial in the development of precision therapeutic strategies to improve patient outcomes. TRIAL REGISTRATION EUPHRATES trial NCT01046669. FUNDING 5P30AG024824, IK2CX002766, R01HL144599, K24HL159247, R01HL158626, R01HL173531, R35GM142992, R35GM145330, R35GM136312, K23HL166880, R35HL140026.

Viral lower respiratory tract disease (LRTD) is a major cause of global morbidity and mortality, but pathogen-specific estimates remain limited. We aimed to quantify the global burden of viral LRTD episodes, hospitalisations, and severe clinical outcomes by aetiology from 2010 to 2021. We used the Global Burden of Disease (GBD) 2021 modelling framework to generate estimates of viral LRTD incidence, hospitalisations, and clinical outcomes from 2010 to 2021 for 204 countries and territories, 21 regions, and seven super-regions. Aetiology-specific estimates were generated for five viral categories: influenza, respiratory syncytial virus (RSV), human metapneumovirus (hMPV), SARS-CoV-2, and “other” viral pathogens. Data inputs included surveillance systems, clinical informatics, published literature, surveys, and vital registration. Viral LRTD incidence was estimated using DisMod-MR 2.1, a GBD Bayesian meta-regression tool for disease incidence. We then applied location-specific admission scalars, adjusted for healthcare access, to derive hospitalisations. Aetiology-specific clinical outcome proportions—including intensive care unit (ICU) need, invasive mechanical ventilation (IMV) need, and in-hospital mortality—were estimated using meta-regression—Bayesian, regularised, trimming models. These were then applied to the hospitalisation estimates. COVID-19 incidence and hospitalisations were estimated using established GBD COVID-19 methods. All estimates are reported with 95

RATIONALE:Sepsis is a leading cause of mortality and involves a dysregulated host response to infection. The host and microbe have historically been considered independently in studies of sepsis, limiting our understanding of key relationships driving mortality. OBJECTIVES:We sought to identify host and microbial factors associated with sepsis mortality and build prognostic classifiers. METHODS:We studied 321 critically ill adults and adjudicated sepsis status. From whole blood collected within 24 hours of admission, we performed transcriptional profiling, and from plasma, we performed proteomic and metagenomic analyses. We evaluated associations between in-hospital mortality and gene expression, protein levels, and microbial metagenomic data and built support vector machine-based prognostic classifiers. MEASUREMENTS AND MAIN RESULTS:In patients with sepsis, mortality was associated with increased expression of genes related to neutrophil degranulation, lower expression of genes related to T-cell signaling, and higher IL-8 levels. Mortality was also associated with greater microbial mass and greater bacterial relative dominance. Similar findings were observed in a broader group that also included patients with culture-negative sepsis or indeterminate sepsis status. An integrated host-microbe metagenomic classifier predicted sepsis mortality with an area under the curve (AUC) of 0.79, and a host transcriptomic classifier performed comparably, with an AUC of 0.75. Both performed better (P < 0.05 by paired DeLong tests) that the Acute Physiology, Age, Chronic Health Evaluation III score (AUC of 0.69). CONCLUSIONS:Taken together, our findings provide a conceptual advance in the understanding of host and microbial factors associated with mortality in critical illness and demonstrate a new approach to mortality prediction in sepsis.

Suspected infection requiring hospitalisation has highly heterogenous presentation. Yet, variances in host response and its implications are largely unknown. In this multicentre cohort of 3802 individual patients presenting to the Emergency Department (ED) with suspected infection requiring hospitalisation, we apply uniform manifold approximation and projections and K-means clustering to 29 plasma proteins to identify biologically discrete host response clusters. In this work, we first describe two large clusters, called "Dysregulated" and "Undifferentiated", with abnormal protein concentrations and adverse outcomes in the former. Through further clustering, we identify 4 sub-clusters in the Dysregulated cluster, each with discrete biological signatures, clinical correlates, and outcomes. Clusters 3 and 4 are characterised by renal impairment and viral infections respectively. Clusters 5 and 6 are associated with bacterial culture positivity, with the former consistent with an immunosuppressed signature and worse outcomes, and the latter with gram-negative bacteria, higher IL-6 and IL-8, and better outcomes despite higher vasopressor use. These clusters are a biologically driven approach to characterising acute suspected infection and may lead to more targeted therapeutics.

Background Almost all large-scale trials of disease-modifying therapeutic agents in critical care have failed to show benefit for patients, which may be explained in part by the clinical and biological heterogeneity inherent in virtually all critical illness syndromes. Enrichment strategies have been developed to separate responders from non-responders and better target treatments. In patients with the acute respiratory distress syndrome, a critical illness syndrome involving severe lung inflammation, latent class analysis and other clustering approaches have led to the discovery of subgroups (phenotypes) that appear to respond differently to treatment based on retrospective analyses of published clinical trials and observational cohorts. The next step is to test these phenotypes in a prospective trial. Rapid, point-of-care analytical methods have now made such a trial possible. There is a need to advance treatment for patients with acute respiratory distress syndrome and other critical illness syndromes by incorporating a phenotype-based approach into prospective trial design. The hyperinflammatory and hypoinflammatory phenotypes, that have been identified in acute respiratory distress syndrome, will be the first to be included in such a trial, with scope for further phenotypes to be studied over time. Future work This Efficacy and Mechanism Evaluation report, through expert consensus, describes a new Phase II, multiarm, adaptive platform randomised controlled trial design that tests multiple pharmacological therapies in a population of patients with acute respiratory distress syndrome stratified by baseline inflammatory phenotype. This report also reviews issues to be considered in developing precision medicine trials in critical care, which are designed with newly developed clinical phenotypes in mind. This work has been used to develop the Precision medicine Adaptive Network platform Trial in Hypoxaemic acutE respiratory failuRe precision medicine trial in acute respiratory distress syndrome, which has been funded and will begin recruitment in June 2025. Limitations This report is the result of expert consensus review, rather than utilising strict review methodologies (e.g. Delphi consensus process). However, expert consensus has been found to generate similar results to consensus processes when a high degree of agreement is reached and > 70% agreement was reached for all included recommendations. Funding This article presents independent research funded by the (NIHR) Efficacy and Mechanism Evaluation programme as award number NIHR154493.

Rationale : The Hyperinflammatory phenotype of critical illness has been associated with higher 90-day mortality. However, whether baseline Hyperinflammatory classification is associated with deaths later in hospitalization remains unknown; and we hypothesized that this association weakens over the course of critical illness. Methods: We analyzed three cohorts of critically ill adult patients where molecular phenotypes were assigned previously using latent class analysis: the Reevaluation of Systemic Early Neuromuscular Blockade in ARDS (ROSE, Hyper=397, N=1006) trial, and two observational sepsis cohorts, Molecular Diagnosis and Risk Stratification of Sepsis (MARS; Hyper=590, N=1485) and Early Assessment of Renal and Lung Injury (EARLI; Hyper=288, N=818). Cox models with time-varying coefficients were used to estimate mortality hazard ratios over time of baseline Hyperinflammatory classification. We used change-point analyses on these daily estimates to identify a timepoint when the hazards ratio dropped-off markedly, and used the identified timepoint to dichotomize patient into either early- or late-death groups. We fitted logistic regression models to evaluate the association of mortality with baseline Hyperinflammatory phenotype or APACHE score independently in the two death groups. Results: Time-varying hazard ratios indicated a rapidly decreasing association of the Hyperinflammatory phenotype with death in all three cohorts [Figure A-C]. In each cohort, transition cutoffs for reduced mortality association of baseline Hyperinflammatory phenotype were as follows: Day 4 in ROSE, Day 9 in MARS, and Day 5 in EARLI, leading to 33%, 46%, and 37% patients in the early-death group in each cohort respectively. The association of the Hyperinflammatory phenotype with mortality was weaker in the late-death compared to early-death group (ROSE: OR 2.3, [CI:1.7-3.1] vs 8.5, [CI:5.5 -13.3]; MARS: OR 1.4, [CI:1.1-1.9] vs 3.6, [CI:2.7-4.9]; and EARLI: OR 1.9, [CI:1.3-3.0] vs 4.8, [CI:2.8-8.4]). Baseline Hyperinflammatory classification predicted early deaths better than late deaths (AUROC: ROSE 0.74 vs 0.59, MARS: 0.66 vs 0.54, EARLI: 0.69 vs 0.57). The predictive performance of APACHE score for mortality was comparable to the Hyperinflammatory phenotype for early but not late deaths [Figure D-F], with the APACHE models superior at predicting late deaths. Conclusions: Across three cohorts, we observed that the prognostic value of baseline Hyperinflammatory phenotype assignment diminishes rapidly over time, with a marked drop off around Day 5. Our findings suggest that Hyperinflammatory-specific characteristics are more likely to contribute to deaths that occur earlier in critical illness. Our results suggest that shorter-term mortality outcomes may be more pertinent for trials evaluating Hyperinflammatory-specific interventions in critically ill patients.

RATIONALE: Studies have shown differential response to PEEP in ARDS molecular phenotypes, with higher PEEP associated with survival in the Hyperinflammatory phenotype. We hypothesize the Hyperinflammatory phenotype has more systemically-induced lung injury which leads to greater lung recruitability than direct injury. Here, we evaluated whether vasopressor-use would be a simpler surrogate for systemically-induced lung injury and tested for heterogeneity of treatment effect (HTE) with PEEP in baseline vasopressor-use groups. METHODS: We analyzed 2,797 patients from LUNG-SAFE (observational) and 549 patients from ALVEOLI (RCT), which tested high-versus low-PEEP. In LUNG-SAFE, we created tertiles of PEEP averaged across days 1-3, discarded the middle tertile, and labeled the first/third tertiles as low-PEEP (median=5.0 [5.0-6.0]cmH2O) and high-PEEP (10 [10.0-12.0] cmH2O) groups. PEEP groups in ALVEOLI were labelled per randomization. Vasopressor-use classification was determined by baseline status (yes/no). To evaluate HTE, we regressed 90-day mortality against vasopressor-use groups, PEEP groups, and their interaction term. The LUNG-SAFE model was adjusted for baseline BMI, age, sex and PaO2 level. Since ALVEOLI was randomized, that model was left unadjusted. RESULTS: In LUNG-SAFE, vasopressor-use comprised 144 (51%) patients with higher 90-day mortality than non-vasopressor patients (46% vs 32%; p<0.001). 42% of patients in the low-PEEP and 63% of the high-PEEP groups were on vasopressors. We observed a significant treatment interaction (p<0.001), with high-PEEP associated with lower mortality in the vasopressor-use group and the opposite effect in the non-vasopressor group. In ALVEOLI, vasopressor-use comprised 144 (26%) patients and had higher mortality than the non-vasopressor group (41 vs 22%; p<0.001). Of these, 71 (49.6%) and 73 (50.7%) were in the high-PEEP and low-PEEP arms respectively. The treatment interaction was non-significant in ALVEOLI (p=0.067); however, numerically we observed the same patterns, with high-PEEP associated with lower mortality in the vasopressor-use and higher in the non-vasopressor-use group (Table). Nested logistic regression within the vasopressor-use group showed remarkably similar odds ratios for mortality in the high-PEEP arm in both LUNG-SAFE (OR=0.75, CI:0.51-1.07) and ALVEOLI (OR=0.72, CI:0.37-1.39). Based on these findings, with α=0.05 and β=0.8, a sample of 1,180 patients would be needed to detect an 8% mortality reduction in future ARDS trials of patients on vasopressors at enrollment. CONCLUSIONS: Vasopressor use at baseline predicted differential response to PEEP strategies, with a benefit from high PEEP in patients on vasopressors. These results suggest that vasopressor use may be a viable, univariate method for enriching patients who may benefit from high PEEP in future trials.

Clinicians aim to provide treatments that will result in the best outcome for each patient. Ideally, treatment decisions are based on evidence from randomised clinical trials. Randomised trials conventionally report an aggregated difference in outcomes between patients in each group, known as an average treatment effect. However, the actual effect of treatment on outcomes (treatment response) can vary considerably between individuals, and can differ substantially from the average treatment effect. This variation in response to treatment between patients-heterogeneity of treatment effect-is particularly important in critical care because common critical care syndromes (eg, sepsis and acute respiratory distress syndrome) are clinically and biologically heterogeneous. Statistical approaches have been developed to analyse heterogeneity of treatment effect and predict individualised treatment effects for each patient. In this Review, we outline a framework for deriving and validating individualised treatment effects and identify challenges to applying individualised treatment effect estimates to inform treatment decisions in clinical care.

INTRODUCTION: Hyper- and hypoinflammatory phenotypes have consistently been identified by latent class analysis in numerous cohorts of patients with ARDS and sepsis. In retrospective analysis of clinical trials of simvastatin in ARDS, corticosteroids in COVID-19 and activated protein C in severe sepsis, enrichment of treatment effect was observed in the hyperinflammatory phenotype. We hypothesized that the treatment effect of acetaminophen would be enriched in the hyperinflammatory phenotype among sepsis patients with respiratory or circulatory organ dysfunction enrolled in a recently completed trial of acetaminophen for sepsis (ASTER, Ware et al. JAMA 2024). METHODS: Patients enrolled in ASTER were retrospectively classified into the hyper- and hypoinflammatory phenotypes using a previously validated parsimonious classifier that included baseline plasma IL-8, sTNFR1, serum bicarbonate, and need for vasopressors. Multivariate logistic regression models were fit with key trial endpoints (days alive and free of respiratory, cardiovascular and renal organ support to 28-days, 28-day and 90-day mortality, and 28-day ventilator-free days) as the dependent variable and the interaction term of treatment group and phenotype as the independent variable. Treatment effect for acetaminophen was also evaluated separately for the hyper- and hypoinflammatory phenotypes. RESULTS: Of the 447 patients enrolled in ASTER, 392 had biomarkers and clinical data available for parsimonious classification: 108 (28%) patients were classified as hyperinflammatory and 284 (72%) patients as hypoinflammatory. Consistent with prior studies, mortality was higher in the hyperinflammatory phenotype at both 28-days (34% vs 17%, p < 0.001) and 90-days (43% vs. 26%, p = 0.002). There was a significant interaction (p=0.050) between acetaminophen treatment effect and inflammatory phenotype for days alive and free of organ support, but contrary to our hypothesis, the acetaminophen treatment effect was evident only in the hypoinflammatory group (Table). Although not statistically significant, the acetaminophen treatment effect was also larger in the hypoinflammatory phenotype across other the clinical outcomes tested (Table). CONCLUSIONS: In contrast to prior studies that have shown enrichment for treatment effect of simvastatin, corticosteroids and activated protein C in sepsis and ARDS patients with the hyperinflammatory phenotype, we did not identify an enrichment of treatment effect in the hyperinflammatory phenotype in the ASTER clinical trial. The treatment effect of acetaminophen appeared to be more beneficial in the hypoinflammatory phenotype, comprising the majority of patients enrolled. These findings indicate that future clinical trials of acetaminophen in sepsis should not prospectively enrich for the hyperinflammatory phenotype.

Rationale: Tight glycemic control (TGC) with insulin has not consistently shown benefit in critically ill patients. We previously reported that the subset of children with a hyperinflammatory subphenotype benefited from TGC in the HALF-PINT (Heart and Lung Failure - Pediatric Insulin Titration) study of hyperglycemic children with heart and lung failure and the IIT-SBPP (Intensive Insulin Treatment - Severely Burned Pediatric Patients) study in severely burned pediatric patients. However, whether this effect was mediated through a reduction in inflammation or some other biologic process is not fully understood. Objectives: To deepen the understanding of inflammatory subphenotypes and explore the biologic mechanisms underlying heterogeneous response to TGC. Methods: Plasma cytokine measurements and whole-blood transcriptomics from 740 blood samples collected on Pre- and Post-treatment Study Days 0, 2, and 4 from 293 HALF-PINT participants (n = 250 hypoinflammatory and n = 43 hyperinflammatory) were used to identify cytokine and gene expression signatures of differential responses to TGC. Measurements and Results: Patients with the hyperinflammatory subphenotype had greater baseline expression of genes relating to inflammation, cell-cycle activity, and immunometabolism. Hyperinflammatory patients treated to a target glucose range of 80-110 mg/dl experienced greater reductions in inflammatory cytokines, innate immune gene expression, and heme metabolism gene expression, as well as an increase in lymphocyte gene expression, compared with those treated to a target range of 150-180 mg/dl. Causal mediation testing indicated that these changes partly explained the observed mortality benefit of TGC in the hyperinflammatory subgroup of patients. Conclusions: These findings expand our understanding of the biology underlying inflammatory subphenotypes and provide biologic insight into the mortality benefit of TGC in hyperinflammatory children.