Machine learning-driven identification of the gene-expression signature associated with a persistent multiple organ dysfunction trajectory in critical illness.

BACKGROUND: Multiple organ dysfunction syndrome (MODS) disproportionately drives morbidity and mortality among critically ill patients. However, we lack a comprehensive understanding of its pathobiology. Identification of genes associated with a persistent MODS trajectory may shed light on underlying biology and allow for accurate prediction of those at-risk. METHODS: Secondary analyses of publicly available gene-expression datasets. Supervised machine learning (ML) was used to identify a parsimonious set of genes associated with a persistent MODS trajectory in a training set of pediatric septic shock. We optimized model parameters and tested risk-prediction capabilities in independent validation and test datasets, respectively. We compared model performance relative to an established gene-set predictive of sepsis mortality. FINDINGS: Patients with a persistent MODS trajectory had 568 differentially expressed genes and characterized by a dysregulated innate immune response. Supervised ML identified 111 genes associated with the outcome of interest on repeated cross-validation, with an AUROC of 0.87 (95% CI: 0.85-0.88) in the training set. The optimized model, limited to 20 genes, achieved AUROCs ranging from 0.74 to 0.79 in the validation and test sets to predict those with persistent MODS, regardless of host age and cause of organ dysfunction. Our classifier demonstrated reproducibility in identifying those with persistent MODS in comparison with a published gene-set predictive of sepsis mortality. INTERPRETATION: We demonstrate the utility of supervised ML driven identification of the genes associated with persistent MODS. Pending validation in enriched cohorts with a high burden of organ dysfunction, such an approach may inform targeted delivery of interventions among at-risk patients. FUNDING: H.R.W.'s NIHR35GM126943 award supported the work detailed in this manuscript. Upon his death, the award was transferred to M.N.A. M.R.A., N.S.P, and R.K were supported by NIHR21GM151703. R.K. was supported by R01GM139967.

Biomarker-based risk model to predict persistent multiple organ dysfunctions after congenital heart surgery: a prospective observational cohort study.

BACKGROUND: Multiple organ dysfunction syndrome (MODS) is an important cause of post-operative morbidity and mortality for children undergoing cardiac surgery requiring cardiopulmonary bypass (CPB). Dysregulated inflammation is widely regarded as a key contributor to bypass-related MODS pathobiology, with considerable overlap of pathways associated with septic shock. The pediatric sepsis biomarker risk model (PERSEVERE) is comprised of seven protein biomarkers of inflammation and reliably predicts baseline risk of mortality and organ dysfunction among critically ill children with septic shock. We aimed to determine if PERSEVERE biomarkers and clinical data could be combined to derive a new model to assess the risk of persistent CPB-related MODS in the early post-operative period. METHODS: This study included 306 patients < 18 years old admitted to a pediatric cardiac ICU after surgery requiring cardiopulmonary bypass (CPB) for congenital heart disease. Persistent MODS, defined as dysfunction of two or more organ systems on postoperative day 5, was the primary outcome. PERSEVERE biomarkers were collected 4 and 12 h after CPB. Classification and regression tree methodology were used to derive a model to assess the risk of persistent MODS. RESULTS: The optimal model containing interleukin-8 (IL-8), chemokine ligand 3 (CCL3), and age as predictor variables had an area under the receiver operating characteristic curve (AUROC) of 0.86 (0.81-0.91) for differentiating those with or without persistent MODS and a negative predictive value of 99% (95-100). Ten-fold cross-validation of the model yielded a corrected AUROC of 0.75 (0.68-0.84). CONCLUSIONS: We present a novel risk prediction model to assess the risk for development of multiple organ dysfunction after pediatric cardiac surgery requiring CPB. Pending prospective validation, our model may facilitate identification of a high-risk cohort to direct interventions and studies aimed at improving outcomes via mitigation of post-operative organ dysfunction.

Identification of a pediatric acute hypoxemic respiratory failure signature in peripheral blood leukocytes at 24 hours post-ICU admission with machine learning.

Background: There is no generalizable transcriptomics signature of pediatric acute respiratory distress syndrome. Our goal was to identify a whole blood differential gene expression signature for pediatric acute hypoxemic respiratory failure (AHRF) using transcriptomic microarrays within twenty-four hours of diagnosis. We used publicly available human whole-blood gene expression arrays of a Berlin-defined pediatric acute respiratory distress syndrome (GSE147902) cohort and a sepsis-triggered AHRF (GSE66099) cohort within twenty-four hours of diagnosis and compared those children with a PaO2/FiO2 < 200 to those with a PaO2/FiO2 ≥ 200. Results: We used stability selection, a bootstrapping method of 100 simulations using logistic regression as a classifier, to select differentially expressed genes associated with a PaO2/FiO2 < 200 vs. PaO2/FiO2 ≥ 200. The top-ranked genes that contributed to the AHRF signature were selected in each dataset. Genes common to both of the top 1,500 ranked gene lists were selected for pathway analysis. Pathway and network analysis was performed using the Pathway Network Analysis Visualizer (PANEV) and Reactome was used to perform an over-representation gene network analysis of the top-ranked genes common to both cohorts. Changes in metabolic pathways involved in energy balance, fundamental cellular processes such as protein translation, mitochondrial function, oxidative stress, immune signaling, and inflammation are differentially regulated early in pediatric ARDS and sepsis-induced AHRF compared to both healthy controls and to milder acute hypoxemia. Specifically, fundamental pathways related to the severity of hypoxemia emerged and included (1) ribosomal and eukaryotic initiation of factor 2 (eIF2) regulation of protein translation and (2) the nutrient, oxygen, and energy sensing pathway, mTOR, activated via PI3K/AKT signaling. Conclusions: Cellular energetics and metabolic pathways are important mechanisms to consider to further our understanding of the heterogeneity and underlying pathobiology of moderate and severe pediatric acute respiratory distress syndrome. Our findings are hypothesis generating and support the study of metabolic pathways and cellular energetics to understand heterogeneity and underlying pathobiology of moderate and severe acute hypoxemic respiratory failure in children.

Biomarker-based risk model to predict persistent multiple organ dysfunctions after congenital heart surgery â€" A prospective observational cohort study.

Background: Multiple organ dysfunction syndrome (MODS) is an important cause of post-operative morbidity and mortality for children undergoing cardiac surgery requiring cardiopulmonary bypass (CPB). Dysregulated inflammation is widely regarded as a key contributor to bypass-related MODS pathobiology, with considerable overlap of pathways associated with septic shock. The pediatric sepsis biomarker risk model (PERSEVERE) is comprised of seven protein biomarkers of inflammation, and reliably predicts baseline risk of mortality and organ dysfunction among critically ill children with septic shock. We aimed to determine if PERSEVERE biomarkers and clinical data could be combined to derive a new model to assess the risk of persistent CPB-related MODS in the early post-operative period. Methods: This study included 306 patients <18 years old admitted to a pediatric cardiac ICU after surgery requiring cardiopulmonary bypass (CPB) for congenital heart disease. Persistent MODS, defined as dysfunction of two or more organ systems on postoperative day 5, was the primary outcome. PERSEVERE biomarkers were collected 4 and 12 hours after CPB. Classification and Regression Tree methodology was used to derive a model to assess the risk of persistent MODS. Results: The optimal model containing interleukin-8 (IL-8), chemokine ligand 3 (CCL3), and age as predictor variables, had an area under the receiver operating characteristic curve (AUROC) of 0.86 (0.81-0.91) for differentiating those with or without persistent MODS, and a negative predictive value of 99% (95-100). Ten-fold cross-validation of the model yielded a corrected AUROC of 0.75. Conclusions: We present a novel risk prediction model to assess the risk for development of multiple organ dysfunction after pediatric cardiac surgery requiring CPB. Pending prospective validation, our model may facilitate identification of a high-risk cohort to direct interventions and studies aimed at improving outcomes via mitigation of post-operative organ dysfunction. Clinical Trial Registration Number: This study does not meet criteria for a clinical trial per the WHO International Clinical Trials Registry Platform as no intervention was performed.

Biomarkers estimating baseline mortality risk for neonatal sepsis: nPERSEVERE: neonate-specific sepsis biomarker risk model.

BACKGROUND: Prognostic biomarker research neonatal sepsis is lacking. We assessed the utility of a validated pediatric prognostic tool called PERSEVERE II that uses decision tree methodology to predict mortality at discharge in neonates who experienced sepsis. METHODS: Prospective study in a dual-center cohort of neonates with sepsis admitted between June 2020 and December 2021. Biomarker analysis was done on serum samples obtained at the time of evaluation for the event. RESULTS: In a cohort of 59 neonates with a mortality rate of 15.3%, PERSEVERE II was 67% sensitive and 59% specific for mortality, p 0.27. Amongst PERSEVERE II biomarkers, IL-8 showed good prognostic performance for mortality prediction with a cutoff of 300 pg/mL (sensitivity 100%, specificity 65%, negative predictive value 100%, AUC 0.87, p 0.0003). We derived a new decision tree that is neonate specific (nPERSEVERE) with improved performance compared to IL-8 (sensitivity 100%, specificity 86%, negative predictive value 100%, AUC 0.95, p < 0.0001). CONCLUSIONS: IL-8 and nPERSEVERE demonstrated good prognostic performance in a small cohort of neonates with sepsis. Moving toward precision medicine in sepsis, our study proposes an important tool for clinical trial prognostic enrichment that needs to be validated in larger studies. IMPACT: Prognostic and predictive biomarker research is lacking in the newborn intensive care unit. Biomarkers can be used at the time of evaluation for neonatal sepsis (blood culture acquisition) to identify neonates with high baseline mortality risk. Stratification is an important step toward precision medicine in neonatal sepsis.

Electrolyte derangements in critically ill children receiving balanced versus unbalanced crystalloid fluid resuscitation.

BACKGROUND: Adult studies have demonstrated potential harm from resuscitation with 0.9% sodium chloride (0.9%NaCl), resulting in increased utilization of balanced crystalloids like lactated ringers (LR). The sodium and potassium content of LR has resulted in theoretical safety concerns, although limited data exists in pediatrics. We hypothesized that use of LR for resuscitation would not be associated with increased electrolyte derangements compared to 0.9%NaCl. METHODS: A prospective, observational cohort study of critically ill children who received ≥ 20 ml/kg of fluid resuscitation and were admitted to two pediatric intensive care units from November 2017 to February 2020. Fluid groups included patients who received > 75% of fluids from 0.9%NaCl, > 75% of fluids from LR, and a mixed group. The primary outcome was incidence of electrolyte derangements (sodium, chloride, potassium) and acidosis. RESULTS: Among 559 patients, 297 (53%) received predominantly 0.9%NaCl, 74 (13%) received predominantly LR, and 188 (34%) received a mixture. Extreme hyperkalemia (potassium ≥ 6 mmol/L) was more common in 0.9%NaCl group (5.8%) compared to LR group (0%), p 0.05. Extreme acidosis (pH > 7.1) was more common in 0.9%NaCl group (11%) compared to LR group (1.6%), p 0.016. CONCLUSIONS: LR is associated with fewer electrolyte derangements compared to 0.9%NaCl. Prospective interventional trials are needed to validate these findings.

Multi-omic characterization of pediatric ARDS via nasal brushings.

RATIONALE: While nasal brushing transcriptomics can identify disease subtypes in chronic pulmonary diseases, it is unknown whether this is true in pediatric acute respiratory distress syndrome (PARDS). OBJECTIVES: Determine whether nasal transcriptomics and methylomics can identify clinically meaningful PARDS subgroups that reflect important pathobiological processes. METHODS: Nasal brushings and serum were collected on days 1, 3, 7, and 14 from control and PARDS subjects from two centers. PARDS duration was the primary endpoint. MEASUREMENTS AND MAIN RESULTS: Twenty-four control and 39 PARDS subjects were enrolled. Two nasal methylation patterns were identified. Compared to Methyl Subgroup 1, Subgroup 2 had hypomethylation of inflammatory genes and was enriched for immunocompromised subjects. Four transcriptomic patterns were identified with temporal patterns indicating injury, repair, and regeneration. Over time, both inflammatory (Subgroup B) and cell injury (Subgroup D) patterns transitioned to repair (Subgroup A) and eventually homeostasis (Subgroup C). When control specimens were included, they were largely Subgroup C. In comparison with 17 serum biomarkers, the nasal transcriptome was more predictive of prolonged PARDS. Subjects with initial Transcriptomic Subgroup B or D assignment had median PARDS duration of 8 days compared to 2 in A or C (p = 0.02). For predicting PARDS duration ≥ 3 days, nasal transcriptomics was more sensitive and serum biomarkers more specific. CONCLUSIONS: PARDS nasal transcriptome may reflect distal lung injury, repair, and regeneration. A combined nasal PCR and serum biomarker assay could be useful for predictive and diagnostic enrichment. Trial registration Clinicaltrials.gov NCT03539783 May 29, 2018.

Integrated PERSEVERE and endothelial biomarker risk model predicts death and persistent MODS in pediatric septic shock: a secondary analysis of a prospective observational study.

BACKGROUND: Multiple organ dysfunction syndrome (MODS) is a critical driver of sepsis morbidity and mortality in children. Early identification of those at risk of death and persistent organ dysfunctions is necessary to enrich patients for future trials of sepsis therapeutics. Here, we sought to integrate endothelial and PERSEVERE biomarkers to estimate the composite risk of death or organ dysfunctions on day 7 of septic shock. METHODS: We measured endothelial dysfunction markers from day 1 serum among those with existing PERSEVERE data. TreeNet® classification model was derived incorporating 22 clinical and biological variables to estimate risk. Based on relative variable importance, a simplified 6-biomarker model was developed thereafter. RESULTS: Among 502 patients, 49 patients died before day 7 and 124 patients had persistence of MODS on day 7 of septic shock. Area under the receiver operator characteristic curve (AUROC) for the newly derived PERSEVEREnce model to predict death or day 7 MODS was 0.93 (0.91-0.95) with a summary AUROC of 0.80 (0.76-0.84) upon tenfold cross-validation. The simplified model, based on IL-8, HSP70, ICAM-1, Angpt2/Tie2, Angpt2/Angpt1, and Thrombomodulin, performed similarly. Interaction between variables-ICAM-1 with IL-8 and Thrombomodulin with Angpt2/Angpt1-contributed to the models' predictive capabilities. Model performance varied when estimating risk of individual organ dysfunctions with AUROCS ranging from 0.91 to 0.97 and 0.68 to 0.89 in training and test sets, respectively. CONCLUSIONS: The newly derived PERSEVEREnce biomarker model reliably estimates risk of death or persistent organ dysfunctions on day 7 of septic shock. If validated, this tool can be used for prognostic enrichment in future pediatric trials of sepsis therapeutics.

Redefining critical illness.

Research and practice in critical care medicine have long been defined by syndromes, which, despite being clinically recognizable entities, are, in fact, loose amalgams of heterogeneous states that may respond differently to therapy. Mounting translational evidence-supported by research on respiratory failure due to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection-suggests that the current syndrome-based framework of critical illness should be reconsidered. Here we discuss recent findings from basic science and clinical research in critical care and explore how these might inform a new conceptual model of critical illness. De-emphasizing syndromes, we focus on the underlying biological changes that underpin critical illness states and that may be amenable to treatment. We hypothesize that such an approach will accelerate critical care research, leading to a richer understanding of the pathobiology of critical illness and of the key determinants of patient outcomes. This, in turn, will support the design of more effective clinical trials and inform a more precise and more effective practice at the bedside.

Candidate Biomarkers for Sepsis-Associated Acute Kidney Injury Mechanistic Studies.

INTRODUCTION: Sepsis-associated acute kidney injury (SA-AKI) is a frequent complication of sepsis, yet the pathophysiologic mechanisms of SA-AKI are incompletely understood. PERSEVERE is a clinically validated serum biomarker panel with high sensitivity in predicting mortality from sepsis, and recent evidence suggests it can also predict severe, persistent SA-AKI at day 3 of hospitalization among septic children. We developed a murine model of PERSEVERE (mPERSEVERE) to further interrogate the sepsis-related biological underpinnings of SA-AKI using candidate biomarkers within mPERSEVERE. METHODS: Eight-week-old C57BL/6 male mice underwent induction of sepsis by cecal ligation and puncture (CLP). mPERSEVERE biomarkers were collected at 8-hours and kidneys were harvested at 24-hours post-CLP Classification and regression tree analysis (CART) was used to generate a SA-AKI predictive model. Kidney gene expression levels of candidate biomarkers were quantified using real time polymerase chain reaction. RESULTS: Thirty- five mice underwent CLP Among mice identified by mPERSEVERE as high-risk for mortality, 70% developed SA-AKI at 24-hours compared to 22% of low-risk mice. CART analysis identified two mPERSEVERE biomarkers-C-C motif chemokine ligand 3 (CCL3) and keratinocyte-derived chemokine (KC)-as most predictive for SA-AKI with an area under the receiver operating curve of 0.90. In mice that developed SA-AKI, renal expression of KC was significantly increased compared to mice without SA-AKI (p  = 0.013), whereas no difference was seen in renal expression of CCL3 in mice with SA-AKI vs. no SA-AKI. KC and CCL3 localized to renal tubule epithelial cells as opposed to infiltrating immune cells by immunohistochemistry. CONCLUSIONS: The combination of plasma CCL3+KC can predict SA-AKI development in mice at 24-hours following CLP Of these two biomarkers, only renal expression of KC is increased in mice with SA-AKI. Further studies are required to determine if KC directly contributes to the underlying pathobiology of SA-AKI.

Matrix metalloproteinases and their inhibitors in pediatric severe acute pancreatitis.

BACKGROUND: Acute pancreatitis (AP) is increasing in incidence in adult and pediatric patients. Identification of patients at high risk for progression to severe acute pancreatitis (SAP) is crucial, as it can lead to increased mortality and health system cost. Matrix metalloproteinases (MMPs) are endopeptidases which degrade extracellular matrix proteins and increase activity of pro-inflammatory cytokines. Tissue inhibitors of metalloproteinases (TIMPs) regulate MMP activity. Prior limited studies of MMPs and TIMPs have found some to be associated with development of SAP. The aim of this study was to further investigate the role of MMPs and TIMPs in detecting pediatric patients at risk for developing moderately severe AP or SAP. METHODS: Plasma samples were prospectively collected for patients <21 years of age presenting with AP between November 2015 and October 2019, along with healthy controls. Bead-based multiplex assays were utilized to test levels of 12 MMPs and TIMPs. RESULTS: Samples were collected from 7 subjects who developed SAP, 7 with moderately severe AP, 45 with mild AP and 44 healthy controls. MMP-9 (p = 0.04) and TIMP-1 (p = 0.01) levels were significantly higher in SAP patients. A multivariable logistic regression model using MMP-9 and TIMP-1 predicted SAP (AUROC 0.87, 95% CI 0.76-0.98). CONCLUSION: We have demonstrated that MMP9 and TIMP1 levels are increased at AP presentation in pediatric patients who developed SAP during the course of illness. Further studies are needed to validate the use of MMPs and TIMPs as predictive tools for development of SAP in pediatric pancreatitis.

Biomarkers associated with mortality in pediatric patients with cardiac arrest and acute respiratory distress syndrome.

AIMS: To identify plasma biomarkers associated with cardiac arrest in a cohort of children with acute respiratory distress syndrome (ARDS), and to assess the association of these biomarkers with mortality in children with cardiac arrest and ARDS (ARDS + CA). METHODS: This was a secondary analysis of a single-center prospective cohort study of children with ARDS from 2014-2019 with 17 biomarkers measured. Clinical characteristics and biomarkers were compared between subjects with ARDS + CA and ARDS with univariate analysis. In a sub-cohort of ARDS + CA subjects, the association between biomarker levels and mortality was tested using univariate and bivariate logistic regression. RESULTS: Biomarkers were measured in 333 subjects: 301 with ARDS (median age 5.3 years, 55.5% male) and 32 ARDS + CA (median age 8 years, 53.1% male). More arrests (69%) occurred out-of-hospital with a median CPR duration of 11 (IQR 5.5, 25) minutes. ARDS severity, PRISM III score, vasoactive-ionotropic score and extrapulmonary organ failures were worse in the ARDS + CA versus ARDS group. Eight biomarkers were elevated in the ARDS + CA versus ARDS cohort: sRAGE, nucleosomes, SP-D, CCL22, IL-6, HSP70, IL-8, and MIP-1b. sRAGE, SP-D, and CCL22 remained elevated when the cohorts were matched for illness severity. When controlling for severity of ARDS and cardiac arrest characteristics, sRAGE, IL-6 and granzyme B were associated with mortality in the ARDS + CA group. CONCLUSION: sRAGE, IL-6 and granzyme B were associated with cardiac arrest mortality when controlling for illness severity. sRAGE was consistently higher in the ARDS + CA cohort compared to ARDS and retained independent association with mortality.

Pediatric Sepsis Biomarker Risk Model Biomarkers and Estimation of Myocardial Dysfunction in Pediatric Septic Shock.

OBJECTIVES: Sepsis-associated myocardial dysfunction is common in pediatric septic shock and negatively impacts outcomes. Early estimation of sepsis-associated myocardial dysfunction risk has the potential to inform clinical care and improve clinical trial design. The Pediatric Sepsis Biomarker Risk Model II is validated as a biomarker-based enrichment algorithm to discriminate children with septic shock with high baseline mortality probability. The objectives were to determine if Pediatric Sepsis Biomarker Risk Model biomarkers are associated with risk for sepsis-associated myocardial dysfunction in pediatric septic shock and to develop a biomarker-based model to reliably estimate sepsis-associated myocardial dysfunction risk. DESIGN: Secondary analysis of prospective cohort study. SETTING: Single-center, quaternary-care PICU. PATIENTS: Children less than 18 years old admitted to the PICU from 2003 to 2018 who had Pediatric Sepsis Biomarker Risk Model biomarkers measured for determination of Pediatric Sepsis Biomarker Risk Model II mortality probability and an echocardiogram performed within 48 hours of septic shock identification. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Pediatric Sepsis Biomarker Risk Model II mortality probability was calculated from serum biomarker concentrations and admission platelet count. Echocardiograms were reread by a single cardiologist blinded to Pediatric Sepsis Biomarker Risk Model II data, and sepsis-associated myocardial dysfunction was defined as left ventricular ejection fraction less than 45% for primary analyses. Multivariable logistic regression analyzed the association of Pediatric Sepsis Biomarker Risk Model II mortality probability with sepsis-associated myocardial dysfunction. Classification and regression tree methodology was employed to derive a Pediatric Sepsis Biomarker Risk Model biomarker-based model for sepsis-associated myocardial dysfunction. Thirty-two of 181 children with septic shock demonstrated sepsis-associated myocardial dysfunction. Pediatric Sepsis Biomarker Risk Model II mortality probability was independently associated with sepsis-associated myocardial dysfunction (odds ratio, 1.45; 95% CI, 1.17-1.81; p = 0.001). Modeling with Pediatric Sepsis Biomarker Risk Model biomarkers estimated sepsis-associated myocardial dysfunction risk with an area under the receiver operating characteristic curve of 0.90 (95% CI, 0.85-0.95). Upon 10-fold cross-validation, the derived model had a summary area under the receiver operating characteristic curve of 0.74. Model characteristics were similar when sepsis-associated myocardial dysfunction was defined by both low left ventricular ejection fraction and abnormal global longitudinal strain. CONCLUSIONS: A newly derived Pediatric Sepsis Biomarker Risk Model biomarker-based model reliably estimates risk of sepsis-associated myocardial dysfunction in pediatric septic shock, but independent prospective validation is needed.

Advancing precision medicine for acute respiratory distress syndrome.

Acute respiratory distress syndrome (ARDS) is a heterogeneous clinical syndrome. Understanding of the complex pathways involved in lung injury pathogenesis, resolution, and repair has grown considerably in recent decades. Nevertheless, to date, only therapies targeting ventilation-induced lung injury have consistently proven beneficial, and despite these gains, ARDS morbidity and mortality remain high. Many candidate therapies with promise in preclinical studies have been ineffective in human trials, probably at least in part due to clinical and biological heterogeneity that modifies treatment responsiveness in human ARDS. A precision medicine approach to ARDS seeks to better account for this heterogeneity by matching therapies to subgroups of patients that are anticipated to be most likely to benefit, which initially might be identified in part by assessing for heterogeneity of treatment effect in clinical trials. In October 2019, the US National Heart, Lung, and Blood Institute convened a workshop of multidisciplinary experts to explore research opportunities and challenges for accelerating precision medicine in ARDS. Topics of discussion included the rationale and challenges for a precision medicine approach in ARDS, the roles of preclinical ARDS models in precision medicine, essential features of cohort studies to advance precision medicine, and novel approaches to clinical trials to support development and validation of a precision medicine strategy. In this Position Paper, we summarise workshop discussions, recommendations, and unresolved questions for advancing precision medicine in ARDS. Although the workshop took place before the COVID-19 pandemic began, the pandemic has highlighted the urgent need for precision therapies for ARDS as the global scientific community grapples with many of the key concepts, innovations, and challenges discussed at this workshop.

Pediatric sepsis biomarkers for prognostic and predictive enrichment.

Sepsis is a major public health problem in children throughout the world. Given that the treatment guidelines emphasize early recognition, there is interest in developing biomarkers of sepsis, and most attention is focused on diagnostic biomarkers. While there is a need for ongoing discovery and development of diagnostic biomarkers for sepsis, this review will focus on less well-known applications of sepsis biomarkers. Among patients with sepsis, the biomarkers can give information regarding the risk of poor outcome from sepsis, risk of sepsis-related organ dysfunction, and subgroups of patients with sepsis who share underlying biological features potentially amenable to targeted therapeutics. These types of biomarkers, beyond the traditional concept of diagnosis, address the important concepts of prognostic and predictive enrichment, which are key components of bringing the promise of precision medicine to the bedside of children with sepsis.

A Research Agenda for Precision Medicine in Sepsis and Acute Respiratory Distress Syndrome: An Official American Thoracic Society Research Statement.

Background: Precision medicine focuses on the identification of therapeutic strategies that are effective for a group of patients based on similar unifying characteristics. The recent success of precision medicine in non-critical care settings has resulted from the confluence of large clinical and biospecimen repositories, innovative bioinformatics, and novel trial designs. Similar advances for precision medicine in sepsis and in the acute respiratory distress syndrome (ARDS) are possible but will require further investigation and significant investment in infrastructure. Methods: This project was funded by the American Thoracic Society Board of Directors. A multidisciplinary and diverse working group reviewed the available literature, established a conceptual framework, and iteratively developed recommendations for the Precision Medicine Research Agenda for Sepsis and ARDS. Results: The following six priority recommendations were developed by the working group: 1) the creation of large richly phenotyped and harmonized knowledge networks of clinical, imaging, and multianalyte molecular data for sepsis and ARDS; 2) the implementation of novel trial designs, including adaptive designs, and embedding trial procedures in the electronic health record; 3) continued innovation in the data science and engineering methods required to identify heterogeneity of treatment effect; 4) further development of the tools necessary for the real-time application of precision medicine approaches; 5) work to ensure that precision medicine strategies are applicable and available to a broad range of patients varying across differing racial, ethnic, socioeconomic, and demographic groups; and 6) the securement and maintenance of adequate and sustainable funding for precision medicine efforts. Conclusions: Precision medicine approaches that incorporate variability in genomic, biologic, and environmental factors may provide a path forward for better individualizing the delivery of therapies and improving care for patients with sepsis and ARDS.

Recalibration of the Renal Angina Index for Pediatric Septic Shock.

INTRODUCTION: Sepsis-associated acute kidney injury (AKI) is a common diagnosis in children that is associated with poor outcomes. The lack of therapeutic options once present makes early identification of at-risk patients essential. The renal angina index (RAI) has been previously validated to predict severe AKI in heterogeneous populations of critically ill children. The performance of this score specifically in children with septic shock is unknown. METHODS: A secondary analysis of a multicenter, prospective, observational study of 379 children with septic shock to determine the ability of the RAI to predict severe AKI at day 3, and to assess for the potential need for recalibration of the RAI in this unique subset of patients. RESULTS: At the original cutoff of ≥8, the RAI predicted day 3 severe AKI with an area under the receiving operating characteristic (AUROC) curve 0.90 (95% confidence interval [CI]: 0.86 to 93), 95% sensitivity, and 54% specificity. A Youden's index identified a higher optimal cutoff of ≥20 (sensitivity 83%, specificity 80%), and day 1 platelet count <150 × 103/µl was an independent predictor of severe AKI (adjusted odds ratio: 3.2; 95% CI: 1.7 to 6.3; P < 0.001). Recalibration of the RAI to include platelet count and this new threshold restored the sensitivity of the original ≥8 threshold (95%), while improving its specificity (69%). CONCLUSIONS: The RAI appears to be a sensitive and reliable tool for prediction of severe AKI in children with septic shock, although the use of a recalibrated sepsis-specific RAI using a higher cutoff and platelet count may be beneficial.

A Precision Medicine Approach to Biomarker Utilization in Pediatric Sepsis-Associated Acute Kidney Injury.

Sepsis is a leading cause of morbidity and mortality in critically ill children, and acute kidney injury (AKI) is a frequent complication that confers an increased risk for poor outcomes. Despite the documented consequences of sepsis-associated AKI (SA-AKI), no effective disease-modifying therapies have been identified to date. As such, the only treatment options for these patients remain prevention and supportive care, both of which rely on the ability to promptly and accurately identify at risk and affected individuals. To achieve these goals, a variety of biomarkers have been investigated to help augment our currently limited predictive and diagnostic strategies for SA-AKI, however, these have had variable success in pediatric sepsis. In this mini-review, we will briefly outline the current use of biomarkers for SA-AKI, and propose a new framework for biomarker discovery and utilization that considers the individual patient's sepsis inflammatory response. Now recognized to be a key driver in the complex pathophysiology of SA-AKI, understanding the dysregulated host immune response to sepsis is a growing area of research that can and should be leveraged to improve the prediction and diagnosis of SA-AKI, while also potentially identifying novel therapeutic targets. Reframing SA-AKI in this manner - as a direct consequence of the individual patient's sepsis inflammatory response - will facilitate a precision medicine approach to its management, something that is required to move the care of this consequential disorder forward.

Machine Learning Identifies Complicated Sepsis Course and Subsequent Mortality Based on 20 Genes in Peripheral Blood Immune Cells at 24 H Post-ICU Admission.

A complicated clinical course for critically ill patients admitted to the intensive care unit (ICU) usually includes multiorgan dysfunction and subsequent death. Owing to the heterogeneity, complexity, and unpredictability of the disease progression, ICU patient care is challenging. Identifying the predictors of complicated courses and subsequent mortality at the early stages of the disease and recognizing the trajectory of the disease from the vast array of longitudinal quantitative clinical data is difficult. Therefore, we attempted to perform a meta-analysis of previously published gene expression datasets to identify novel early biomarkers and train the artificial intelligence systems to recognize the disease trajectories and subsequent clinical outcomes. Using the gene expression profile of peripheral blood cells obtained within 24 h of pediatric ICU (PICU) admission and numerous clinical data from 228 septic patients from pediatric ICU, we identified 20 differentially expressed genes predictive of complicated course outcomes and developed a new machine learning model. After 5-fold cross-validation with 10 iterations, the overall mean area under the curve reached 0.82. Using a subset of the same set of genes, we further achieved an overall area under the curve of 0.72, 0.96, 0.83, and 0.82, respectively, on four independent external validation sets. This model was highly effective in identifying the clinical trajectories of the patients and mortality. Artificial intelligence systems identified eight out of twenty novel genetic markers (SDC4, CLEC5A, TCN1, MS4A3, HCAR3, OLAH, PLCB1, and NLRP1) that help predict sepsis severity or mortality. While these genes have been previously associated with sepsis mortality, in this work, we show that these genes are also implicated in complex disease courses, even among survivors. The discovery of eight novel genetic biomarkers related to the overactive innate immune system, including neutrophil function, and a new predictive machine learning method provides options to effectively recognize sepsis trajectories, modify real-time treatment options, improve prognosis, and patient survival.