Acute respiratory distress syndrome (ARDS) is a common cause of respiratory failure yet has few pharmacologic therapies, reflecting the mechanistic heterogeneity of lung injury. We hypothesized that damage to the alveolar epithelial glycocalyx, a layer of glycosaminoglycans interposed between the epithelium and surfactant, contributes to lung injury in patients with ARDS. Using mass spectrometry of airspace fluid noninvasively collected from mechanically ventilated patients, we found that airspace glycosaminoglycan shedding (an index of glycocalyx degradation) occurred predominantly in patients with direct lung injury and was associated with duration of mechanical ventilation. Male patients had increased shedding, which correlated with airspace concentrations of matrix metalloproteinases. Selective epithelial glycocalyx degradation in mice was sufficient to induce surfactant dysfunction, a key characteristic of ARDS, leading to microatelectasis and decreased lung compliance. Rapid colorimetric quantification of airspace glycosaminoglycans was feasible and could provide point-of-care prognostic information to clinicians and/or be used for predictive enrichment in clinical trials.

Rationale: Acute respiratory distress syndrome (ARDS) lacks known causal biomarkers. Plasma concentrations of sRAGE (soluble receptor for advanced glycation end products) strongly associate with ARDS risk. However, whether plasma sRAGE contributes causally to ARDS remains unknown. Objectives: Evaluate plasma sRAGE as a causal intermediate in ARDS by Mendelian randomization (MR), a statistical method to infer causality using observational data. Methods: We measured early plasma sRAGE in two critically ill populations with sepsis. The cohorts were whole-genome genotyped and phenotyped for ARDS. To select validated genetic instruments for MR, we regressed plasma sRAGE on genome-wide genotypes in both cohorts. The causal effect of plasma sRAGE on ARDS was inferred using the top variants with significant associations in both populations (P < 0.01, R² > 0.02). We applied the inverse variance–weighted method to obtain consistent estimates of the causal effect of plasma sRAGE on ARDS risk. Measurements and Main Results: There were 393 European and 266 African ancestry patients in the first cohort and 843 European ancestry patients in the second cohort. Plasma sRAGE was strongly associated with ARDS risk in both populations (odds ratio, 1.86; 95% confidence interval [1.54–2.25]; 2.56 [2.14–3.06] per log increase). Using genetic instruments common to both populations, plasma sRAGE had a consistent causal effect on ARDS risk with a β estimate of 0.50 (95% confidence interval [0.09–0.91] per log increase). Conclusions: Plasma sRAGE is genetically regulated during sepsis, and MR analysis indicates that increased plasma sRAGE leads to increased ARDS risk, suggesting plasma sRAGE acts as a causal intermediate in sepsis-related ARDS.

Discovering novel uses for existing drugs, through drug repurposing, can reduce the time, costs, and risk of failure associated with new drug development. However, prioritizing drug repurposing candidates for downstream studies remains challenging. Here, we present a high-throughput approach to identify and validate drug repurposing candidates. This approach integrates human gene expression, drug perturbation, and clinical data from publicly available resources. We apply this approach to find drug repurposing candidates for two diseases, hyperlipidemia and hypertension. We screen >21,000 compounds and replicate ten approved drugs. We also identify 25 (seven for hyperlipidemia, eighteen for hypertension) drugs approved for other indications with therapeutic effects on clinically relevant biomarkers. For five of these drugs, the therapeutic effects are replicated in the All of Us Research Program database. We anticipate our approach will enable researchers to integrate multiple publicly available datasets to identify high priority drug repurposing opportunities for human diseases.

Objectives: Phenotyping is a core task in observational health research utilizing electronic health records (EHRs). Developing an accurate algorithm demands substantial input from domain experts, involving extensive literature review and evidence synthesis. This burdensome process limits scalability and delays knowledge discovery. We investigate the potential for leveraging large language models (LLMs) to enhance the efficiency of EHR phenotyping by generating high-quality algorithm drafts. Materials and Methods: We prompted four LLMs—GPT-4 and GPT-3.5 of ChatGPT, Claude 2, and Bard—in October 2023, asking them to generate executable phenotyping algorithms in the form of SQL queries adhering to a common data model (CDM) for three phenotypes (ie, type 2 diabetes mellitus, dementia, and hypothyroidism). Three phenotyping experts evaluated the returned algorithms across several critical metrics. We further implemented the top-rated algorithms and compared them against clinician-validated phenotyping algorithms from the Electronic Medical Records and Genomics (eMERGE) network. Results: GPT-4 and GPT-3.5 exhibited significantly higher overall expert evaluation scores in instruction following, algorithmic logic, and SQL executability, when compared to Claude 2 and Bard. Although GPT-4 and GPT-3.5 effectively identified relevant clinical concepts, they exhibited immature capability in organizing phenotyping criteria with the proper logic, leading to phenotyping algorithms that were either excessively restrictive (with low recall) or overly broad (with low positive predictive values). Conclusion: GPT versions 3.5 and 4 are capable of drafting phenotyping algorithms by identifying relevant clinical criteria aligned with a CDM. However, expertise in informatics and clinical experience is still required to assess and further refine generated algorithms.

Developing algorithms to extract phenotypes from electronic health records (EHRs) can be challenging and time-consuming. We developed PheMap, a high-throughput phenotyping approach that leverages multiple independent, online resources to streamline the phenotyping process within EHRs. PheMap is a knowledge base of medical concepts with quantified relationships to phenotypes that have been extracted by natural language processing from publicly available resources. PheMap searches EHRs for each phenotype’s quantified concepts and uses them to calculate an individual’s probability of having this phenotype. We compared PheMap to clinician-validated phenotyping algorithms from the Electronic Medical Records and Genomics (eMERGE) network for type 2 diabetes mellitus (T2DM), dementia, and hypothyroidism using 84 821 individuals from Vanderbilt Univeresity Medical Center's BioVU DNA Biobank. We implemented PheMap-based phenotypes for genome-wide association studies (GWAS) for T2DM, dementia, and hypothyroidism, and phenome-wide association studies (PheWAS) for variants in FTO, HLA-DRB1, and TCF7L2. In this initial iteration, the PheMap knowledge base contains quantified concepts for 841 disease phenotypes. For T2DM, dementia, and hypothyroidism, the accuracy of the PheMap phenotypes were >97% using a 50% threshold and eMERGE case-control status as a reference standard. In the GWAS analyses, PheMap-derived phenotype probabilities replicated 43 of 51 previously reported disease-associated variants for the 3 phenotypes. For 9 of the 11 top associations, PheMap provided an equivalent or more significant P value than eMERGE-based phenotypes. The PheMap-based PheWAS showed comparable or better performance to a traditional phecode-based PheWAS. PheMap is publicly available online. PheMap significantly streamlines the process of extracting research-quality phenotype information from EHRs, with comparable or better performance to current phenotyping approaches.