Abstract Rationale In ARDS, the extent of inflammation, as characterized by circulating cytokines and innate immune activation markers, partitions patients into prognostic subphenotypes. Conventional protein markers of endothelial cell (EC) activation, while enriched, may not best distinguish the hyperinflammatory state. We previously demonstrated that endothelial glycocalyx (EGCX) degradation identifies clinically relevant biological heterogeneity in pediatric ARDS. This transmembrane, apical EC layer, when degraded, releases glycan fragments into the circulation, disrupting EC homeostasis and contributing to organ injury. We tested the hypothesis that EGCX degradation identifies both a distinct and principal axis of biological variation in pediatric ARDS. Methods We conducted a retrospective analysis of prospectively collected plasma samples (2018-2020) from a cohort of children with ARDS. Plasma levels of glycosaminoglycans (i.e., heparan and chondroitin sulfate, including their respective sulfation motifs, and hyaluronic acid) were measured using HPLC-MS/MS. We quantified circulating proteoglycans and protein biomarkers reflective of lung epithelial injury, EC activation, and inflammation using a custom multiplex magnetic bead assay. Each patient had 28 biomarkers, which were log-transformed and standardized. Principal component (PC) analysis was employed to identify distinct biological axes. Each PC represented a weighted linear combination of biomarkers, with loading coefficients quantifying the magnitude and direction of each biomarker’s contribution to each PC. Positive coefficients indicated a positive correlation between the biomarker and corresponding PC. For each patient, both PC scores and outcomes were standardized for use in regression analyses. Results This study included 36 children with ARDS. PC1-PC3 were retained based on scree plots and permutation testing, collectively explaining 63% of the total variance (PC1: 40%, PC2: 12%, PC3: 11%; all p 0.0001). The variance captured by PC1 was primarily defined by EGCX constituents, with 6-O- and N-sulfated heparan sulfate exhibiting the largest positive loading coefficients. In contrast, PC2 reflected an inflammatory signature characterized by IL-1β, IL-6, IL-10, and TNF-α (Figure 1A-B). Higher PC1 scores were associated with worse organ dysfunction, as assessed by PELOD-2 (β = 0.34, 95% CI 0.02, 0.66, p = 0.035), and fewer 28-day ventilator-free days (β=-0.33, 95% CI -0.64, -0.02, p = 0.038), adjusting for PC2 and PC3. Conclusions Our findings, while limited by sample size and in need of external validation, suggest that EGCX heparan sulfate degradation identifies a distinct biological axis in pediatric ARDS, separate from other glycans and the canonical inflammatory cytokines that define the hyperinflammatory subphenotype. These results underscore the potential of endothelial glycobiology to guide biologically informed phenotyping and risk stratification in pediatric ARDS. This abstract is funded by: Children’s Discovery and Innovation Institute at UCLA; K12HD047349; K23HD096018
Sallee et al. (Fri,) studied this question.