Abstract Rationale Neutropenic sepsis is a life-threatening condition associated with high rates of morbidity and mortality relative to non-neutropenic populations. Targeted therapies to modulate the host response are needed, however very little is understood of how neutropenia alters sepsis immunobiology. Here, we compared plasma protein biomarker levels between neutropenic and non-neutropenic septic patients with the aim of characterizing neutropenia-specific dysregulated responses. Methods Baseline plasma protein biomarkers from two observational cohorts (VALID; n = 1133, MARS; n = 1396) and one randomized clinical trial (CLOVERS, n = 1552) of critically ill patients with sepsis were analyzed by neutropenic status. Biomarkers associated with systemic inflammation and endothelial activation were chosen for analysis. Since differential cell counts were unavailable, neutropenia was defined as a minimum leukocyte count ≤1000/µL, which is consistent with prior work and has a high specificity for severe neutropenia. To visualize potential group differences, principal component analysis (PCA) plots were constructed from log-transformed data. Untransformed biomarker concentrations were compared using the Wilcoxon rank-sum test. Finally, the proportion of patients with the previously assigned hyperinflammatory phenotype, determined through latent class analysis, were compared using the Chi-square test. As a sensitivity analysis, we repeated the above analyses to include a neutropenic definition of leukocyte count ≤1500/µL. Results A total of 47 (4.1%), 60 (4.3%), and 72 (4.6%) patients met our definition for neutropenia in VALID, MARS, and CLOVERS, respectively. Mean leukocyte counts ranged from 257-469/µL in neutropenic patients and 14,438-16,712/µL in non-neutropenic patients. Across all three cohorts, neutropenia was associated with a distinct protein biomarker signature as evidenced by visual PCA data separation (Figure1 A-C, left panels). At the individual biomarker level, neutropenic patients displayed consistently higher concentrations of inflammatory mediators, including interleukin (IL)-6, IL-8, IL-10 (p 0.001 for all), and soluble tumor necrosis factor receptor-1 (p≤0.011) (Figure1 A-C, right panels). Conversely, concentrations of angiopoietin (ANG)-1 (p 0.001), e-Selectin (p 0.001), and p-Selectin (p 0.001) were lower in neutropenic patients. No consistent trend was observed for ANG-2 or intercellular adhesion molecule-1. The hyperinflammatory phenotype was more frequent amongst neutropenic patients compared to non-neutropenic patients (58-96% vs 22-38%; p 0.001). These findings were consistent in sensitivity analyses that expanded the definition of neutropenia to a leukocyte count ≤1500/µL. Conclusions Compared to non-neutropenic patients, those with neutropenic sepsis exhibit a stronger proinflammatory biomarker profile, lower markers of endothelial activation, and a higher prevalence of the hyperinflammatory phenotype. These findings suggest that neutropenic sepsis represents a distinct biological subgroup warranting consideration in future trials testing immunomodulatory therapies. This abstract is funded by: R35HL177135 (CSC), R01HL126176 (LBW), K01HL157755 (VEK), R01HL173531 (PS), R35GM142992 (PS)
Hartlage et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: