Monocyte Trajectories Endotypes Are Associated With Worsening in Septic Patients

Abstract

Sepsis is a life-threatening organ dysfunction caused by a dysregulated host response to infection. The immune system plays a key role in sepsis onset and remains dysregulated over time in a heterogeneous manner. Here, we decipher the heterogeneity of the first week evolution of the monocyte HLA-DR (mHLA-DR) surface protein expression in septic patients, a key molecule for adaptive immunity onset. We found and verified four distinctive trajectories endotypes in a discovery (n = 276) and a verification cohort (n = 102). We highlight that 59% of septic patients exhibit low or decreasing mHLA-DR expression while in others mHLA-DR expression increased. This study depicts the first week behavior of mHLA-DR over time after sepsis onset and shows that initial and third day mHLA-DR expression measurements is sufficient for an early risk stratification of sepsis patients. These patients might benefit from immunomodulatory treatment to improve outcomes. Going further, our study introduces a way of deciphering heterogeneity of immune system after sepsis onset which is a first step to reach a more comprehensive landscape of sepsis.

Trial registration: ClinicalTrials.gov NCT02803346 NCT02638779.

Keywords: ICU– Intensive Care Unit; endotype; flow cytometry; immune monitoring; immunosuppression; monocyte HLA-DR; sepsis; trajectory.

Figure 1

Sepsis mHLA-DR trajectory endotypes. (A) Discovery dataset: “Non-improvers”, “Decliners”, “Improvers” and “High expressors” endotypes from KmL clustering method are represented in each panel. Black lines and dots correspond to individual patients mHLA-DR trajectories and values at the sampling time point, respectively. Blue line and blue band represent the mean endotype trajectory from a mixed effect model and the associated 95% confidence interval. The green area represents Healthy Volunteers reference interval, as defined in (14). (B) Verification dataset “Non-improvers”, “Decliners”, “Improvers” and “High expressors” endotypes from a clustering de novo with KmL method are represented in each panel. Black lines and points respectively correspond to verification cohort’s patients mHLA-DR trajectory and their sampling time point. The green area represents Healthy Volunteers reference interval, has defined in (14). Green solid line and blue dashed line represent the mean trajectory from a mixed effect model of verification cohort and discovery cohort, respectively. (C) Discovery and Verification cohorts were merged in an “Aggregated dataset” and KML unsupervised clustering was run de novo on mHLA-DR expression values. The figure represents patient trajectories in each panel. Black lines and points correspond to aggregated dataset patients mHLA-DR trajectory and their sampling time point, respectively. Red line and red band represent from a mixed effect model the mean endotype trajectory and the associated 95% confidence interval, respectively. The green area represents Healthy Volunteers reference interval, defined in (14). Mean trend mHLA-DR values are reported in sTable 3 .

Figure 2

Mean trajectory per endotype in aggregated dataset. Mean trajectory of mHLA-DR endotypes in aggregated dataset were together drawn. Non-Improvers (red, plain line), Decliners (gold, long dashed line), Improvers (dark blue, dotted line) and High expressors (light blue, dashed and dotted line). 95% confidence intervals were drawn as areas around mean trend. The green area represents Healthy Volunteers reference interval, has defined in (14).

Figure 3

D28 outcomes in mHLA-DR endotypes. (A) Curves of probabilities over 28 days of IAI (top pane), Death (middle pane) and ICU discharge (bottom pane) in Aggregated dataset’s endotypes: Non-improvers (red), Decliners (gold), Improvers (dark blue) and High expressors (light blue). These probabilities were estimated through a survival model of IAI with Death and ICU discharge as competing risks and are expressed as cumulated probabilities. Probabilities at the time of the latest remaining patient at risk were reported on the right hand side of curves. (B) Forest plot of Fine-Gray regression sub-distribution Hazard Ratio (sHR) of outcome by endotype, in comparison with Non-improvers. sHR were reported graphically (black point) and numerically along with 95% Confidence Interval (CI, horizontal bars). sHR significantly different from 1 were reported in bold and p value (p.) was numerically reported.