Macrophage activation markers are associated with infection and mortality in patients with acute liver failure

Abstract

Background and aims: Acute liver failure is a multisystem disorder with a high mortality and frequent need for emergency liver transplantation. Following massive innate immune system activation, soluble markers of macrophage activation are released during liver damage and their association with disease severity and prognosis requires exploration.

Methods: Patients ALF from the United States Acute Liver Failure Study Group (USALFSG, n = 224) and King's College Hospital (n = 40) together with healthy controls (HC, n = 50) were recruited. Serum from early (Days 1-3) and late (>Day 3) time points were analysed for MAMs by enzyme-linked immunosorbent assay correlated to markers of illness severity and 21-day spontaneous survival. Surface expression phenotyping was performed via Flow Cytometry on CD14⁺ monocytes.

Results: All MAMs serum concentrations were significantly higher in ALF compared to controls (p < .0001). sCD206 concentration was higher in early and late stages of the disease in patients with bacteraemia (p = .002) and infection in general (p = .006). In MELD-adjusted multivariate modelling, sCD206 and sCD163 were independently associated with mortality. CD14⁺ monocyte expression of CD206 (p < .001) was higher in patients with ALF compared with controls and correlated with SOFA score (p = .018). sCD206 was independently validated as a predictor of infection in an external cohort.

Conclusions: sCD206 is increased in serum of ALF patients with infections and poor outcome and is upregulated on CD14⁺ monocytes. Later measurements of sCD163 and sCD206 during the evolution of ALF have potential as mechanistic predictors of mortality. sCD206 should be explored as a biomarker of sepsis and mortality in ALF.

Keywords: acute liver failure; innate immunity; macrophages; organ failure; sepsis.

FIGURE 1

Macrophage activation markers in the serum of patients with acute liver failure (ALF) measured by enzyme‐linked immunosorbent assay (ELISA). Six molecules were measured from the cohort of patients with ALF in the acute liver failure study group early (Days 1–3 following admission to hospital) and late (Day 3+) in the course of disease and healthy controls (HC) from volunteers at King's College Hospital. (A, E) Secretory leukocyte peptidase inhibitor (SLPI) and soluble macrophage mannose receptor (sMR) concentrations are higher in both early and late groups compared to HC. (B, F) soluble cluster of differentiation 163 (sCD163) and soluble programmed death ligand 1 (sPD‐L1) concentrations are higher ALF (both groups) compared to HC and higher in early compared to late groups. (C) Soluble Mer tyrosine kinase (sMerTK) concentrations are higher in the early group compared to HC and to the late group. (D) Osteopontin (OPN) concentrations are higher in the early group compared to the late one. *p < .05, **p < .01, ***p < .001, ****p < .0001.

FIGURE 2

Macrophage activation markers in the serum of patients with acute liver failure (ALF) measured by enzyme‐linked immunosorbent assay (ELISA) early (Days 1–3) and late (Day3+) during admission. Aetiologies of ALF considered were acetaminophen (APAP), autoimmune hepatitis (AIH), drug‐induced liver injury (DILI), indeterminate and other aetiology (vascular, ischaemic). (A) SLPI is lowest in patients with autoimmune hepatitis compared to APAP in the early stages of admission while there is no significant difference between different aetiologies in the late course of the disease. (B, C) sCD163 and sMerTK are lower in APAP compared to other aetiologies only in late samples. sMerTK is higher in DILI and indeterminate causes compared to APAP in early and late stages. (D) OPN is higher in patients with APAP compared to autoimmune hepatitis, DILI, indeterminate and other aetiologies in the early stages. The significance is maintained after the third day of admission with higher values in APAP compared to AIH and DILI. (E) sMR is higher in DILI in early stages and in other aetiologies later during admission compared to APAP. (F) sPD‐L1 levels are lower in autoimmune hepatitis, DILI and indeterminate aetiologies compared to those with APAP in early stage of disease. **p < .01, ***p < .001, ****p < .0001.

FIGURE 3

Comparison of macrophage activation markers depending on transplantation status measured by enzyme‐linked immunosorbent assay (ELISA). (A, D, F) Secretoty leukocyte peptidase inhibitor (SLPI), Osteopontin (OPN) and soluble programmed death ligand 1 (sPD‐L1) concentrations are lower in those patients with acute liver failure (ALF) who were transplanted compared to those who survived and died. (C) Soluble Mer tyrosine kinase (sMerTK) concentrations are higher in those ALF who died compared to those who survived. (E) Soluble macrophage mannose receptor (sMR) concentrations are higher in those transplanted ALF compared to those who survived. **p < 0.01***p < 0.001****p < 0.0001.

FIGURE 4

Soluble macrophage mannose receptor (sMR/sCD206) was measured from the cohort of patients with acute liver failure (ALF) in the ALF study group early (Days 1–3 following admission to hospital) and late (Day 3+) divided into those with positive or negative bacteraemia, presence or absence of infection. (A, B) Soluble macrophage mannose receptor (sMR) concentrations are higher in patient with positive bacteraemia compared to those with negative ones at early and late measurements. (D) Soluble macrophage mannose receptor (sMR) concentrations are not significantly different between those with or without presence of infection at an early time point. (E) Soluble macrophage mannose receptor (sMR) concentrations are higher in those with presence of infection compared to those without at a late time point. (C, F) Soluble macrophage mannose receptor (sMR) concentrations are significant in bacteraemia (AUROC = .753, p = 0.0004) and in the presence of infection (AUROC = 0.65, p = 0.007) in the late stage of disease. **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 5

Soluble macrophage mannose receptor (sMR/sCD206) was measured in patients with acute liver failure (ALF) from the validation cohort on Days 1 and 3–7 following hospital admission. (A) Soluble macrophage mannose receptor (sMR) concentrations are higher in ALF (n = 40) compared to HC (n = 16); (B) no difference was found between admission levels and sequential samples taken between Day 3 and 7 after admission; (C) no difference in sMR was found between ALF patients spontaneously survived compared to patients deceased or transplanted (Tx) within 90 days from admission (samples from both Day 1 and Day 3+); (D) sMR directly correlates with Sequential Organ Failure Assessment (SOFA) score (top) and Model for End‐Stage Liver Disease (MELD) (bottom); (E) sMR is increased in patients with positive culture (blood, sputum or urine) both at baseline and at Day 3+; (F) ROC Curve representing sMR performance in detecting infection in ALF patients at Day 1 and Day 3+. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 6

Phenotypic characterisation of CD14⁺ cells selected with flow cytometry from peripheral blood mononuclear cells (PBMCs) with acute liver failure (ALF) and healthy controls (HC). (A) Gating strategy from all cells, single cells, live/dead and CD14⁺ cells; (B) CD206, PD‐L1 and HLA‐DR expression in CD14+ cells in ALF compared to HC. (C) Surface expression of CD206 is negatively correlated with MELD/SOFA suggesting increased monocyte/macrophage shedding of these molecules with worsening disease in ALF. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.