Immunophenotyping patients with sepsis and underlying haematological malignancy reveals defects in monocyte and lymphocyte function

To the Editor,

Sepsis is a common reason for intensive care unit (ICU) admission of patients with haematological malignancy [1]. The main focus is placed on neutropenia, with little attention paid to other white cell lineage such as monocytes and lymphocytes. Immune dysfunction in these cells is well-described in non-cancer septic patients and associated with an increased mortality risk [2,3,4]. Features typically associated include impaired monocyte antigen presentation and co-stimulation (HLA-DR, CD80, CD86), increased immune checkpoint inhibition (lymphocyte PD-1 and monocyte PD-L1), impaired lymphocyte proliferation/ maturation (IL-7 receptor), activation (CD28 and CTLA-4), and viability [2,3,4]. The primary objective of this feasibility study was to ascertain whether these cells are similarly affected in haematology patients with sepsis.

We conducted a prospective observational study in patients with or without haematological malignancy admitted to the ICU with sepsis. Peripheral blood mononuclear cells (PBMC) were isolated and assessed by multi-parameter flow cytometry, and serum immune analytes by ELISA (Additional file 1: Methods). A focused analysis was performed of cell surface markers associated with sepsis-induced immunosuppression [2,3,4].

We included 11 haematology ICU patients, 33 non-haematology ICU patients (and 17 healthy volunteers as a reference). Patient demographics are detailed in Additional file 1: Table S2. Compared to non-haematology patients, haematology patients were of similar age and had a similar SOFA score. However, compared to non-haematology patients, haematology patients had lower neutrophils (p < 0.0001), lymphocytes (p = 0.03), and monocytes (p = 0.005). Hospital mortality was similar between both groups (27% non-haematology vs. 36% haematology) (Fig. 1, Additional file 1: Fig. S1).

Fig. 1

Differences in clinical variables, monocyte and lymphocyte function between non-haematology and haematology patients. Comparison of patients admitted to the Intensive Care Unit with a non-haematology (Non-HO, n = 33), or haematology (HO, n = 11) diagnosis. Healthy volunteers (n = 17) are included as a reference. Innate immune response (a.) including neutrophil count (i.), monocyte count (ii.), and monocyte phagocytosis as measured by pHRodo (iii.), with example contour plot of non-HO (iv.) and HO (v.). Adaptive immune response (b.–c.) including lymphocyte count (b.i.) CD4 lymphocyte IL-7 receptor (IL-7R) expression (b.ii.) and correlation plot of IL-7R with percentage cell death of non-HO (b.iii.) and HO (b.iv.), apoptosis (c.i.), and programmed cell death receptor-1 (PD-1) expression (c.ii.) correlation plot of PD-1 with percentage cell death of non-HO (c.iii.) and HO (c.iv.) patients. LPS-induced cytokine release (d.) including IL-1β (i.), TNF-α (ii.) and IL-10 (iii.). Data compared using Mann Whitney test. Only p < 0.1 shown

There was a trend towards decreased monocyte phagocytosis (p = 0.055) among haematology patients. Viability in lymphocyte CD4 and CD8 cell populations and CD4 IL-7R levels were lower among haematology patients (Fig. 1, Additional file 1: Figs. S2, S3). A positive correlation was seen between PD-1 expression and cell death in CD4 lymphocytes in non-haematology patients but not haematology patients (Fig. 1).

Serum TNF-α was higher among haematology patients, although monocyte intracellular TNF-α levels were similar. Following ex vivo whole blood stimulation with LPS, serum IL-1β (p = 0.043) and TNF-α (p = 0.001) increased significantly in non-haematology patients, but not in haematology patients. (Fig. 1).

We present novel data demonstrating immune dysfunction in monocytes and lymphocytes taken from haematology patients with sepsis; over and above that seen in non haematology patients. This included impaired monocyte phagocytosis, and impaired release of TNF-α and IL-1β (canonical cytokines associated with monocyte function) on whole blood stimulation with LPS. Intriguingly, monocyte HLA-DR, a robust functional marker of immunoparesis in critically ill patients [4], was not different in haematology patients.

Mechanisms of lymphocyte death are likely to differ between haematology and non-haematology patient cohorts. The association between CD4 lymphocyte PD-1 expression and cell death is also described in patients with sepsis [4]. We found a positive correlation between PD-1 expression in CD4 lymphocytes in non-haematology patients but not in haematology patients.

Existing therapies to improve clinical outcomes in the critically ill haematology patient with sepsis are limited. Further research is required to gain a better understanding of the immune phenotype in this population, providing a rational for individualized sepsis treatment.