Abstract
Rationale: Sustained sepsis-associated immunosuppression is associated with uncontrolled infection, multiple organ dysfunction, and death.
Objectives: In the first controlled biomarker-guided immunostimulatory trial in sepsis, we tested whether granulocyte–macrophage colony-stimulating factor (GM-CSF) reverses monocyte deactivation, a hallmark of sepsis-associated immunosuppression (primary endpoint), and improves the immunological and clinical course of patients with sepsis.
Methods: In a prospective, randomized, double-blind, placebo-controlled, multicenter trial, 38 patients (19/group) with severe sepsis or septic shock and sepsis-associated immunosuppression (monocytic HLA-DR [mHLA-DR] <8,000 monoclonal antibodies (mAb) per cell for 2 d) were treated with GM-CSF (4 μg/kg/d) or placebo for 8 days. The patients' clinical and immunological course was followed up for 28 days.
Measurements and Main Results: Both groups showed comparable baseline mHLA-DR levels (5,609 ± 3,628 vs. 5,659 ± 3,332 mAb per cell), which significantly increased within 24 hours in the GM-CSF group. After GM-CSF treatment, mHLA-DR was normalized in 19/19 treated patients, whereas this occurred in 3/19 control subjects only (P < 0.001). GM-CSF also restored ex-vivo Toll-like receptor 2/4–induced proinflammatory monocytic cytokine production. In patients receiving GM-CSF, a shorter time of mechanical ventilation (148 ± 103 vs. 207 ± 58 h, P = 0.04), an improved Acute Physiology and Chronic Health Evaluation-II score (P = 0.02), and a shorter length of both intrahospital and intensive care unit stay was observed (59 ± 33 vs. 69 ± 46 and 41 ± 26 vs. 52 ± 39 d, respectively, both not significant). Side effects related to the intervention were not noted.
Conclusions: Biomarker-guided GM-CSF therapy in sepsis is safe and effective for restoring monocytic immunocompetence. Use of GM-CSF may shorten the time of mechanical ventilation and hospital/intensive care unit stay. A multicenter trial powered for the improvement of clinical parameters and mortality as primary endpoints seems indicated.
Clinical trial registered with www.clinicaltrials.gov (NCT00252915).
Many nonsurvivors from sepsis die in the later course of sepsis in a state of functional failure of cellular immunity.
This study demonstrates that immunostimulation with granulocyte–macrophage colony-stimulating factor effectively restores markers of monocytic immunocompetence. A beneficial impact on the clinical course of patients with severe sepsis and septic shock was also observed.
Sepsis-associated immunosuppression is characterized by impaired innate and adaptive immune responses, including enhanced apoptosis and dysfunction of lymphocytes, impaired phagocyte functions, monocytic deactivation with diminished HLA class II surface expression, and altered ex vivo cytokine production (3, 5, 8, 9, 11–21). Therefore, we proposed the term “immunoparalysis” to describe the hosts' global inability to mount effective antimicrobial immune responses in this state (22, 23). This immunosuppression may be induced by exogenous factors, such as endotoxin and immunosuppressive drugs (15, 22, 24–26), or by endogenous mediators, such as IL-10 (26–32), corticosteroids (30, 33, 34), and catecholamines (35). We and others have demonstrated an association between low levels of monocytic HLA-DR (mHLA-DR) surface expression and immune cell dysfunctions in patients with sepsis. Based on these findings, we proposed reduced mHLA-DR expression as a global biomarker of sepsis-associated immunosuppression (22, 23). Prolonged downregulation of mHLA-DR has been associated with reduced survival in patients with sepsis (9, 10, 12, 16, 17, 21, 22, 33, 36–43). However, other authors did not find this association (discussed in Refs. 36, 44). These contrasting results may be due to both the timing of mHLA-DR determination and the testing procedures used in the past. Although a large trial investigating whether diminished mHLA-DR expression is associated with adverse outcome from sepsis is still needed, reduction of immunosuppressive drugs or immunostimulation in an effort to prevent secondary/nosocomial infections has been proposed (9, 15–17, 22, 45).
Granulocyte–macrophage colony-stimulating factor (GM-CSF) is a 23-kD growth factor with potent immunostimulatory effects. GM-CSF enhances antimicrobial host defense by improving survival, proliferation, differentiation, phagocytosis, and bacterial killing of neutrophils and monocytes/macrophages. Furthermore, GM-CSF may promote migration and adhesion of neutrophils (46). GM-CSF has been demonstrated to increase mHLA-DR expression and endotoxin-induced proinflammatory cytokine production in ex vivo whole blood cultures of patients with severe sepsis (47, 48). Moreover, recent pilot trials indicated that long-lasting monocyte deactivation in sepsis may be reversed in vivo by application of immunostimulants, such as GM-CSF (49), IFN-γ (41, 50), or by removal of inhibitory factors (51).
Controlled, biomarker-guided studies on immunostimulatory therapies for the treatment of patients with sepsis have not been performed thus far. The availability of standardized monitoring of mHLA-DR may now allow biomarker-guided, multicenter trials (43). In this prospective, randomized, double-blind, placebo-controlled, multicenter trial, we tested the efficacy of GM-CSF to reverse sepsis-associated immunosuppression as assessed by restoration of mHLA-DR expression.
A total of 38 patients with severe sepsis or septic shock and sepsis-induced immunosuppression were included in a prospective, randomized, double-blind, placebo-controlled, multicenter trial. Sepsis-induced immunosuppression was defined as reduced mHLA-DR levels of less than 8,000 monoclonal antibodies (mAb) per cell (43) in two consecutive measurements (2 d before inclusion; Figure 1). The primary outcome measure was mHLA-DR expression. Secondary endpoints included course of disease severity, cellular immunity, inflammatory markers, and length of hospital/intensive care unit (ICU) stay. Patient enrollment was from November 2005 to January 2007. Patients from three tertiary care academic centers situated in Berlin, Germany, were included (24-bed medical ICU [CC13], 12-bed surgical ICU [CC8], 14-bed anesthesiological ICU [CC7]). Inclusion and exclusion criteria and patient allocation is given in the online supplement (Table E1 and Figure E1). Fatal outcomes were categorized as sepsis related or sepsis unrelated. Intention-to-treat and per-protocol analyses were performed (Figure E1). The study was approved by the local Ethics Committee on human research and designed in adherence to the Declaration of Helsinki. Written informed consent was obtained from patients or legal representatives.
Figure 1. Study paradigm. APACHE = Acute Physiology and Chronic Health Evaluation; GM-CSF = granulocyte–macrophage colony-stimulating factor; mAB = monoclonal antibody; SAPS = Simplified Acute Physiology Score; SOFA = Sequential Organ Failure Assessment; TISS = Therapeutic Intervention Scoring System.
[More] [Minimize]After inclusion, patients were randomized by the respective hospitals' pharmacy in permutated blocks of four (stratified for age and sex) to receive either GM-CSF or placebo (0.9% sodium chloride). Beginning on the day thereafter, GM-CSF group patients received subcutaneous injections of 4 μg/kg body weight per day GM-CSF (Leukine; Berlex Laboratories, Richmond, CA) for 5 consecutive days. GM-CSF or placebo was continued for 3 additional days at either 8 μg/kg/day GM-CSF (if mHLA-DR was ≤15,000 mAb per cell at Day 5), or 4 μg/kg/day (mHLA-DR >15,000 mAb per cell) (Figure 1). Follow up was performed until Day 28, hospital discharge, or death. Physicians in charge were unaware of group assignments and provided treatment without interference by the researchers. All participating ICUs have established standard operating procedures to guide ICU therapy in adherence to current international guidelines. This includes central venous oxygen saturation guided fluid and vasopressor regimen, steroid regimen (continuous intravenous infusion of 240 mg/d hydrocortisone in volume-refractory hemodynamic failure), analgosedation and ventilation regimen (pressure control mode), and empiric antibiotic therapy based on current recommendations.
For detailed information, see the online supplement.
Disease severity was assessed using the following scores: Acute Physiology and Chronic Health Evaluation II (APACHE-II), Sequential Organ Failure Assessment (SOFA), Simplified Acute Physiology Score-II (SAPS-II), and Therapeutic Intervention Scoring System (TISS)-28 (Figure 1).
To detect a recovery rate (mHLA-DR increase from <8,000 to >15,000 mAb per cell) of 80% (GM-CSF group) versus a spontaneous recovery rate of 30% (placebo group) in the intervention interval, with an 80% power at a two-sided P value of 0.05, 19 patients per group needed to be included in the trial. No interim analyses were performed. For statistical analyses, MedCalc 9.0.1 software was used (MedCalc, Mariakerke, Belgium). Analysis of variance (ANOVA) for repeated measures, Student's t test, and Fisher's exact test were used, as appropriate. Subgroup analyses were performed using ANOVA with Fisher's post hoc test. All data were checked for normal distribution (Kolmogorov-Smirnov test). Non-normally distributed data were log transformed. Results refer to the intention-to-treat population and are reported as mean (±SD), if not indicated otherwise. A P value of less than 0.05 was considered significant.
A total of 135 patients were screened, and all patients eligible for study inclusion entered the analysis (Figure E1): 38 white ICU patients (19 medical ICU, 15 surgical ICU, 4 anesthesiological ICU) with severe sepsis and septic shock (82% male; mean age, 63.6 ± 13.7 yr, mean APACHE-II score, 21.9 ± 6.3) were included. Major etiologies of sepsis were pneumonia and peritonitis (Table 1); 14 patients (37%) had evidence for gram-positive infection, 12 patients (32%) with mixed gram-positive/gram-negative infection, 8 patients (21%) with gram-negative infection, and 3 patients (8%) had a fungal infection. At baseline, the intervention and control groups did not differ in regard to age, sex, body mass index, presence of shock, source of sepsis/etiology, distribution of microbial pathogens, days on ICU until study inclusion, presence of acute renal failure, need for renal replacement therapy, norepinephrine dose, and APACHE-II, TISS-28, SAPS-II, and SOFA scores (all P > 0.05) (Table 1). The onset of severe sepsis and septic shock in the majority of study patients was at the time of ICU admission/referral. However, as exact quantification of onset of sepsis can be difficult to achieve, this must be interpreted with caution. Major concomitant diseases at baseline included arterial hypertension (57%), hyperlipoproteinemia (34%), diabetes mellitus (32%), chronic obstructive pulmonary disease (21%), coronary artery disease (18%), and chronic renal insufficiency (18%, minimum stage 3). Between-group differences in the distribution of concomitant diseases at baseline were not identified.
Characteristics | GM-CSF Group (n = 19) | Placebo Group (n = 19) | P Value |
|---|---|---|---|
| Age, years | 64.0 ± 13.6 | 63.3 ± 14.2 | NS* |
| Sex, male (%) | 16/19 (84) | 15/19 (79) | NS† |
| Body mass index | 27.4 ± 6.3 | 26.5 ± 4.8 | NS* |
| Septic shock at baseline (%) | 11/19 (58) | 10/19 (53) | NS† |
| Major source of sepsis at baseline (%) | |||
| Pneumonia | 11/19 (58) | 10/19 (52) | NS† |
| Peritonitis | 6/19 (32) | 5/19 (26) | NS† |
| Other | 2/19 (11) | 4/19 (21) | NS† |
| Mortality rate at study Day 28 (%) | 3/19 (16) | 4/19 (21) | NS† |
| Days on ICU until study inclusion | 6.0 ± 3.3 | 8.47 ± 8.9 | NS* |
| Length of ICU stay, days | 40.9 ± 26.1 | 52.1 ± 39.6 | NS* |
| Total intrahospital stay, days | 58.8 ± 32.6 | 68.9 ± 45.6 | NS* |
| Need for RRT | |||
| ARF at baseline (%) | 12/19 (63) | 11/19 (58) | NS† |
| Days on RRT | 14.4 ± 10.2 | 11.5 ± 10.2 | NS* |
| Time on ventilator, Days 1–9, hours | 147.9 ± 102.8 | 207.2 ± 57.5 | 0.037* |
| Norepinephrine dose, μg/kg/min | |||
| Study Day 1 | 0.19 ± 0.17 | 0.18 ± 0.17 | NS* |
| Study Day 9 | 0.12 ± 0.13 | 0.14 ± 0.13 | NS* |
| Days 1 vs. 9 | NS* | NS* | — |
| APACHE-II | |||
| Study Day 1 | 21.3 ± 6.1 | 22.5 ± 6.6 | NS* |
| Study Day 9 | 16.7 ± 5.9 | 20.8 ± 7.4 | (0.06)* |
| Days 1 vs. 9 | P = 0.02* | NS* | — |
| SOFA | |||
| Study Day 1 | 7.2 ± 4.0 | 9.5 ± 3.7 | NS* |
| Study Day 9 | 5.2 ± 3.4 | 7.5 ± 5.3 | NS* |
| Days 1 vs. 9 | P = NS* | NS* | — |
| SAPS-II | |||
| Study Day 1 | 38.8 ± 8.9 | 45.6 ± 13.4 | NS* |
| Study Day 9 | 37.3 ± 10.6 | 45.0 ± 14.9 | NS* |
| Days 1 vs. 9 | NS* | NS* | — |
| TISS-28 | |||
| Study Day 1 | 42.4 ± 9.5 | 46.5 ± 9.9 | NS* |
| Study Day 9 | 38.1 ± 13.8 | 42.7 ± 11.2 | NS* |
| Days 1 vs. 9 | NS* | NS* | — |
A total of 37 patients completed the 9-day treatment interval. One patient in the GM-CSF group died at study Day 8 from reasons related to sepsis-induced hemodynamic failure, and thus did not complete the treatment interval. All other patients were treated as established in the protocol, and received either a subcutaneous injection of 4 μg/kg body weight GM-CSF per day or placebo for 8 consecutive days, and thus completed the study protocol. Dose escalation to 8 μg/kg of body weight GM-CSF per day was performed in two patients of the intervention arm, because mHLA-DR expression was less than 15,000 mAb per cell at Day 5. In the placebo group, two individuals were excluded from the per-protocol analysis due to protocol violation and a suspected severe adverse event, respectively (Figure E1).
The course of IL-6 and IL-10 levels and platelet counts was not different between the study groups (Table 2). A tendency toward higher baseline IL-6 and IL-10 levels in the GM-CSF group was observed (both not significant [NS]). During immunotherapy, the plasma level of tumor necrosis factor (TNF)-α moderately increased in the GM-CSF group (P = 0.02), whereas it was unchanged in control subjects. At Day 9, no significant between-group differences were indentified (Table 2). Baseline procalcitonin (PCT) levels were higher in the GM-CSF group (Table 2, NS). Although PCT levels decreased in both study groups, the decrease was more prominent in patients receiving GM-CSF (Table 2). As expected, GM-CSF plasma levels significantly increased in the treatment group until study Day 5 (Table 2, P = 0.04 vs. baseline). In contrast, GM-CSF levels remained unchanged over the treatment interval in patients receiving placebo.
Normal Range | Study Day 1 (Baseline) | Study Day 9 (After Therapy) | Between-Group P Value | P Value Days 1 vs. 9 | ||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Study Day 5 | Study Day 1 | Study Day 5 | Study Day 9 | |||||||
| mHLA-DR, mAb/cell | >15,000 | NS | P < 0.001 | P < 0.0001 | ||||||
| GM-CSF | 5,609 ± 3,628 | 43,676 ± 24,517 | 50,907 ± 28,568 | P < 0.0001 | ||||||
| Placebo | 5,659 ± 3,332 | 7,814 ± 5,787 | 10,426 ± 8,424 | NS | ||||||
| IL-6, pg/ml | <5 | NS | NS | NS | ||||||
| GM-CSF | 147.2 ± 256.5 | 87.4 ± 164.5 | 136.6 ± 240.7 | NS | ||||||
| Placebo | 78.6 ± 144.9 | 116.8 ± 230.9 | 190.3 ± 339.0 | NS | ||||||
| IL-10, pg/ml | <5 | NS | NS | NS | ||||||
| GM-CSF | 27.4 ± 64.9 | 22.9 ± 44.4 | 26.4 ± 52.6 | NS | ||||||
| Placebo | 17.8 ± 19.5 | 12.8 ± 9.5 | 16.0 ± 19.6 | NS | ||||||
| TNF-α, pg/ml | <10 | NS | NS | NS | ||||||
| GM-CSF | 12.7 ± 4.4 | 11.6 ± 5.8 | 20.9 ± 13.5 | P = 0.02 | ||||||
| Placebo | 22.3 ± 22.2 | 18.3 ± 15.2 | 22.0 ± 16.7 | NS | ||||||
| Procalcitonin, μg/L | <0.5 | NS | NS | NS | ||||||
| GM-CSF | 17.3 ± 43.8 | 5.3 ± 14.2 | 5.2 ± 2.1 | (P = 0.13) | ||||||
| Placebo | 8.7 ± 18.2 | 3.1 ± 6.0 | 4.4 ± 9.4 | NS | ||||||
| Platelet count per nl | 150–400 | NS | NS | NS | ||||||
| GM-CSF | 235.0 ± 132.8 | 266.0 ± 144.1 | 271.2 ± 155.5 | NS | ||||||
| Placebo | 167.9 ± 109.8 | 216.6 ± 147.4 | 225.2 ± 170.1 | NS | ||||||
| GM-CSF level, pg/ml | <2.5 | NS | P = 0.05 | P = 0.03 | ||||||
| GM-CSF | 2.1 ± 2.2 | 40.3 ± 77.1 | 25.5 ± 62.9 | (P = 0.12) | ||||||
| Placebo | 2.1 ± 2.0 | 1.9 ± 2.4 | 1.8 ± 1.4 | NS | ||||||
After one dose of GM-CSF, neutrophil numbers increased significantly in the treated versus control patients (Figure 2A). Similarly, beginning on study Day 4, monocyte numbers were higher in the GM-CSF group (Figure 2B). The numbers of CD4+ and CD8+ T-lymphocyte subsets increased over time in patients treated with GM-CSF, but not in control patients (ANOVA for repeated measurements, P < 0.05 for both subsets), and, after therapy at Day 9, T-cell counts were significantly higher in the intervention group (Figures 2C and 2D). In contrast, the course of natural killer cells and B lymphocytes was not significantly different between groups (Figures 2E and 2F). For assessment of monocytic function, we analyzed mHLA-DR expression and ex vivo cytokine release in response to the Toll-like receptor (TLR) 4 stimulant endotoxin and TLR2 stimulant, Pam3CSK4 (Figures 3A–3E and 4A–4D). A rapid increase of mHLA-DR expression was observed in all patients in the GM-CSF group as early as after one treatment (Figure 3A and Table 2). Only two patients did not reach normal levels (>15,000 mAb per cell) within 5 days, and required dose escalation. The effect remained stable over the treatment interval, and mHLA-DR levels normalized in 19 out of 19 patients treated with GM-CSF, whereas mHLA-DR normalized in only 3 out of 19 patients of the placebo group over the intervention interval (P < 0.001). For individual data dot-plot, please refer to Figure E2a. Endotoxin-induced monocytic TNF-α release was significantly higher in patients treated with GM-CSF within the first 5 days of treatment (Figure 3B; individual data plot given in Figure E2b). In addition, significantly higher endotoxin-induced IL-6 (Figure 3C) and IL-8 (Figure 3D) production was observed in individuals treated with GM-CSF at study Days 5 and 9. By contrast, IL-10 release showed a trend toward lower levels in the treatment group (Figure 3E). Similarly, TLR2-induced monocytic TNF-α, IL-6, and IL-8 production recovered faster in patients treated with GM-CSF, whereas IL-10 release after Pam3Cys stimulation significantly declined over the treatment interval in patients receiving GM-CSF (Figures 4A–4D). Overall, the data indicate that monocyte proinflammatory cytokine release and mHLA-DR expression were restored under GM-CSF treatment. The course of mHLA-DR in survivors versus nonsurvivors is given as an individual data dot plot (Figure E2c).
Figure 2. Impact of granulocyte–macrophage colony-stimulating factor (GM-CSF) on leukocyte counts and leukocyte subpopulations—rising numbers of neutrophils, monocytes, and T lymphocytes. Patients with sepsis and sepsis-associated immunosuppression were treated with GM-CSF or placebo for 8 days. Leukocyte counts and respective subpopulations were assessed in patients receiving GM-CSF (solid symbols) and placebo (open symbols) at study Days 1–9. Absolute numbers (means ± SD) of (A) neutrophils, (B) monocytes, (C) CD4+ T-lymphocytes, (D) CD8+ T-lymphocytes, (E) B-lymphocytes, and (F) natural killer (NK) cells. Dashed lines indicate lower limits of normal. Repeated measurements analysis of variance for all subsets P < 0.05, except for NK cells and B-lymphocytes. *P < 0.05; **P < 0.01; ***P < 0.001 for between-group comparisons at the indicated study days.
[More] [Minimize]
Figure 3. Recovery of monocytic HLA-DR (mHLA-DR) expression and Toll-like receptor (TLR) 4–induced cytokine release under treatment with granulocyte–macrophage colony-stimulating factor (GM-CSF). Patients with sepsis and sepsis-associated immunosuppression were treated with GM-CSF or placebo for 8 days. Course of (A) mHLA-DR expression (dashed line indicates 15,000 mAb/cell) and (B) monocytic tumor necrosis factor (TNF)-α release after 4 hours of endotoxin stimulation at study Days 1 to 9. At study Days 1, 5, and 9, monocytic cytokine release was assessed after 4 hours (24 hr for IL-10) of endotoxin stimulation: (C) IL-6, (D) IL-8, (E) IL-10 (all means ± SD). Solid and gray symbols/plots represent GM-CSF (n = 19), and open symbols/plots indicate placebo (n = 19) individuals. *P < 0.05; **P < 0.01; ***P < 0.001 between groups at the indicated study day. #P < 0.05; ##P < 0.01; ###P < 0.001 versus Day 1 in the same group. §P < 0.05; §§P < 0.01 versus Day 5 in the same group (repeated measurement analysis of variance, all P < 0.05).
[More] [Minimize]
Figure 4. Recovery of Toll-like receptor (TLR) 2–induced monocytic cytokine release by granulocyte–macrophage colony-stimulating factor (GM-CSF) treatment. Patients with sepsis and sepsis-associated immunosuppression were treated with GM-CSF or placebo for 8 days. At study Days 1, 5, and 9, TLR2-induced monocytic cytokine release was assessed. Gray box-plots represent GM-CSF (n = 19), and open box-plots indicate placebo (n = 19) individuals. Monocytic cytokine release was assessed 4 hours (24 hr for IL-10) after Pam3CSK4 stimulation: (A) IL-6, (B) IL-8, (C) IL-10, (D) tumor necrosis factor (TNF)-α. *P < 0.05, **P < 0.01 between groups at the indicated study day; #P < 0.05, ##P < 0.01, ###P < 0.001 versus Day 1 in the same group; §P < 0.05, §§§P < 0.001 versus Day 5 in the same group (repeated measurement analysis of variance, all P < 0.05).
[More] [Minimize]Time of mechanical ventilation (P = 0.037), length of ICU stay, and length of intrahospital stay (both NS) were all shorter in patients receiving GM-CSF (Table 1). Patients in the GM-CSF group required more vasopressor support (norepinephrine) at baseline. After the intervention interval, the required norepinephrine dose was lower in the treatment group (Table 1). Over the intervention interval, the respective norepinephrine need was reduced by 37% in the GM-CSF group and by 22% in control patients, respectively (Table 1, NS). The number of patients with acute renal failure at baseline and the number of days on renal replacement therapy was not different between the study groups (Table 1). Patients in the control group had a tendency toward higher baseline APACHE-II, SOFA, SAPS-II, and TISS-28 scores. This, however, did not reach statistical significance (all P ≥ 0.07, Table 1). Mean APACHE-II, SOFA, and other scores improved in both study groups over the treatment interval (Table 1). Nevertheless, observed improvements were greater in patients receiving GM-CSF. Over the intervention interval, APACHE-II, SOFA, SAPS II, and TISS-28 scores were reduced by 22, 28, 4, and 10%, respectively, in the GM-CSF group, whereas the respective scores were reduced by 8, 21, 1, and 8% in control patients. After the intervention interval (at study Day 9), a significant decline in APACHE-II scores was observed in the GM-CSF group (P = 0.02, Table 1), but not in the placebo group compared with baseline. When compared with the control group, a trend toward significantly lower APACHE-II scores was found in the GM-CSF group at study Day 9 (P = 0.06).
The 28-day mortality was not different between the study groups (16 vs. 21%, NS; Table 1). In total, seven patients died within the 28-day observation interval, and five of them died from sepsis-related reasons (hemodynamic failure, NS between groups). Six (86%) of the nonsurvivors died while still hospitalized (n = 1 in rehabilitation center, NS between groups). Importantly, as the study was not powered to assess mortality, the respective data should be interpreted with caution. Disease severity, as assessed by clinical scores (APACHE-II, SAPS-II, SOFA, and TISS-28) continued to improve through the 28-day follow-up period in both study groups. At study Day 28, there were no differences between the study groups (P ≥ 0.25 for all comparisons). Importantly, side effects, which may theoretically evolve under immunostimulatory therapy with GM-CSF (e.g., anaphylactic or skin reactions, increased capillary leakage, and metabolic disturbances), were not observed.
We demonstrate in the first immune monitoring–based, randomized, placebo-controlled trial that immunostimulation with GM-CSF is a safe and effective measure to restore mHLA-DR expression and cytokine release in patients with sepsis and sepsis-associated immunosuppression. Furthermore, GM-CSF shortened the time of mechanical ventilation, as well as intrahospital and ICU stay (Table 1). A trend toward improved disease severity (clinical scores) was observed under treatment, whereas 28-day mortality was not different between the two study groups (Table 1).
Cellular immunity is of major importance for an effective clearance of bacteria. Disturbed monocytic phagocytosis, diminished cytokine expression and antigen presentation, lymphocyte dysfunction, and apoptosis may typically be observed in the later phase of sepsis (3, 6, 16, 17, 20, 52). Today, many nonsurvivors from sepsis die from complications in this later stage (3, 8, 11–21). This immune intervention trial was specifically designed to reconstruct monocytic immunity in patients in the immune hyporeactive phase of sepsis. Thus, study patients were included after about a week of standard ICU therapy (Table 1). As sepsis-associated immunosuppression can be transient, repeated standardized mHLA-DR measurements (43) before inclusion confirmed the respective immunological state.
Treatment with GM-CSF induced an increase in the numbers of neutrophils and monocytes (Figures 2A and 2B), and a fast and effective increase of mHLA-DR (Figure 3A). After the intervention interval, mHLA-DR reached normal limits in all patients receiving GM-CSF (Figure 3A and Table 2). mHLA-DR expression in two patients was not normalized after 5 days of treatment with 4 μg/kg/day of GM-CSF. After dose escalation, as intended in the study protocol (Figure 1), the respective patients' mHLA-DR expression normalized. This might indicate a relatively low rate of partial responders to 4 μg/kg/day of GM-CSF. Normalization of mHLA-DR was associated with an increase in monocytic proinflammatory cytokine production (IL-6, TNF-α), and a decrease in monocytic antiinflammatory cytokine production (IL-10) in response to TLR2 (Pam3CSK4) and TLR4 ligands (endotoxin) (Figures 3C–3E and 4A–4D). In addition, we analyzed T-cell numbers and T-cell cytokine production after stimulation with mitogens and recall antigens. Both CD4+ and CD8+ T cells increased significantly under GM-CSF therapy (Figures 2C and 2D). However, significant differences in T-cell function were not noted between the study groups. In the future, this should be investigated in more detail, as recovery of T-cell functionality by immunostimulatory approaches may be of importance to restore adequate antimicrobial defenses in sepsis.
The clinical course differed between GM-CSF and control patients. A shorter time of mechanical ventilation was observed in the treatment group (Table 1). This is in line with findings in animal models, indicating that GM-CSF might decrease the intensity of pulmonary inflammation (53–56). In addition to time on the ventilator, intrahospital and ICU stay was shorter in patients receiving GM-CSF. This, however, did not reach statistical significance, even in the subgroup of patients with pneumonia (n = 21). Some limitations of this study demand further discussion. First, although the study groups did not differ in regard to important baseline characteristics, including source of sepsis and microbial pathogens, a tendency toward higher baseline disease severity scores was noted in control patients (Table 1). We are unable to rule out that this might have contributed to the observed clinical effects of GM-CSF. Nevertheless, IL-6, IL-10, and PCT levels, which correlate with mortality in sepsis (28, 32, 38, 57, 58), were higher in the GM-CSF group. It seems unlikely that the observed clinical effects were attributable to a slightly lower disease severity at baseline. Another limitation is the overall sample size and the fact that we did not assess other indices reflecting monocyte immunity, such as phagocytosis and antigen presentation, or clinical endpoints, such as (secondary) infection rate. This needs to be investigated in subsequent analyses. Moreover, the mortality rate of the study population seems rather low. This may be due to the fact that patients were included in the later phase of sepsis. Patients dying in the early “shock phase” of the disease were thus not assessed.
Our study differs from other sepsis trials in that we used a standardized biomarker (i.e., the mHLA-DR expression) as inclusion criteria. Monitoring of immune function was performed to assess the immunological dynamics and the course of cell-mediated immunity in our study population. To us, this seems a prerequisite for risk stratification, longitudinal follow-up, and the design and testing of immunological interventions. As of today, the course of mHLA-DR may best reflect the impact of a given immunological intervention in sepsis.
In conclusion, we demonstrate that treatment using GM-CSF is a safe and effective measure to restore mHLA-DR expression and cytokine release (as indicators of monocytic function) in patients with severe sepsis/septic shock and sepsis-associated immunosuppression. A trend toward an improved course of disease severity was observed in patients receiving GM-CSF. A larger analysis with clinical outcome measures and mortality as primary endpoints is needed.
The authors thank all nurses, medical doctors, and intensive care unit supervisors, as well as all technicians and medical doctors of the Institute of Medical Immunology for their dedicated help and support.
| 1. | Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:1303–1310. |
| 2. | Angus DC, Pereira CA, Silva E. Epidemiology of severe sepsis around the world. Endocr Metab Immune Disord Drug Targets 2006;6:207–212. |
| 3. | Annane D, Bellissant E, Cavaillon JM. Septic shock. Lancet 2005;365:63–78. |
| 4. | Annane D, Aegerter P, Jars-Guincestre MC, Guidet B. Current epidemiology of septic shock: the CUB-Rea Network. Am J Respir Crit Care Med 2003;168:165–172. |
| 5. | Russell JA. Management of sepsis. N Engl J Med 2006;355:1699–1713. |
| 6. | Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, Reinhart K, Angus DC, Brun-Buisson C, Beale R, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit Care Med 2008;36:296–327. |
| 7. | Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med 2003;348:1546–1554. |
| 8. | Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 2003;348:138–150. |
| 9. | Carlet J, Cohen J, Calandra T, Opal SM, Masur H. Sepsis: time to reconsider the concept. Crit Care Med 2008;36:964–966. |
| 10. | Pachot A, Monneret G, Brion A, Venet F, Bohe J, Bienvenu J, Mougin B, Lepape A. Messenger RNA expression of major histocompatibility complex class II genes in whole blood from septic shock patients. Crit Care Med 2005;33:31–38. Discussion, 236–237. |
| 11. | Pachot A, Lepape A, Vey S, Bienvenu J, Mougin B, Monneret G. Systemic transcriptional analysis in survivor and non-survivor septic shock patients: a preliminary study. Immunol Lett 2006;106:63–71. |
| 12. | Monneret G, Lepape A, Voirin N, Bohe J, Venet F, Debard AL, Thizy H, Bienvenu J, Gueyffier F, Vanhems P. Persisting low monocyte human leukocyte antigen-DR expression predicts mortality in septic shock. Intensive Care Med 2006;32:1175–1183. |
| 13. | Heidecke CD, Weighardt H, Hensler T, Bartels H, Holzmann B. Immune paralysis of T-lymphocytes and monocytes in postoperative abdominal sepsis: correlation of immune function with survival. Chirurg 2000;71:159–165. |
| 14. | Munford RS, Pugin J. Normal responses to injury prevent systemic inflammation and can be immunosuppressive. Am J Respir Crit Care Med 2001;163:316–321. |
| 15. | Monneret G, Venet F, Pachot A, Lepape A. Monitoring immune dysfunctions in the septic patient: a new skin for the old ceremony. Mol Med 2008;14:64–78. |
| 16. | Schefold JC, Hasper D, Volk HD, Reinke P. Sepsis: time has come to focus on the later stages. Med Hypotheses 2008;71:203–208. |
| 17. | Pugin, J. Immunostimulation is a rational therapeutic strategy in sepsis. Novartis Found Symp 2007;280:21–27. Discussion, 27–36, 160–164. |
| 18. | Wesche DE, Lomas-Neira JL, Perl M, Chung CS, Ayala A. Leukocyte apoptosis and its significance in sepsis and shock. J Leukoc Biol 2005;78:325–337. |
| 19. | Monneret G, Debard AL, Venet F, Bohe J, Hequet O, Bienvenu J, Lepape A. Marked elevation of human circulating CD4+CD25+ regulatory T cells in sepsis-induced immunoparalysis. Crit Care Med 2003;31:2068–2071. |
| 20. | Cohen J. The immunopathogenesis of sepsis. Nature 2002;420:885–891. |
| 21. | Hoesel LM, Gao H, Ward PA. New insights into cellular mechanisms during sepsis. Immunol Res 2006;34:133–141. |
| 22. | Volk HD, Reinke P, Falck P, Staffer G, von Baehr R. Prognostic parameters for the clinical outcome of septic disease in immunosuppressed patients. Clin Transplant 1989;3:246–252. |
| 23. | Volk HD, Reinke P, Docke WD. Clinical aspects: from systemic inflammation to ‘immunoparalysis’. Chem Immunol 2000;74:162–177. |
| 24. | Wolk K, Docke WD, von Baehr V, Volk HD, Sabat R. Impaired antigen presentation by human monocytes during endotoxin tolerance. Blood 2000;96:218–223. |
| 25. | Wolk K, Kunz S, Crompton NE, Volk HD, Sabat R. Multiple mechanisms of reduced major histocompatibility complex class II expression in endotoxin tolerance. J Biol Chem 2003;278:18030–18036. |
| 26. | Wolk K, Docke W, von Baehr V, Volk H, Sabat R. Comparison of monocyte functions after LPS- or IL-10–induced reorientation: importance in clinical immunoparalysis. Pathobiology 1999;67:253–256. |
| 27. | Fumeaux T, Pugin J. Role of interleukin-10 in the intracellular sequestration of human leukocyte antigen-DR in monocytes during septic shock. Am J Respir Crit Care Med 2002;166:1475–1482. |
| 28. | Marchant A, Alegre ML, Hakim A, Pierard G, Marecaux G, Friedman G, De Groote D, Kahn RJ, Vincent JL, Goldman M. Clinical and biological significance of interleukin-10 plasma levels in patients with septic shock. J Clin Immunol 1995;15:266–273. |
| 29. | Koppelman B, Neefjes JJ, de Vries JE, de Waal Malefyt R. Interleukin-10 down-regulates MHC class II alphabeta peptide complexes at the plasma membrane of monocytes by affecting arrival and recycling. Immunity 1997;7:861–871. |
| 30. | Reith, W., LeibundGut-Landmann S, Waldburger JM. Regulation of MHC class II gene expression by the class II transactivator. Nat Rev Immunol 2005;5:793–806. |
| 31. | Woiciechowsky C, Asadullah K, Nestler D, Eberhardt B, Platzer C, Schoning B, Glockner F, Lanksch WR, Volk HD, Docke WD. Sympathetic activation triggers systemic interleukin-10 release in immunodepression induced by brain injury. Nat Med 1998;4:808–813. |
| 32. | Moore KW, de Waal Malefyt R, Coffman RL, O'Garra A. Interleukin-10 and the interleukin-10 receptor. Annu Rev Immunol 2001;19:683–765. |
| 33. | Le Tulzo Y, Pangault C, Amiot L, Guilloux V, Tribut O, Arvieux C, Camus C, Fauchet R, Thomas R, Drenou B. Monocyte human leukocyte antigen-DR transcriptional downregulation by cortisol during septic shock. Am J Respir Crit Care Med 2004;169:1144–1151. |
| 34. | Schwiebert LM, Schleimer RP, Radka SF, Ono SJ. Modulation of MHC class II expression in human cells by dexamethasone. Cell Immunol 1995;165:12–19. |
| 35. | Basta PV, Moore TL, Yokota S, Ting JP. A beta-adrenergic agonist modulates DR alpha gene transcription via enhanced cAMP levels in a glioblastoma multiforme line. J Immunol 1989;142:2895–2901. |
| 36. | Fumeaux T, Pugin J. Is the measurement of monocytes HLA-DR expression useful in patients with sepsis? Intensive Care Med 2006;32:1106–1108. |
| 37. | Monneret G, Elmenkouri N, Bohe J, Debard AL, Gutowski MC, Bienvenu J, Lepape A. Analytical requirements for measuring monocytic human lymphocyte antigen DR by flow cytometry: application to the monitoring of patients with septic shock. Clin Chem 2002;48:1589–1592. |
| 38. | Hynninen M, Pettila V, Takkunen O, Orko R, Jansson SE, Kuusela P, Renkonen R, Valtonen M. Predictive value of monocyte histocompatibility leukocyte antigen-DR expression and plasma interleukin-4 and -10 levels in critically ill patients with sepsis. Shock 2003;20:1–4. |
| 39. | Lekkou A, Karakantza M, Mouzaki A, Kalfarentzos F, Gogos CA. Cytokine production and monocyte HLA-DR expression as predictors of outcome for patients with community-acquired severe infections. Clin Diagn Lab Immunol 2004;11:161–167. |
| 40. | Saenz JJ, Izura JJ, Manrique A, Sala F, Gaminde I. Early prognosis in severe sepsis via analyzing the monocyte immunophenotype. Intensive Care Med 2001;27:970–977. |
| 41. | Docke WD, Randow F, Syrbe U, Krausch D, Asadullah K, Reinke P, Volk HD, Kox W. Monocyte deactivation in septic patients: restoration by IFN-gamma treatment. Nat Med 1997;3:678–681. |
| 42. | Tschoeke SK, Moldawer LL. Human leukocyte antigen expression in sepsis: what have we learned? Crit Care Med 2005;33:236–237. |
| 43. | Docke WD, Hoflich C, Davis KA, Rottgers K, Meisel C, Kiefer P, Weber SU, Hedwig-Geissing M, Kreuzfelder E, Tschentscher P, et al. Monitoring temporary immunodepression by flow cytometric measurement of monocytic HLA-DR expression: a multicenter standardized study. Clin Chem 2005;51:2341–2347. |
| 44. | Perry SE, Mostafa SM, Wenstone R, Shenkin A, McLaughlin PJ. Is low monocyte HLA-DR expression helpful to predict outcome in severe sepsis? Intensive Care Med 2003;29:1245–1252. |
| 45. | Schefold JC, Hasper D, Reinke P, Monneret G, Volk HD. Consider delayed immunosuppression into the concept of sepsis. Crit Care Med 2008;36:3118. |
| 46. | Hamilton JA. Colony-stimulating factors in inflammation and autoimmunity. Nat Rev Immunol 2008;8:533–544. |
| 47. | Flohe S, Borgermann J, Dominguez FE, Majetschak M, Lim L, Kreuzfelder E, Obertacke U, Nast-Kolb D, Schade FU. Influence of granulocyte-macrophage colony-stimulating factor (GM-CSF) on whole blood endotoxin responsiveness following trauma, cardiopulmonary bypass, and severe sepsis. Shock 1999;12:17–24. |
| 48. | Flohe S, Lendemans S, Selbach C, Waydhas C, Ackermann M, Schade FU, Kreuzfelder E. Effect of granulocyte-macrophage colony-stimulating factor on the immune response of circulating monocytes after severe trauma. Crit Care Med 2003;31:2462–2469. |
| 49. | Nierhaus A, Montag B, Timmler N, Frings DP, Gutensohn K, Jung R, Schneider CG, Pothmann W, Brassel AK, Schulte Am Esch J. Reversal of immunoparalysis by recombinant human granulocyte-macrophage colony-stimulating factor in patients with severe sepsis. Intensive Care Med 2003;29:646–651. |
| 50. | Nakos G, Malamou-Mitsi VD, Lachana A, Karassavoglou A, Kitsiouli E, Agnandi N, Lekka ME. Immunoparalysis in patients with severe trauma and the effect of inhaled interferon-gamma. Crit Care Med 2002;30:1488–1494. |
| 51. | Schefold JC, von Haehling S, Corsepius M, Pohle C, Kruschke P, Zuckermann H, Volk HD, Reinke P. A novel selective extracorporeal intervention in sepsis: immunoadsorption of endotoxin, interleukin 6, and complement-activating product 5a. Shock 2007;28:418–425. |
| 52. | Kox WJ, Volk T, Kox SN, Volk HD. Immunomodulatory therapies in sepsis. Intensive Care Med 2000;26:S124–S128. |
| 53. | Ballinger MN, Paine R III, Serezani CH, Aronoff DM, Choi ES, Standiford TJ, Toews GB, Moore BB. Role of granulocyte macrophage colony-stimulating factor during gram-negative lung infection with Pseudomonas aeruginosa. Am J Respir Cell Mol Biol 2006;34:766–774. |
| 54. | Baleeiro CE, Christensen PJ, Morris SB, Mendez MP, Wilcoxen SE, Paine R III. GM-CSF and the impaired pulmonary innate immune response following hyperoxic stress. Am J Physiol Lung Cell Mol Physiol 2006;291:L1246–L1255. |
| 55. | Lieschke GJ, Stanley E, Grail D, Hodgson G, Sinickas V, Gall JA, Sinclair RA, Dunn AR. Mice lacking both macrophage- and granulocyte-macrophage colony-stimulating factor have macrophages and coexistent osteopetrosis and severe lung disease. Blood 1994;84:27–35. |
| 56. | Paine R III, Preston AM, Wilcoxen S, Jin H, Siu BB, Morris SB, Reed JA, Ross G, Whitsett JA, Beck JM. Granulocyte-macrophage colony-stimulating factor in the innate immune response to Pneumocystis carinii pneumonia in mice. J Immunol 2000;164:2602–2609. |
| 57. | Harbarth S, Holeckova K, Froidevaux C, Pittet D, Ricou B, Grau GE, Vadas L, Pugin J. Diagnostic value of procalcitonin, interleukin-6, and interleukin-8 in critically ill patients admitted with suspected sepsis. Am J Respir Crit Care Med 2001;164:396–402. |
| 58. | Panacek, EA, Marshall JC, Albertson TE, Johnson DH, Johnson S, MacArthur RD, Miller M, Barchuk WT, Fischkoff S, Kaul M, et al.; Monoclonal Anti-TNF: a Randomized Controlled Sepsis Study Investigators. Efficacy and safety of the monoclonal anti-tumor necrosis factor antibody F(ab′)2 fragment afelimomab in patients with severe sepsis and elevated interleukin-6 levels. Crit Care Med 2004;32:2173–2182. |
