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).

SummaryTolerance of monocytes/macrophages to endotoxin (lipopolysaccharide [LPS]) can be induced both in vivo and in vitro by LPS itself. Exposure to LPS, even at a very low dose, induces a downregulation of cytokine response to a second high dose LPS challenge. To learn more about the unknown mechanisms of this phenomenon, we studied the role of antiinflammatory cytokines in this process. Preculture of human peripheral blood monocytes for 24 hours with low concentrations of LPS induced hyporesponsiveness to high-dose LPS rechaUenge with respect to tumor necrosis factor (TNF) c~ and interleukin (IL) 10 but not IL-1RA production. These results suggest that LPS tolerance reflects a functional switch of monocytes rather than a general LPS hyporesponsiveness. IblO and transforming growth factor (TGF) 31 showed additive effects in replacing LPS for induction of LPS hyporesponsiveness in vitro. Additionally, neutralizing anti-IL-10 and anti-TGF-~ monoclonal antibodies prevented induction of LPS tolerance. In vitro induced LPS tolerance looks like the ex vivo LPS hyporesponsiveness of monocytes from septic patients with fatal outcome: downregulation of LPS-induced TNF-oe and IL-10 production but not of IL-1RA secretion. LPS hyporesponsiveness in septic patients was preceded by expression of 11.-10 at both the mRNA and protein level. In summary, our data suggests that IL-10 and TGF-~ mediate the phenomenon of LPS tolerance in vitro and perhaps in vivo (septic patients), too.M onocytes and macrophages are highly sensitive for LPS, reacting on encounter with even traces by synthesis of inflammatory mediators. Among them, a network of proinflammatory cytokines is of special importance for the coordinated activation of the immune system in response to microbial invasion. When local infection develops into sepsis, a strong systemic reaction occurs, and cytokines normally restricted to areas of local injury can be detected systemically causing symptoms of septic shock (1). Although whole-body inflammation is frequently initiated by LPS, a single sublethal injection of LPS also renders the host temporarily refractory to subsequent LPS challenge as well as to other inflammatory stimuli. This phenomenon, referred to as endotoxin tolerance may be at least partially due to a reduced capacity of monocytes/macrophages to synthesize cytokines upon reexposure to LPS (2). The phenomenon of in vivo LPS desensitization has been reproduced in vitro using murine or human monocytes/macrophages (3-5). The mechanisms of LPS tolerance are not clear so far. Modulation of LPS-binding sites (CD14), secretion of inhibitory mediators by desensitized monocytes, and an altered intracellular signaling have been discussed (3,(6)(7)(8).Antiinflammatory mediators such as I1.-10, TGF-B, or IL-1RA which are produced also by monocytes after LPS contact but with delayed kinetics, are thought to be important for downregulating the inflammatory cascade, and they are known to be protective in animal models of sepsis. We wondered whether (a) the production of an...

Inflammatory cells, in particular monocytes/macrophages, release pro-inflammatory mediators in response to several infectious and non-infectious stimuli. The excessive release of these mediators, resulting in the development of whole body inflammation, may play an important role in the pathogenesis of sepsis and septic shock. TNF-alpha, acting synergistically with cytokines such as IL-1, GM-CSF and IFN-gamma, is the key mediator in the induction process of septic shock, as shown in several experimental models. Based on this concept and on the encouraging results obtained in several experimental models, a number of clinical sepsis trials targeting the production or action of TNF-alpha or IL-1 have been performed in recent years. Unfortunately, these trials have failed to demonstrate a therapeutic benefit. One reason for this may be the lack of exact immunologic analyses during the course of septic disease. Recently, we demonstrated that there is a biphasic immunologic response in sepsis: an initial hyperinflammatory phase is followed by a hypo-inflammatory one. The latter is associated with immunodeficiency which is characterized by monocytic deactivation, which we have called "immunoparalysis". While anti-inflammatory therapy (e.g. anti-TNF antibodies, IL-1 receptor antagonist, IL-10) makes sense during the initial hyperinflammatory phase, immune stimulation by removing inhibitory factors (plasmapheresis) or the administration of monocyte activating cytokines (IFN-gamma, GM-CSF) may be more useful during "immunoparalysis".