Abstract 714

Introduction:

Hemophagocytic lymphohistiocytosis (HLH) is a rare multisystem hyperinflammatory syndrome associated with immune dysregulation due to genetic or acquired defects in cytotoxic natural killer cell and CD8 T cell function. Although the mechanism of activity is currently unknown, etoposide has been the mainstay of HLH therapy during the past two decades, since its initial empiric use. Jordan, et al (Blood 2004;104(3):735-43) demonstrated that the clinical and laboratory manifestations of HLH are recapitulated in perforin-deficient (prf−/−) mice infected with lymphocytic choriomeningitis virus (LCMV). In this animal model, ineffective cytotoxic T effector cells fail to down-modulate stimulatory signals provided by antigen presenting cells (APCs). Excessive T cell stimulation leads to massive cytokine production which drives systemic macrophage activation resulting in HLH-like disease pathology. Thus, at least three critical events occur in the pathogenesis of HLH: 1) Abnormal increase in antigen presentation, 2) Abnormal increase in CD8 T cell activation and cytokine secretion, and 3) Pathological macrophage activation and hemophagocytosis. We hypothesize that our unique animal model of HLH can be used to test the activity and mechanisms of action of current and novel therapeutic approaches for this disease.

Methods:

Five days after perforin-deficient (prf−/−) or wild-type (WT) mice were infected with LCMV, they were given intraperitoneal (IP) injections of either: etoposide (VP16), dexamethasone (DEX), other chemotherapeutic agents, or irrelevant carrier controls. Outcome measures included: serial measurements of disease severity using a clinical scoring system, post-infection survival, serial measurements of serum interferon-gamma (IFNγ), determination of hemoglobin levels 15 days post-infection, and flow cytometric analyses of whole spleen 8 days post-infection to assess total cellularity, T cell activation, and macrophage infiltration. Effectiveness of antigen presentation by APCs from LCMV-infected drug-treated mice was assessed by ex vivo IFNγ production using effector T cells previously generated in vitro. Osmotic pumps were used to deliver exogenous IFNγ to WT mice to assess the effect of drug therapy on IFNγ-mediated macrophage activation in the absence of viral infection.

Results:

VP16 was an effective single agent, whereas DEX was not, in our murine model of HLH leading to significant improvements in: disease severity (p<0.03 after day 22 post-infection, VP16 v. control), survival (p<0.02), peak serum IFNγ levels (p<0.009), and hemoglobin levels (p<0.002). Flow cytometric analyses of whole spleen from animals treated with VP16 showed decreases in total cellularity (p<0.0002, VP16 v. control) as well as absolute numbers of CD8 T cells (p<0.002), virus-specific CD8 T cells (p<0.002), and macrophages (p<0.0007). The decrease in T cell numbers was not caused by a direct effect of these drugs on antigen presentation by APCs, and there was no effect of drug treatment on IFNγ-mediated macrophage dependent pathology in the absence of viral infection.

Conclusions:

We present data using an animal model to test treatments for HLH by examining their effects on different aspects of HLH pathology. Our studies indicate that VP16 acts primarily via cytolytic effects on dividing T cells. This leads to a diminished pool of activated but ineffective responding T cells and attenuation of hypercytokinemia. Normalization of peak serum IFNγ levels results in decreased tissue infiltration of activated macrophages with less hemophagocytosis. Thus, improvement in HLH-like disease severity and survivability after treatment with VP16 is a direct effect of deleting activated IFNγ-producing T cells. Furthermore, this data serves to validate the use of this murine model of HLH pathogenesis to define the mechanism(s) of current and novel anti-HLH therapeutic agents. We envision using this model in the future to design rational anti-HLH therapy by combining cytolytic, immunosuppressive, and/or selective biological agents that have complementary modes of action.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution