Platelet P2Y12 Inhibitors Reduce Systemic Inflammation and Its Prothrombotic Effects in an Experimental Human Model

Abstract

Objective: Clinical studies suggest that platelet P2Y12 inhibitors reduce mortality from sepsis, although the underlying mechanisms have not been clearly defined in vivo. We hypothesized that P2Y12 inhibitors may improve survival from sepsis by suppressing systemic inflammation and its prothrombotic effects. We therefore determined whether clopidogrel and the novel, more potent P2Y12 inhibitor, ticagrelor, modify these responses in an experimental human model.

Approach and results: We randomized 30 healthy volunteers to ticagrelor (n=10), clopidogrel (n=10), or no antiplatelet medication (controls; n=10). We examined the effect of P2Y12 inhibition on systemic inflammation, which was induced by intravenous injection of Escherichia coli endotoxin. Both P2Y12 inhibitors significantly reduced platelet-monocyte aggregate formation and peak levels of major proinflammatory cytokines, including tumor necrosis factor α, interleukin-6, and chemokine (C-C motif) ligand 2. In contrast to clopidogrel, ticagrelor also significantly reduced peak levels of IL-8 and growth colony-stimulating factor and increased peak levels of the anti-inflammatory cytokine IL-10. In addition, ticagrelor altered leukocyte trafficking. Both P2Y12 inhibitors suppressed D-dimer generation and scanning electron microscopy revealed that ticagrelor also suppressed prothrombotic changes in fibrin clot ultrastructure.

Conclusions: Potent inhibition of multiple inflammatory and prothrombotic mechanisms by P2Y12 inhibitors demonstrates critical importance of platelets as central orchestrators of systemic inflammation induced by bacterial endotoxin. This provides novel mechanistic insight into the lower mortality associated with P2Y12 inhibitors in patients with sepsis in clinical studies.

Trial registration: ClinicalTrials.gov NCT01846559.

Keywords: antiplatelet agents; clopidogrel; fibrin; infection; sepsis; thrombosis; ticagrelor.

© 2015 American Heart Association, Inc.

Figures

Figure 1. Levels of pro-inflammatory cytokines TNFα (A), IL-6 (B), CCL2 (C), G-CSF (D), IL-8 (E), IL-10 (F) and hsCRP (G) before and after 1 week of antiplatelet treatment and following LPS administration (t = 0 hours)

Data expressed as mean ± SEM (n=10 in each group). The overall effect of LPS and the effect of ticagrelor and clopidogrel (both compared to control at each time point) determined using 2-way ANOVA with Dunnett’s correction for multiple comparisons for the cytokines (*p

Figure 2. Platelet-monocyte (A) and platelet-neutrophil (B) aggregate formation and platelet P-selectin expression (C) at baseline, immediately before LPS administration and 6 hours after LPS administration, in unstimulated samples and samples stimulated by 30 μM ADP ex vivo

Data expressed as mean ± SEM (n=10 in each group). The overall effect of LPS and the effect of ticagrelor and clopidogrel (both compared to control at each time point) were determined using 2-way ANOVA with Dunnett’s correction for multiple comparisons (*p

Figure 3. Leukocyte (A), neutrophil (B), monocyte (C) and platelet (D) counts before and after 1 week of antiplatelet treatment and following LPS administration (t = 0 hours)

Data expressed as mean ± SEM (n=10 in each group). The overall effect of LPS and the effect of ticagrelor and clopidogrel (both compared to control at each time point) were determined using 2-way ANOVA with Dunnett’s correction for multiple comparisons (*p

Figure 4. Levels of D-dimer (A) and fibrin clot maximum absorbance (B) (a measure of clot density) and lysis area under the curve (C) (a complex measure that assesses both clot formation and lysis) determined by turbidimetry (expressed as percentage change from baseline value) following treatment and LPS administration

Data expressed as mean ± SEM (n=10 in each group). The overall effect of LPS and the effect of ticagrelor and clopidogrel (both compared to control at each time point) were determined using 2-way ANOVA with Dunnett’s correction for multiple comparisons (*p

Figure 5. Representative electron microscope images of fibrin clots formed from plasma ex vivo in each treatment group immediately before and 6 hours after LPS administration

Clots were prepared in duplicate and 4 photographs were taken of each clot at each time point. In the control group, there is an increase in fibrin network density following LPS whereas in the clopidogrel and ticagrelor groups this is not apparent.

Figure 6. Fibrin fiber diameter (A) and fibrin network density (B) (determined by electron microscopy) following LPS administration

Data expressed as mean ± SEM (n=8 in each group). The overall effect of LPS and the effect of ticagrelor and clopidogrel (both compared to control at each time point) were determined using 2-way ANOVA with Dunnett’s correction for multiple comparisons (*p