Plasma levels of the growth arrest-specific gene 6 product (Gas6) and antiplatelet drug responsiveness in healthy subjects.

Abstract

Aspirin targets the thromboxane (Tx) A2 pathway and clopidogrel the adenosine diphosphate (ADP) pathway of platelet activation process. The biological effects of aspirin and clopidogrel vary from one subject to another and this variability has been implicated in the risk of recurrent ischemic events (1). We recently examined aspirin and clopidogrel responsiveness in 96 healthy subjects (2) and found aspirin pseudo-resistance (APR) in 30% of cases. As in the other subjects, the production of TxB2 (the stable breakdown product of TxA2) was abolished by aspirin in these pseudo-resistant subjects, but their platelet closure times (CT) determined with the PFA-100 � device (collagen/epinephrine cartridge) were normal, despite aspirin intake (100 mg day )1 ) for 7 days. This suggested that platelets continued to aggregate despite adequate inhibition of TxA2 production by aspirin, and that additional platelet amplification pathways therefore compensated for the lack of TxA2. Several studies have linked low PFA-100 � (Dade Behring, Dudingen, Switzerland) CT values to vascular events in aspirin-treated patients (3-5), suggesting that APR may be a risk factor for recurrent cardiovascular events. The importance of the growth arrest-specific gene 6 product (Gas6) in platelet activation and thrombus stabilization has been demonstrated in mice (6-8), but the expression of Gas6 and its receptors in human platelets is controversial (6,9). We and others have detected Gas6 protein with an ELISA method in human and mouse plasma (9-11). We recently showed that plasma Gas6 levels in healthy subjects do not influence platelet aggregation ex vivo in response to selected platelet agonists (11), but the role of Gas6 in the variability of biological responses to antiplatelet drugs has not been investigated. Plasma Gas6 levels were determined in the 96 healthy volunteers enrolled in the above-mentioned study (2) with an ELISA method (10) in baseline samples, and were expressed as a percentage of the Gas6 level in a normal plasma pool (11). Statistical analysis used the Kruskall-Wallis test and a logistic regression model to test the relationship between antiplatelet drug responsiveness and quartiles of Gas6 levels, after adjustment for potentially confounding variables. Plasma Gas6 levels ranged from 47.7% to 167.1% of the value obtained with the normal plasma pool (median 85%, interquartile range (IR): 68-103%). These values are compat- ible with those found in another population of healthy male volunteers (11). Subjects in the first and second quartiles of plasma Gas6 levels had median values of 64.3% and 79.2%, respectively, while subjects in the third and fourth quartiles had values of 88.9% and 115.7% (P < 0.001). Subjects with APR had a median Gas6 level of 88% (IR: 78-133%), while aspirin- sensitive subjects had a median Gas6 level of 79% (IR: 57- 132%, P ¼ 0.033). The proportion of APR subjects increased from 8.0% in quartile 1 of plasma Gas6 levels to 21.7%, 45.8% and 43.4% in quartiles 2-4, respectively (P ¼ 0.009). Table 1 shows the results of univariate and multivariate logistic regression analysis testing the association of APR and quartiles