The factor VII-activating protease (FSAP), a newly discovered serine-protease present in human plasma, has 2 main functions in hemostasis: it is a potent activator of prourokinase1 and accelerates coagulation by activating factor VII, independently of tissue factor.2 The FSAP Marburg I polymorphism (1601G>A) was recently evaluated as a candidate risk factor for venous thromboembolism (VTE), since it impairs the capacity of FSAP to activate prourokinase while preserving its capacity to activate factor VII.3
Hoppe et al4 first reported the Marburg I polymorphism to be associated with an increased risk of VTE,2 but these data were not confirmed by Van Minkelen et al.5 Since such different findings were possibly related to the different selection of the control group (formed by blood donors in the former study and by unselected healthy subjects in the latter), Hoppe et al6 reviewed their data using a different control group of non-blood donors and confirmed the association between the Marburg I polymorphism and VTE only for patients with idiopathic events (ie, those occurring in the absence of triggering factors [OR = 2.7; 95% CI, 1.2-6.1]).4
To further investigate the role of the Marburg I polymorphism as a risk factor for VTE, we carried out a large case control study of 418 patients (161 men and 257 women) who had a first, objectively confirmed VTE and were referred to our Thrombosis Center for a thrombophilia screening, and 422 healthy controls (173 men and 249 women) who were partners or friends of the whole population of thrombosis patients seen at the center. The Marburg I polymorphism was evaluated by amplification refractory mutation system (primers and conditions available on request). The median age at VTE for patients and at blood sampling for controls was 39 years (range, 14-76 years) and 42 years (range, 16-84 years), respectively. We confirmed a statistically significant association between VTE and such established risk factors as factor V Leiden, prothrombin 20210G>A, antithrombin, protein C and protein S deficiencies, hyperhomocysteinemia, and oral contraceptive use (data not shown).
Table 1 shows the prevalence of the Marburg I polymorphism, which was very similar in patients and controls, either considering all VTEs (5.3% vs 5.2%; OR = 1.0; 95% CI, 0.51-1.9) or only the 190 idiopathic events (4.4% vs 5.2%; OR = 1.1; 95% CI, 0.5-2.7). All carriers of FSAP Marburg I were heterozygous for the variant.
In conclusion, the present study ruled out a strong association between the FSAP 1601G>A polymorphism (Marburg I) and unselected or idiopathic VTE. Whether the Marburg I polymorphism determines a weak effect on thrombotic risk could be observed only in very large study populations, and remains to be determined.
