In this issue of Blood, Skjeflo et al demonstrate an association between plasma complement factor 5 (C5) levels and risk of future venous thromboembolism (VTE) in a nested case-control study of patients enrolled in the Tromsø population-based cohort study.1
Resembling the coagulation cascade in organization, the complement system is a key component of innate immunity. Experimental studies have demonstrated complex interplay between the complement system and the coagulation cascade promoting VTE. As with the coagulation cascade, there are 3 pathways, which can activate independently or converge to cleave C5. Cleavage of C5 results in release of a potent anaphylatoxin, C5a, as well as C5b to initiate formation of C5b-C9, the membrane attack complex (MAC). Cleavage of C5 potentiates hemostasis via the actions of both C5a and the MAC. Neutrophil recruitment and priming, through increased expression of tissue factor on the neutrophils, occurs in response to C5a.2,3 This recruitment and priming can lead to release of neutrophil extracellular traps, which further promote thrombosis through platelet aggregation and activation as well as thrombin formation.4,5 The MAC contributes to the activation of the coagulation cascade through platelet activation leading to the release of platelet factor V and assembly of the prothrombinase complex.6
In a prior study, complement factor 3 (C3) levels were associated with an increased risk of future VTE.7 As with the study by Skjeflo et al, the association of C3 with VTE remained with only slight modification of the risk estimates after adjustment for general inflammation, as measured by C-reactive protein, and after adjustment for body mass index. Cleavage of C3 to C3b generates C5 convertase, resulting in cleavage of C5 to C5a and C5b. Thus, higher levels of C3 result in higher levels of C5a and the MAC. In the current study, the association between C5 and risk of VTE remained, with only slight modification of the risk estimate, after similarly adjusting for C-reactive protein and body mass index. Accordingly, the agreement in findings between these 2 studies support the noted associations between complement and the development of VTE.
An additional finding of importance of this study relates to determination of drivers of plasma C5 levels. Although the acute phase reactant C-reactive protein can activate complement, the magnitude of increase in complement proteins (ie, C3) during inflammatory stimuli (eg, infection) is less pronounced. In fact, plasma C5 levels did not differ from baseline in 1 preclinical study, but were elevated at the site of inflammation.8 In this study, the authors found only a weak linear relationship between C-reactive protein and C5 levels, indicating that chronic inflammation explains only a minor part of the association between C5 and VTE. In addition, using a protein quantitative trait loci analysis, the authors found no significant SNP in either whole genome or cis-restricted analyses of 1 033 970 variants in the whole-exome data set. Although experimental studies suggest a role for C5 in hemostasis, Skjeflo et al demonstrate C5 as a risk factor for VTE within the general population. This transition from bench to bedside not only strengthens the evidence for the role of complement in hemostasis, but also provides rationale for future studies confirming this association as well as those considering novel therapeutic interventions. Inhibition of C5 in patients with paroxysmal nocturnal hemoglobinuria improves outcomes and decreases the risk of clinical thromboembolism (relative reduction 85%).9,10 The findings in the study by Skjeflo et al raise the question of whether there is a subgroup of VTE patients that could similarly benefit from complement inhibition.
In summary, Skjeflo et al demonstrate important evidence toward the role of complement in hemostasis, particularly the risk of VTE. These findings contributing to this growing body of literature and provide rationale for continued scientific investigation with the goal of improving outcomes in patients with and at risk of VTE.
Conflict-of-interest disclosure: Research funds from AstraZeneca and Astellas Pharma Global paid to the institution not related to the scope of this project; research funds from National Institutes of Health, National Heart, Lung, and Blood Institute (NHLBI; 1K01HL136893-01), American Cancer Society IRG-18-158-61-04, and National Institutes of Health, NHLBI Loan Repayment Program not related to the scope of this project; expert review for Covington & Burling Loan Repayment Program and Luther & Associates not related to the scope of this project.