aPL are a heterogeneous family of autoantibodies reacting against proteins with affinity for negatively charged phospholipids (PL). In particular, beta-2 glycoprotein I (β2GPI) provides, together with prothrombin, the most important epitope targeted by aPL. Three configurations of β2GPI have been described: circulating plasma β2GPI exists in a circular form; upon binding to suitable anionic surfaces as cardiolipin (CL) and other PL or to LPS, the molecule opens into a J-shaped fishhook configuration; recently an intermediate S shape of β2GPI has been observed. β2GPI consists of 5 domains (D): DI-IV comprise 60 amino acids and contain two disulfide bridges each, while DV is aberrant, as it includes 82 amino acids due to a 6-residue insertion and a 19-residue C-terminal extension cross-linked by an additional disulfide bond. DI has been identified as the most relevant antigenic target involved in β2GPI/anti-β2GPI antibody binding. This epitope is a cryptic and conformation-dependent structure: in the circular conformation of β2GPI, DI interacts with DV and the critical epitope is thus hidden. When β2GPI adopts the S shape, the epitope is covered by DIII–IV carbohydrate chains, thus preventing antibodies from binding β2GPI. Upon opening to a J configuration, the critical epitope is exposed, becoming available for antibody binding. Several factors might lead to the surface exposition of the critical epitope, such as oxidative stress. Indeed, under oxidative conditions, disulfide bonds form in the molecule leading to the unmasking of the critical B-cell structure [6].

aPL exert a thrombogenic effect by interfering with both soluble components and cells involved in the coagulation cascade [5]. aPL promote aggregation and activation of platelets, neutralizing β2GPI interaction with von Willebrand factor and enhancing the expression of platelet membrane glycoprotein (GP) IIb/IIIa. Moreover, aPL induce a proinflammatory and procoagulant endothelial phenotype upregulating cellular adhesion molecules, promoting the synthesis of endothelial nitric oxide synthase (eNOS) and of proinflammatory cytokines as IL-6 and TNF-α. Lastly, aPL have been shown to significantly increase in both ECs and monocytes the expression of TF, the major initiator of the clotting cascade. Many molecules have been advocated as potential mediators of β2GPI interaction with target cells: annexin A2, toll-like receptor (TLR) 2 and 4, heparan sulfate, and ApoER2’. There is general agreement that NFκB and p38MAPK are involved in the downstream signaling pathways engaged by aPL [7].

aPL exert a thrombophilic effect also at a placental level, inducing intraplacental thrombosis and infarction, leading to an impairment of the maternal-fetal blood exchange. aPL were also shown to contribute to the prothrombotic state by activating complement and disrupting the anticoagulant annexin A5 shield on trophoblast and endothelial cell (EC) monolayers [7].

Hereby, cutting-the-edge therapeutic strategies are widely discussed, presenting the pioneer reports about the efficacy of novel pharmacological agents in APS based on the recent advances in understanding of aPL pathogenic mechanisms.

Fondaparinux and idraparinux are synthetic pentasaccharides homologous to heparin-binding site; their activity is limited on factor Xa. Both administered subcutaneously, fondaparinux is given daily, whereas idraparinux has a longer half-life, which allows a weekly administration. Given the low affinity to platelet factor 4, fondaparinux and idraparinux can be safely prescribed to patients with a history of heparin-induced thrombocytopenia. Fondaparinux has been licensed for VTE prophylaxis; idraparinux has not yet been patented: even though it was as effective as warfarin in preventing recurrence, it was associated with a significantly higher bleeding rate. Neither fondaparinux nor idraparinux has been yet evaluated in the setting of APS [8].

Most recently, a novel class of anticoagulants has been synthesized: all administered orally, such pharmacological agents inhibit a single enzyme of the coagulation cascade, being thus called direct oral anticoagulants (DOA). Dabigatran is a potent, competitive, reversible direct thrombin inhibitor, which binds to thrombin and blocks its interaction with substrates. Direct FXa inhibitors include rivaroxaban, apixaban, and edoxaban. All these agents are highly selective, reversible, competitive, and dose dependent. They represent an advance over VKA mainly in terms of a better quality of life for patients: since they display a predictable anticoagulant effect, DOA are administered at a fixed dose. In addition, not being metabolized by the cytochrome P450 system, they do not interact with dietary constituents or alcohol and have few reported drug interactions, therefore not requiring routine monitoring of anticoagulant intensity. However, these novel DOA do not allow overcoming some other limitations affecting treatment with VKA. The main issue lies in the significant bleeding risk that any anticoagulant regimen carries and in the absence of an available pharmacological reversal agent [9].

The role of these emerging anticoagulants in APS management is still to be determined: there are three ongoing randomized controlled clinical trials evaluating rivaroxaban in the management of APS, as compared to low-intensity anticoagulation. The RAPS trial has been promoted by a UK group; it is a phase II/III study that has recruited 156 APS patients with a history of venous thromboembolism. A Spanish phase III trial has been started in Spain on 218 patients with venous or arterial events [10]. Most recently, an Italian trial considering triple positive APS patients only is going to start recruiting.

HCQ is an antimalarial drug with anti-inflammatory and antithrombotic properties. In addition, HCQ has been shown to exert immunomodulatory effects: it prevents activation of TLR3, TLR7, and TLR9; inhibits antigen processing and presentation; and reduces circulating immune complexes [11, 12]. In in vitro models of thrombotic APS, HCQ has been demonstrated to inhibit GPIIb/IIIa expression on aPL-activated platelets [13], to reverse the formation of aPL-β2GPI-PL bilayer complexes [14], and to prevent the aPL-induced disruption of the annexin A5 shield [15]. Its antithrombogenic properties have been confirmed in in vivo models of APS: HCQ injection in mice induced a dose dependent decrease in thrombus size [16]. HCQ might be effective even in obstetric APS, as it was shown to reverse the binding of aPL to human placental syncytiotrophoblasts [17] and the aPL-inhibition of trophoblast IL-6 secretion [18].

Clinical data on the effectiveness of HCQ in preventing aPL-related thrombotic events have been derived from studies in SLE cohorts. Recently, a cross-sectional study on 77 APS patients and 56 asymptomatic aPL carriers from an SLE registry proved that the probability of a thrombotic event was decreased by LDASA or HCQ use [19]. In a still unpublished work, Petri and coworkers observed a decrease in arterial as well as venous thrombosis in aPL-positive lupus patients receiving HCQ [20].

In primary thrombotic APS, HCQ has been to date evaluated as an adjunctive pharmacological tool: patients receiving a combo regimen comprising HCQ plus oral anticoagulation experienced less recurrences compared to those on anticoagulants only. However, the extrapolation of data is affected by the limitations biasing this work: the study cohort was limited to 40 patients, and the follow-up lasted 36 months only [21]. HCQ is currently catalyzing much attention in APS: an ongoing study is assessing the effect of HCQ on annexin A5 resistance assay in aPL patients with or without SLE; a randomized controlled trial promoted by the international research organization APS ACTION is evaluating HCQ in the primary prevention of thrombosis in aPL asymptomatic carriers at 5 years of follow-up [22].

Even though there is limited clinical evidence of its antithrombotic effects in primary APS, treatment guidelines consider HCQ as a potential adjunctive therapy, particularly in consideration of its excellent safety profile [2].

Statins inhibit cholesterol synthesis in the mevalonate pathway by blocking the 3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase. The use of statins in the treatment of APS might thus be beneficial in the prevention of thrombosis since hypertriglyceridemia and low HDL cholesterol levels provide the most frequent cardiovascular risk factor reported in APS patients [23]. However, these pharmacological compounds have also been shown to exert a wide array of additional pleiotropic antithrombotic and anti-inflammatory effects in APS. In vitro, fluvastatin, simvastatin, and rosuvastatin have been demonstrated to inhibit TF synthesis in EC [24]; fluvastatin and simvastatin were both reported to suppress anti-β2GPI antibody-induced endothelial adhesiveness and to reduce monocyte adhesion to the endothelium [25], while rosuvastatin inhibited the upregulation of VCAM induced by aPL [26]. In vivo, fluvastatin reduced the size of the thrombus induced by aPL infusion and the leukocyte adhesion to EC. In models of obstetric APS, pravastatin did not prevent aPL-mediated functional effects on human trophoblasts in vitro [27], while in vivo simvastatin and pravastatin reduced fetal death [28]. These findings have been later confirmed in ex vivo studies. A trial in 42 patients with APS treated with fluvastatin for 30 days reported a decline in several thrombogenic and inflammatory mediators in monocytes [29]. More recently, Erkan observed a significant reduction of half of the evaluated proinflammatory and procoagulant parameters (IL-1β, VEGF, TNF-α, IP10, CD40L, and TF) in a cohort of 41 aPL asymptomatic carriers after 3 months of treatment with fluvastatin [30].

As a whole, statins could have a role a potential treatment for APS, beyond their lipid-lowering activity with no intrinsic additional risk of bleeding [31].

In vitro, vitamin D exerts an antithrombotic and immunomodulator function by inhibiting β2GPI-mediated TF expression [32]. Retrospective studies indicated that the prevalence of vitamin D deficiency in APS patients ranges from 10 to 50 %, while insufficiency may occur in up to 70 % of patients [32–36]. Low vitamin D levels correlate with arterial and venous thrombosis as well as with non-criteria APS manifestations [32–34]. Conversely, most studies supported no association between low vitamin D levels and obstetric APS [32–35]. Prospective studies are warranted to evaluate the therapeutic value of supplementation with vitamin D in aPL carriers.

In vivo and in vitro models suggest the therapeutic potential of intravenous immunoglobulins (IVIg) in APS. IVIg were shown to inhibit aPL, by partially neutralizing LA activity and preventing aCL binding to cardiolipin by Fab. IVIg exert an anti-idiotype activity, with inactivation of idiotype-bearing B-cell clones. Furthermore, IVIg have been demonstrated to increase IgG catabolism, to modulate complement activation, to block Fcγ receptor on macrophages, and to downregulate proinflammatory cytokines [37]. Treatment with IVIg resulted in an inhibition of aPL thrombogenic effects, with a reduction of circulating aCL levels [38]. There are few reports about successful treatment with IVIg in the management of aPL-related clinical manifestations, mainly hematogical (thrombocytopenia, hemolytic anemia, hypoprothrombinemia). Treatment response was observed in all cases but one [39]. Recently, IVIg were found to be effective in preventing recurrent thrombosis among five patients with refractory APS in a 5-year open study [40].

The pivotal role exerted by B cells in APS has been progressively deciphered: B lymphocytes contribute to APS etiopathogenesis by producing autoantibodies, inducing the formation of germinal centers and the synthesis of cytokines. Accordingly, in NZW x BXSB mice, treatment with IgG against B-cell activating factor (BAFF) receptor did not prevent the development of aCL even though it prevented aPL-related thrombotic vasculopathy, prolonging survival [41]. In the same murine models, IgG against cytotoxic T-lymphocyte antigen 4 immunoglobulin (CTLA4) affected initiation but not development of APS. These data suggest the potential efficacy of belimumab and abatacept in APS [42]. Interestingly, in belimumab-treated patients with systemic lupus erythematosus (SLE) a positive-to-negative conversion rate was reported for aCL [43]. To date, clinical experience of B-cell inhibitory agents specifically in APS patients are restricted to the use of rituximab, a chimeric monoclonal antibody targeting CD20 on the surface of B cells. Successful treatment with rituximab has been reported in anecdotal reports and in one case series from the BIOGEAS registry. In this multicenter Spanish registry, a therapeutic response was observed in 92 % of 12 cases [44]. In 2012, a review collected all the published cases, identifying 27 APS patients treated with rituximab [45]. The anti-CD20 monoclonal resulted in a decrease of aPL titers; among those receiving rituximab because of thrombotic recurrences, clinical improvement was observed in all cases. Moreover, rituximab was beneficial for a plethora of aPL-related clinical manifestations. At this regard, an open-label phase IIa descriptive pilot study (RITAPS) has been carried out in 20 patients with non-criteria APS manifestations refractory to conventional treatments [46]. Rituximab resulted to be effective in controlling some but not all non-criteria manifestations, without substantial change in aPL profile. Notably, caution should be paid to its use in APS: episodes of severe acute thrombotic exacerbations (lacunar infarctions and transverse myelitis) have been reported in two APS/SLE patients receiving rituximab [47].

Rituximab has been used in 20 cases of catastrophic APS (CAPS), in different combination with anticoagulation, high doses of steroids, plasma exchange, and IVIg. Despite the difficulties in determining the effects of rituximab, a lower mortality compared to larger series emerged [48].

The complement system, in particular the mediator C5a, has been shown to play a central role in APS. C5a, a potent anaphylatoxic, proinflammatory, and chemotactic molecule, was demonstrated to induce the expression of TF on EC [49] and neutrophils [50]. In vivo, C5a was involved in deposition of fibrin in a growing thrombus induced by aPL injection [51]. C5a-induced recruitment and activation of neutrophils lead to trophoblast injury and angiogenic factor imbalance in aPL-induced fetal injury [52]. Eculizumab is a humanized monoclonal antibody which binds to the C5 protein with high affinity, thereby inhibiting its cleavage to C5a and C5b and preventing the generation of membrane attack complex [53]. In vivo, anti-C5 antibody was shown to attenuate thrombosis and to prevent aPL-mediated pregnancy loss [49]. To date, eculizumab has been administered to few CAPS patients in which all the other therapeutic strategies proved to be ineffective. A favorable response was described in two cases [54, 55] and a negative outcome in the others [22]. Eculizumab has also been investigated as a tool to manage APS patients after renal transplantation. In a first report on three consecutive kidney transplant recipients with posttransplant aPL-mediated thrombotic microangiopathy (TMA) resistant to plasmapheresis, treatment with eculizumab improved TMA [56]. In another case series of three patients treated with anticoagulation and eculizumab, no systemic thrombotic events or early graft losses were reported after a follow-up ranging from 4 months to 4 years [57].

Defibrotide is a polydisperse mixture of 90 % single-stranded and 10 % double-stranded phosphodiester oligonucleotides derived from the controlled depolymerization of porcine intestinal mucosal DNA. This pharmacological compound acts by upregulating the release of prostacyclin and prostaglandin E2, reducing concentrations of leukotriene B4, inhibiting monocyte superoxide anion generation, stimulating expression of thrombomodulin in human vascular EC, and modulating platelet activity [58]. More recently, defibrotide was shown to downregulate TF expression on monocytes [59]. To date, it has been used in two patients with CAPS: in one case this treatment was successful, while the second patient died [60, 61].

TIFI is a 20 amino acid synthetic peptide that spans Thr101–Thr120 of ULB0-HCMVA from human cytomegalovirus, which shares similarities with the PL-binding site in β2GPI molecule, DV. TIFI is not targeted by aPL; in vitro evidence suggest that TIFI inhibits the binding of labeled β2GPI to human EC and mouse monocytes [62]. Similarly, the peptide prevented anti-β2GPI reactivity toward human trophoblast monolayers. These findings were also confirmed in animal models: the infusion of this synthetic peptide inhibited aPL-mediated thrombosis by decreasing the thrombus size produced in response to aPL and by reducing the binding of fluoresceinated β2GPI to EC. Pregnant naïve mice treated with TIFI were protected from fetal loss induced by human aPL IgG [63].

Accordingly, a synthetic β2GPI-DI was shown to inhibit aPL-mediated prothrombic effects both in vivo and in vitro [64].

MBB2 is a novel single-chain variable fragment (scFv)-Fc monoclonal antibody targeting DI of human, rat, and mouse β2GPI. When infused to experimental animals, MBB2 caused blood clots in rat mesenteric microcirculation after LPS priming. A non-complement-fixing variant of MBB2, MBB2ΔCH2, has also been developed. MBB2ΔCH2 displays the same antigen specificity of MBB2 but, lacking the CH2 domain, is unable to activate the complement cascade. MBB2ΔCH2 has been shown to prevent the aPL procoagulant effects in vivo by competing with circulating aPL for binding to β2GPI. In vivo, the CH2-deleted monoclonal antibody significantly reduced mesenteric thrombus formation and vessel occlusion [65].

Such antibody can be used in the future even in the management of obstetric APS, as MBB2ΔCH2 leaves bound β2GPI untouched. This is important because cell-bound β2GPI exerts a key role at the placental level, as supported by the defective embryo implantation and placental morphogenesis displayed by β2GPI^−/− mice. Hence, any tool able to block the pathogenic effects of aPL leaving untouched β2GPI represents a useful approach.

It can also be postulated that antagonists of the receptors involved in β2GPI cell binding may exert a therapeutic potential [66]. The use of antagonists or neutralizing monoclonal antibodies acting on TLR2/4 might be speculated in APS. In addition, DV of β2GPI binds the A1 ligand-binding type A module of ApoER2’; a dimer composed of two A1 molecules joined by a flexible linker has been shown to inhibit anti-β2GPI antibody/dimerized β2GPI immune complexes from binding negatively charged PL and ApoER2’ in vitro, more potently than A1 in the monomeric form [67]. More recently, proofs of the effectiveness of this dimeric molecule were obtained in vivo, in two animal models of APS. Indeed, treatment with A1-A1 efficiently reduced thrombus size in vivo in the presence of chronic autoimmune anti-β2GPI antibody in lupus-prone (NZW3BXSB)F1 male mice as well as in wild-type mice after infusion with anti-β2GPI antibodies [68].

Similarly, blockers of the intracellular mediators involved in aPL-activated signaling pathways may reverse the prothrombotic phenotype: NFkB and p38MAPK inhibitors have been shown to be effective in preventing aPL-mediated prothrombotic and proinflammatory effects in vitro [69]. More recently, the NFkB inhibitor DHMEQ was demonstrated to ameliorate the prothrombotic state induced by the infusion of the monoclonal antibody WB-6 in normal BALB/c mice [70].

It is can be speculated that TF inhibition may prevent thrombosis in APS [71]. Currently, there are few drugs available on the market blocking TF expression: ACE inhibitors, dilazep, defibrotide, and dipyridamole. In particular, both dilazep and dipyridamole have been shown to block the upregulation of TF specifically induced by polyclonal IgG purified from APS patients in monocytes [72, 73]. However, their role in APS management has been scarcely documented. Dual antiplatelet treatment (different combination of LDASA, ticlopidine, clopidogrel, cilostazol) has been recently proposed: a Japanese study on 82 APS patients with refractory arterial events documented no recurrences among those subjects receiving two antiplatelet agents [74].