Antiphospholipid syndrome is an autoimmune disorder characterized by vascular thrombosis and/or pregnancy morbidity associated with persistent antiphospholipid antibody positivity. Cases fulfilling the Sydney criteria for obstetric morbidity with no previous thrombosis are known as obstetric antiphospholipid syndrome (OAPS). OAPS is the most identified cause of recurrent pregnancy loss and late-pregnancy morbidity related to placental injury.
1. Introduction
Antiphospholipid syndrome (APS) is a systemic autoimmune disorder characterized by the occurrence of vascular thrombosis—arterial or venous—and/or pregnancy morbidity
[1]. These clinical manifestations, according to the Sydney classification criteria, should be associated with the persistent positivity for antiphospholipid antibodies (aPL)
[1][2]. When no other underlying autoimmune disease exists, the syndrome is defined as primary APS. Otherwise, when other autoimmune disorders coexist, mainly systemic lupus erythematosus (SLE), it is named secondary or associated APS
[3][4]. In its most severe form, a life-threatening and infrequent condition known as catastrophic APS (CAPS), patients develop multiple organ dysfunction caused by a rapid onset of small vessel thromboses
[5]. Women that fulfil the Sydney criteria who did not have previous thrombotic events are identified as obstetric antiphospholipid syndrome (OAPS) patients
[6][7].
According to the Sydney clinical criteria, pregnancy morbidity includes (i) at least three consecutive miscarriages before week 10 of gestation, (ii) one or more fetal losses (FL) at ≥ 10 gestational weeks and (iii) stillbirth or prematurity due to eclampsia or severe pre-eclampsia (PE) or placental insufficiency before the 34th week of gestation
[2]. Nevertheless, as occurs in other autoimmune diseases, many patients fail to completely meet the Sydney clinical criteria. These cases are heterogeneously defined across the literature
[8], and it classified as aPL-related obstetric morbidity (OMAPS)
[7][9][10][11][12] (
Table 1).
Table 1. Clinical and laboratory descriptions for OAPS and its variants OMAPS and NC-OAPS.
Antiphospholipid antibodies are a heterogeneous group of immunoglobulins directed against negatively charged phospholipids, cofactors or phospholipid–cofactor complexes that are usually present in monocyte, trophoblast, endothelial and platelet cell membranes
[13][14][15]. The laboratory criteria for APS include the detection of lupus anticoagulant (LA), IgG/IgM anticardiolipin antibodies (aCL) with titers >40 GPL or MPL or >99th percentile or IgG/IgM anti-β2glycoprotein-1 (aβ2GPI) antibodies with titers >40 AU or >99th percentile. Positivity for at least one of the mentioned parameters must be detected at least twice 12 weeks apart
[2]. Cases with low titers of aPL or only one positive test, and therefore not in agreement with the Sydney recommendations, are classified as non-criteria OAPS (NC-OAPS; see
Table 1). These women could present autoantibodies against other phospholipids, such as phosphatidic acid, phosphatidylserine, phosphatidylinositol, phosphatidylglycerol, phosphatidylethanolamine or phosphatidylcholine; cofactors such as prothrombin, annexin A5 or vimentin; IgA aCL or IgA aβ2GPI. All of them are referred to as non-criteria aPL
[8][16].
Antiphospholipid antibodies are detected in 1–5% of healthy women of reproductive age
[17]. aPL/APS is the most frequently treatable acquired thrombophilia during pregnancy
[18], and its prevalence has increased in the last years. In fact, 10–29% of women with poor obstetric outcomes are carriers of aPL
[4].
In the last years, the study of the pathogenic mechanisms leading to the different clinical manifestations of APS has increased, and from these studies, a two-hit hypothesis to explain the development of APS manifestations has been proposed. The presence of aPL is considered the “first hit” that results in the activation of inflammatory—mainly complement-mediated—and coagulation pathways when it is accompanied by another procoagulant trigger, the “second hit”
[15][19]. The expression of a tissue factor (TF) in endothelial and innate immune cells, the activation of nuclear factor κB (NF-κB), the release of proinflammatory cytokines, platelet activation, the production of intracellular reactive oxygen species (ROS) and the induction of the complement cascade are some of the most studied molecular events that explain the development of this syndrome
[19][20][21].
The accurate diagnosis of OAPS is a prerequisite to optimize clinical management and to avoid poor obstetric outcomes, such as fetal growth restriction (FGR) and/or early onset PE. When the standard of care is used, the rate of pregnancy success can rise up to 80% of cases
[4][18][20][22][23][24]. On the other hand, the significant psychological sequelae that may affect the couple, as well as the feeling of blame related to recurrent pregnancy losses and the fear to have more losses, should be considered.
2. Pathophysiology of OAPS
Histological studies found that placental infarction, impaired spiral artery remodeling, decidua inflammation and the deposition of complement split products were the most common features in the placentas of aPL-positive women (
Table 2). These pathological manifestations suggest the role of thrombotic, antiangiogenic and inflammatory factors in the pathological process of the disease
[25][26][27]. aPL have been shown to have a direct embryotoxic effect by affecting both the embryo and the uterus, resulting in decreased implantation or embryo growth
[28][29]. aPL display this embryotoxicity by acting on different cellular targets: innate immune cells (neutrophils, monocytes and platelets); endothelial cells and trophoblast cells (
Figure 1). The clinical manifestations of OAPS, such as miscarriage, fetal loss, PE or placental insufficiency, could be in response to the action of aPL to any one of these cells. The effect of aPL on immune or endothelial cells leads to blood clotting and decreased angiogenesis. aPL disrupt trophoblasts cell biology with affectation from implantation to the remodeling of spiral arteries
[30][31].
Figure 1. Cellular and molecular mechanisms of action of aPL in the pathophysiology of OAPS. aPL affect different cellular processes from blastocyst implantation in the uterus mucosa to trophoblast proliferation and differentiation and, eventually, the impairment of fetal growth due to antiangiogenic and prothrombotic activation. aPL induces inflammation via TLR4/MyD88 (1) in trophoblasts and endothelium and immune cells. Complement component deposition (2) on the endothelium and on trophoblasts drives inflammation and MAC formation, leading to cell death. Trophoblasts apoptosis is also produced when aPT antibodies expose PS on the external trophoblast cellular membrane (3). Inactivation of eNOS and dysfunction in ROS production are observed when the ApoE2 receptor binds to the aβGPI–βGPI complex (4). Trophoblast perturbation also affects decidual NK activity, crucial for embryo implantation (5). aPL in the lumen vessels of placental arteries and veins has pleiotropic actions: induction of leukocyte adhesion (6) on inflamed vasculature that drives neutrophil infiltration (7) inside the decidua and NET formation in response to ROS production (8). These mechanisms altogether enhance thromboinflammation with activation of the coagulation cascade initiated by TF (9). Legend depicts cells and molecules. Image created in BioRender.com (accessed on 13 December 2021).
Table 2. Histopathological findings suggestive of the OAPS.
Whether aPL from obstetric or thrombotic APS patients elicits different mechanisms of action is unresolved. Only APS patients with thrombotic manifestations carry an increased risk of subclinical atherosclerosis, while patients with OAPS do not, thus leading to the idea that distinctive pathogenic mechanisms might sustain the two conditions
[39]. In general, the major mechanism of hypercoagulability in the APS is the aPL-mediated upregulation of TF via the activation of the TLR and NF-κB pathways. These different pathogenic actions were investigated in two in vitro studies. In the first study, immunoglobulins from APS patients with vascular morbidity or from obstetric morbidity differently activated these monocyte signaling pathways
[40]. aPL against β2GPI from thrombotic APS patients were the only ones to induce TF production by monocytes
[40]. In the second study, purified IgG from patients with obstetric but not IgG from non-obstetric APS inhibited trophoblast invasion
[41]. Evidence on the different genetic risk factors for the two APS types has also been investigated
[42].
2.1. aPL Affects Implantation and Trophoblasts Growth
One of the key early events in the establishment of pregnancy is the development of trophoblast subpopulations from the trophectoderm of the implanting blastocyst. After implantation, the trophectoderm of the blastocyst rapidly proliferates and generates the trophoblast, the unique cell type of the placenta. The development and function of the placenta is supported by the maternal decidua, the uterine mucosa
[43]. Trophoblasts differentiate to either extravillous trophoblast cells (EVT) or syncytiotrophoblasts. EVT participate in the remodeling of the uterine spiral arteries that supply the placenta in addition to anchor the placenta to the maternal decidua. Reduced trophoblast invasion and spiral artery transformation are associated with obstetrical disorders such as miscarriages and PE displayed by women with aPL. For their physiological development, trophoblasts require the secretion of angiogenic factors. In in vitro studies, it has been shown that aβ2GPI autoantibodies perturbed the production of vascular endothelial growth factor (VEGF) and placental growth factor (PlGF)
[44] while clearly inhibiting the production of VEGF by endometrial cells
[45][46]. aβ2GPI autoantibodies also prevent EVT function
[47]. aPL antibodies may also affect the syncytiotrophoblast after being internalized by members of the low-density lipoprotein receptor (LDLR) family
[48]. Unlike the syncytiotrophoblast, aPL were not internalized by EVT. It would indicate that the aPL mechanisms of action in EVT are different from that in the syncytiotrophoblast, and it would suggest that aPL trigger intracellular signaling cascades through EVT surface receptor interactions
[49]. At the cell surface of human first-trimester trophoblasts, aβ2GPI autoantibodies activate Toll-like receptor (TLR)-4/Myd88, and the signaling is transduced to the Nalp3 inflammasome, leading to a proinflammatory state by the production of interleukin (IL)-1β and IL-8
[50][51]. A second intracellular mechanism implies the upregulation of miR-146a-3p, leading to the activation of TLR-8 and the secretion of IL-8
[52]. In addition, the mechanisms of the inflammatory response in trophoblasts induced by aβ2GPI autoantibodies involve the impairment of negative regulators of TLR signaling and autophagy
[53]. In turn, autophagy suppression of trophoblasts cells activates decidual NK (dNK) cytotoxicity and inhibits trophoblasts invasion
[54]. In the first-trimester, human decidual leukocytes are primary dNK cells
[55]. A dysregulated dNK activation during this decidual vasculature remodeling is associated with PE in humans
[56].
On the other hand, a clinical study suggested that statins may improve pregnancy outcomes in refractory OAPS by increasing the placental blood flow
[57]. The proposed mechanism was that statins increased the endothelial nitric oxide synthetase (eNOS) activity, raising the serum nitric oxide (NO) levels
[58]. These studies demonstrated a placental vasculoprotective role for eNOS/NO. The pathogenicity of aβ2GPI autoantibodies was observed in a murine model, where autoantibodies against domain I of β2GPI decreased the bioavailable NO by antagonizing the activity of eNOS. The dimerization of β2GPI by the autoantibodies enables its binding to ApoE receptor 2, whose activation triggers the inhibition of eNOS activity
[59]. Thus, aPL-induced leukocyte–endothelial cell adhesion and thrombosis are caused by eNOS antagonism.
2.2. aPL Activates Complement in the Pathophysiology of OAPS
The placenta from OAPS women exhibited high levels of C4a and C3b deposition, which indicated the possibility of complement activation in the pathogenesis of pregnancy complications in OAPS women
[36]. Some clinical studies have found low plasma C3 and C4 complement levels as independent predictors of lower neonatal birth weight and premature delivery
[60][61], and higher serum C3 levels in the first gestational trimester was a protective factor for fetal loss
[62]. The PROMISSE study showed that Bb and C5b-9 were significantly higher in patients with poor obstetric outcomes, as indicative of the participation of complement activation in obstetric complications
[63]. Studies on murine models have shown that complement activation is essential and causative in aPL antibody-induced fetal injury
[64]. Deposition of the human aPL IgG in the mouse’s decidua with subsequent complement activation, neutrophil infiltration and local TNF secretion caused fetal loss and growth retardation
[64][65][66]. Complement activation by aβ2GPI autoantibodies induced thrombus formation subsequent to a priming inflammatory factor
[67] and required the activation of C3 and C5
[68]. C5 and its cleavage product, anaphylotoxin C5a, are pivotal members of complement activation by aPL. Mice lacking C5 or those treated with an anti-C5 mAb showed less neutrophil infiltration into the decidua and were protected from aPL-induced pregnancy complications
[64]. C5a is a potent chemotactic factor and activator of neutrophils. In mice depleted of neutrophils, treatment with aPL did not cause pregnancy loss or growth restriction, nor were inflammatory infiltrates within the decidua. Furthermore, without neutrophil infiltration, there is less C3 deposition
[64]. The ultimate action of aPL activation of a complement is the generation of the C5b-9 membrane attack complex (MAC). The deposition of MAC on the endothelium or on trophoblasts leads to endothelial injury and enhances trophoblast apoptosis
[69].
Moreover, in response to aPL-generated C5a, neutrophils, monocytes, platelets and endothelial cells express TF potentiating inflammation in the decidua, leading to miscarriages
[70]. TF is the major cellular initiator of the coagulation protease cascade and plays important roles in both thrombosis and inflammation. The coagulation cascade is initiated by the complex of TF with factor VIIa (FVIIa). The TF–FVIIa complex activates its substrates: factor X and factor IX. Activated FX (FXa) then converts prothrombin to thrombin, which cleaves fibrinogen and activates platelets, leading to the formation of thrombi. TF also mediates aPL-induced trophoblast injury by two pathogenic mechanisms: (i) complexes of TF with FVIIa and/or FVIIa-FXa induce the production of TNF-α, interleukins and adhesion molecules by cleaving protease-activated receptors (PARs)
[71], and (ii) TF generated by neutrophils contributes to respiratory burst with the generation of ROS
[72]. Both mechanisms lead to fetal death in experimental models of APS.
2.3. Thrombosis versus Complementopathy
Most of the clinical and pathological findings associated with aPL/OAPS seem to have an underlying inflammatory basis more than a primary thrombotic cause
[26]. The underlying mechanisms of complement-mediated thrombosis are not fully clarified, but thromboinflammation is a prominent clinical feature of complement activation
[73]. Although OAPS is generally treated with anticoagulation therapy, heparin also protects pregnant APS patients from complications through inhibition of the complement
[74]. Anticoagulants fondaparinux, a specific inhibitor of FXa, or hirudin, a direct inhibitor of thrombin, did not inhibit the generation of complement split products or prevent pregnancy loss, demonstrating that anticoagulation therapy is insufficient protection against OAPS
[74]. Heparin has long been known to possess anticomplement activity. From early studies in 1929, several mechanisms for this effect have been identified, including the inhibition of C1q binding to immune complexes, interference with the interactions of C4 with C1s and C2, blockade of the formation of the C3 amplification convertase by the alternative pathway and inhibition of the formation of the MAC. Girardi’s data
[74] indicated that heparins prevent obstetrical complications in women with APS, because they block activation of the complement induced by aPL antibodies targeted to decidual tissues rather than by their anticoagulant effects.
2.4. Different aPL with Different Mechanisms of Action
The anionic phospholipid phosphatidylserine (PS) is a procoagulant factor when exposed on the external face of cellular membranes. Annexin A5 binds to PS and forms a protecting anticoagulant complex. However, OAPS patients show reduced levels of Annexin A5 on placental villi, because aβ2GPI autoantibodies disrupt PS–Annexin A5 complexes and promote a thrombogenic state
[75]. The ability of β2GPI immune complexes to disrupt the anticoagulant complex is inhibited by hydroxychloroquine (HCQ)
[76], which reverses the thrombogenic properties of aPL
[77].
As it is described above for eNOS inactivation, the dimerization of β2GPI by aβ2GPI autoantibodies enables its binding to glycoprotein Ibα and ApoE receptor 2 on platelets and potentiates platelets activation
[78] and triggers thrombosis in a murine model of APS
[79]. The effect of aβ2GPI antibodies on platelet aggregation and ATP release involve heparinase activation
[80]. Heparinase inhibition reduces aβ2GPI-dependent platelet activation
[80]. In APS patients, activated platelets release microparticles that contribute to recurrent miscarriages by affecting trophoblasts and endothelial cells
[81], although the exact mechanism is unclear.