Antiphospholipid syndrome

Abstract

Antiphospholipid syndrome (APS) is an autoimmune condition characterized by the occurrence of thrombosis (arterial and/or venous), often multiple, and/or pregnancy morbidity.

Thrombosis is one of the major disease mechanisms, mainly caused by activating endothelial cells, monocytes, and platelets. At present, the management of APS patients with a history of thrombosis is based on long-term antithrombotic therapy, due to the high rate of recurrent thrombosis (29% per year without treatment). Obstetrical APS includes heterogeneous pregnancy complications whose pathogenesis has been increasingly elucidated in the past years. This is due to the current management and treatment, as 80% of APS patients achieve a live birth. The standard approach of APS is not supported by extensive evidence and the best options for refractory and incomplete cases need to be clarified. New and promising molecules are under investigation.

Introduction

Antiphospholipid syndrome (APS) is an autoimmune condition characterized by the occurrence of thrombosis (arterial and/or venous), often multiple, and/or morbidity in pregnancy (recurrent miscarriages, fetal deaths (FDs), and late pregnancy complications such as preeclampsia (PreE) and intrauterine growth restriction (IUGR)), in the presence of antiphospholipid antibodies (aPLs), typically the antibodies included in the classification criteria for APS, lupus anticoagulant (LA), anticardiolipin (aCL), or anti-β2 glycoprotein-I (anti-β2GPI) antibodies ( Table 1 ), although other “non-criteria” antibodies may also play a role. APS may be associated with other autoimmune diseases, mainly systemic lupus erythematosus (SLE), but it can also be seen in patients having no other definable rheumatologic condition (primary APS). Occasionally, it can accompany other conditions, such as infections, drugs, or malignancies .

Table 1

Revised classification criteria for antiphospholipid syndrome.

Vascular thrombosis	≥1 clinical episode of arterial, venous, or small vessel thrombosis. Thrombosis must be objectively confirmed. For histopathological confirmation, thrombosis must be present without inflammation of the vessel wall
Pregnancy morbidity	1 ≥1 unexplained death of a morphologically normal fetus ≥10 weeks of gestation 2 ≥1 premature delivery of a morphologically normal fetus <34 weeks of gestation because of: • severe preeclampsia or eclampsia defined according to standard definition; • recognized features of placental insufficiency 3 ≥ 3 unexplained consecutive miscarriages <10 weeks of gestation, with maternal and paternal factors (anatomic, hormonal, or chromosomal abnormalities) excluded
Laboratory criteria	Presence of antiphospholipid antibodies (aPL), on two or more occasions at least 12 weeks apart and no more than 5 years prior to clinical manifestations, as demonstrated by ≥1 of the following: a Lupus anticoagulant; b Medium to high-titer (>40 GPL or MPL, or >99th percentile) anticardiolipin IgG or IgM; c Anti-β2 glycoprotein-I (anti-β2GPI) IgG or IgM >99th percentile

In this review, we discuss recent advances in the diagnosis of APS including new insights into the pathogenesis and the implications for the identification of new biomarkers. In addition, the importance of recent longitudinal observational studies in understanding the natural history of the syndrome and for embedding research in clinical practice will be elucidated.

Thrombotic APS

Pathogenesis

Although the full pathogenesis of APS is not yet clear, the key mechanisms have been described recently. Thrombosis is one of the major disease features, driven by multiple mechanisms including activation of endothelial cells, monocytes, platelets, coagulation, and complement pathways in addition to inhibition of fibrinolytic and anticoagulation pathways . Recent evidence indicates that vasculopathy, enhanced mainly by severe intimal hyperplasia, can also play a role in arterial vascular occlusion (due to stenotic lesions) and pregnancy morbidity . In support of this hypothesis, Canaud et al. recently demonstrated that the vascular endothelium of proliferating intrarenal vessels from patients with APS nephropathy showed indications of activation of the mammalian target of rapamycin (mTOR) pathway. In cultured vascular endothelial cells, IgG antibodies in patients with APS stimulated mTOR through the phosphatidylinositol 3-kinase (PI3K)-AKT pathway. Patients with APS nephropathy who required renal transplantation and were treated with sirolimus (also known as rapamycin, which can inhibit mTOR) showed no recurrence of vascular lesions and decreased vascular proliferation on biopsy compared with patients with aPLs who were not treated with sirolimus. Of 10 patients treated with sirolimus, seven (70%) had a functioning renal allograft 144 months after transplantation versus three of 27 untreated patients (11%). This study conducted in patients with primary and secondary APS nephropathy, which is mainly mediated by vasculopathy rather than thrombosis, revealed that the activation of the mTOR enzyme stimulates intimal hyperplasia, leading to the formation of the chronic vascular lesions seen in APS. Confirmatory studies and more detailed investigation of the steps leading to mTOR pathway recruitment and the molecular consequences of its activation are still needed. However, there is growing evidence that these enzymes can also induce a prothrombogenic phenotype leading to thrombosis . This might represent an obstacle in proceeding further to test drugs such as sirolimus in patients with APS .

Diagnosis and thrombotic risk assessment

aPL profile and thrombotic risk assessment

The relationship between aPLs and thrombosis has been widely investigated, and a strong association between aPLs and thrombotic events has been confirmed . However, the strength of these results from clinical studies is limited by factors such as differences in the study design and eligibility criteria, and by the diversity of aPLs in terms of types, isotypes, cutoff, and laboratory methods used for their detection.

In clinical practice, it is common to question whether patients with similar clinical manifestations but different patterns and/or combinations of positive aPL test results should be considered equivalent. Patients may have single, double, or triple aPL positivity; have single or multiple isotypes of aCL and anti-β2GPI; or show low versus high titers of antibodies. Whether heterogeneity of “aPL profiles” represents a spectrum of different categories in terms of prognosis and treatment is still a matter of debate.

Recent studies suggest that multiple positive tests for aPLs are more frequently associated with thromboembolic events than a single positive test. However, the presence of a single positive test still has to be considered when assessing thrombotic risk, especially in the context of concomitant autoimmune disease or other cardiovascular risk factors .

Available studies suggest that LA is the single aPL most strongly related to thrombosis . Data from a systematic review of >7000 patients and controls from 25 studies showed that LA has an odds ratio (OR) for thrombosis 5–16 times higher than controls. By contrast, isolated positivity for other aPLs was weakly associated with clinical manifestations of APS . Although LA positivity increased the risk of stroke 48-fold and the risk of myocardial infarction 11-fold, anti-β2GPI only doubled the risk of stroke, with no effect on myocardial infarction . The role of aCL in the absence of LA was also analyzed, and no associated increased risk for stroke or myocardial infarction was reported .

By contrast, some observational studies suggest that patients with SLE and isolated medium–high titer, and persistently positive aCL (defined as more than two-third of positive serial determinations), had an increased risk of thrombosis, while those with occasional aCL positivity did not .

Combined aPL positivity has been associated with an increased thrombotic risk . The so-called triple-positive population (LA + aCL + anti-β2GPI) has been postulated to be the highest risk group . Recently, our group evaluated several combinations of aPL specificities in an attempt to determine the profile that provides the best diagnostic accuracy, not only for APS as a whole but also for thrombosis and pregnancy morbidity independently . Testing for six aPLs resulted in 23 possible combinations. The profile of LA + anti-β2GPI + anti-phosphatidylserine–prothrombin, aPS/PT, had the best diagnostic accuracy for APS as a whole and individually for both thrombosis and/or pregnancy loss (OR 3.73 (95% confidence interval (CI) 1.82–5.38) and OR 3.75 (95% CI 2.13–6.62), and/or OR 4.82 (95% CI 2.17–10.72), respectively), and the best specificity when compared with all the other possible combinations . Multiple aPL positivity also seems to play a role in terms of antibody persistence, and individuals testing positive for more than one aPL tend to have more stable antibody levels on repeated determinations .

In a pediatric population, a statistically significant association between first thromboembolism and persistently positive aPLs, with an overall sixfold increase in the risk for thrombosis, was also reported. These data suggest that the detection of persistent aPLs is clinically meaningful in children with, or at risk of, thromboembolism .

The clinical value of different aPL profiles when assessing the risk of recurrence is also a matter of debate. A recent meta-analysis by Garcia et al. showed that patients with first venous thromboembolism (VTE) who have aPLs have a higher risk of recurrent VTE when compared with patients without aPLs . However, testing for aPLs varied with respect to the antibody sought (aCL, anti-β2GPI, and/or LA), with some studies testing for one aPL, and others testing for more than one, limiting the conclusions drawn.

The aPL risk categories according to profile are shown in Table 2 .

Table 2

Antiphospholipid antibody (aPL) risk categories according to profile.

High-risk aPL profile	• Lupus anticoagulant positivity • Triple positivity (lupus anticoagulant + anti-cardiolipin + anti-β2 glycoprotein-I antibodies) • Isolated persistently positive anti-cardiolipin antibodies at medium–high titers
Low-risk aPL profile	• Isolated, intermittently positive anti-cardiolipin or anti-β2 glycoprotein- I at low-medium titers

Non-criteria aPLs

It has been proposed that several autoantibodies besides aCL, LA, and anti-β2GPI are relevant to APS. These antibodies are directed against other plasma proteins from the coagulation cascade (i.e., PT and/or PS–PT complexes), to specific domains of β2GPI, or interfere with the anticoagulant activity of annexin A5 (A5) . The clinical utility of these newly developed assays and their clinical value in assessing thrombotic risk are being investigated.

IgA isotypes

The issue of the value of immunoglobulin A (IgA) aPL and whether it should be part of the routine diagnostic algorithm has been a subject of intense debate. The current laboratory criteria for APS omit the use of IgA isotypes for both aCL and anti-β2GPI tests. Some available data support testing for IgA aPLs. As data are based on retrospective studies, case reports, and case series, clear-cut recommendations are difficult to make. In addition, comparison between these studies is difficult due to differences in design, population studied, the non-standardized assays used, and the different cutoff chosen for the definition of positivity. As a result and in the absence of well-designed prospective studies, the usefulness of IgA aPL testing in assessing the thrombotic risk in APS continues to be debatable.

Several studies failed to prove the usefulness of IgA aCL and IgA anti-β2GPI testing, because of low prevalence of these antibodies, because they are found along with other aPLs in most cases, and, mainly, because of failure to enhance the diagnostic accuracy of routine testing . Recent studies suggested that isolated IgA anti-β2GPI may identify additional patients with clinical features of APS, and hence recommended testing for these antibodies when other aPLs are negative and APS is suspected. It is important to note that of the 5892 samples tested in this study, only 57 (<1%) were exclusively positive for IgA anti-β2GPI, restricting the application of these recommendations to a limited population of patients.

Based on the published evidence, IgA aPL testing should only be considered for thrombotic risk assessment in selected cases, in the presence of clinical signs and symptoms of APS, mainly associated with SLE, and, particularly, when other aPL tests are negative .

Antiprothrombin antibodies

Antiprothrombin antibodies have been proposed as potential new biomarkers for thrombosis and/or pregnancy morbidity in the setting of APS. Antiprothrombin antibodies are commonly detected by enzyme-linked immunosorbent assay (ELISA), using prothrombin coated onto irradiated plates (aPT), or prothrombin in complex with phosphatidylserine (aPS/PT), as antigen. Although these antibodies can coexist in the same patient, aPT and aPS/PT seem to belong to different populations of autoantibodies .

The clinical value of antiprothrombin antibodies, detected as either aPT or aPS/PT, has been evaluated with contradictory results . Most of the studies support an association between antibodies directed to prothrombin, particularly aPS/PT, and clinical manifestations of APS. A recent systematic review based on data from >7000 patients and controls suggests that while both aPT and aPS/PT increased the risk of thrombosis, aPS/PT seems to represent a stronger risk factor for both arterial and/or venous thrombosis than aPT . The possibility of antiprothrombin antibodies, particularly aPS/PT, being an additional tool for risk stratification is being actively debated, especially when trying to improve the identification of APS patients negative for the criteria aPLs.

Autoantibodies to domain 1 of β2GPI

Several studies have investigated the epitope distribution of anti-β2GPI antibodies with the aim of identifying the pathogenic specificities . The β2GPI molecule has five homologous domains (D): D1–D5. Most of the antibodies have been reported to bind to epitopes located in the domains β2GPI-D1, D4, and D5 and these antibodies may have different clinical interpretations . The main epitope associated with APS has been reported to be cryptic and a conformation-dependent structure that involves different regions of D1 . In an international, multicenter evaluation, an association between anti-β2GPI-D1 antibodies and history of (mostly venous) thrombosis was found . Recent studies have demonstrated that patients with multiple positive test results, usually considered at a higher risk of developing clinical complications, tend to have higher prevalence and higher titers of anti-β2GPI-D1 antibodies . The hypothesis of antibodies specifically binding the domain I of the β2GPI molecule, as opposed to the whole molecule, as a promising biomarker with a better diagnostic accuracy when compared with current assays for anti-β2GPI is scientifically proven, although it needs to be evaluated further.

Risk factors other than aPLs

Recently, the role of vascular risk factors in the development of clinical events in patients with APS has been established. Patients with aPLs presenting with thrombosis frequently have one or more additional cardiovascular risk factors such as hypertension, smoking, hypercholesterolemia, or estrogen use .

When focusing on arterial events, Matyja-Bednarczyk et al. recently showed that livedo reticularis, as well as hypertension and hypercholesterolemia, increased the risk of arterial thrombosis in APS . Moreover, the interaction between aPLs and smoking and oral contraceptives has been elucidated in a case-control study by Urbanus et al. The authors showed that the risk of stroke doubled among LA-positive women who smoked, as compared with nonsmokers, and the risk of stroke among oral contraceptive users multiplied more than sevenfold. All LA-positive women who suffered a myocardial infarction were also smokers.

Coexistent SLE may also have an impact on the risk of thrombosis as SLE is a risk factor for thrombosis per se. Patients with SLE have a higher-than-expected incidence of vascular events, which are not explained completely by traditional vascular risk factors . The combination of SLE and aPLs is particularly a matter of concern, because aPL positivity has been shown to increase the risk of thrombosis in patients with SLE and a diagnosis of SLE appears to further enhance the likelihood of vascular events in patients with aPLs. Indeed, in aPL-positive SLE patients, the annual risk of first thrombosis is higher than in healthy aPL-positive subjects without other cardiovascular risks (4% vs. <1%) .

Observational studies suggest that manifestations of APS other than those included in the clinical classification criteria, such as heart valve lesions , livedo reticularis, and thrombocytopenia may be associated with thrombosis; however, these associations are not considered strong enough to guide clinical decisions.

Thrombotic risk assessment should also be considered in patients with a history of pregnancy morbidity due to aPLs. Lefèvre et al. showed that patients with obstetric APS have a higher thrombotic event rate than healthy women (3.3 vs. 0–0.5/100 patient-years), even if treated with low-dose aspirin (LDA) . In a 10-year observational study of 1592 women with no history of thrombosis who had experienced three consecutive spontaneous abortions before the 10th week of gestation or one FD at or beyond the 10th week of gestation, Gris et al. reported that LA was a risk factor for unprovoked proximal and distal deep and superficial vein thrombosis. Recently, a case–control study including 57 women with primary APS and recurrent early pregnancy loss (REPL) confirmed these results, indicating that a history of REPL associated with aPLs was a risk factor for subsequent long-term thrombosis .

The global APS score and mathematical approaches in risk stratification

One of the unsolved questions is why some aPL carriers never develop any APS manifestation, some develop thrombosis while others present with morbidity during pregnancy and a small number of individuals develop a catastrophic form of APS. Therefore, assessing the risk of developing APS manifestations for an individual with aPLs is very important for physicians .

Three score models have been proposed to quantify the risk of thrombosis/obstetric events in APS . The main aim of these scores is to help clinicians stratify patients according to their risk, identifying those who have a higher likelihood of developing new events and therefore likely to benefit from preventive approaches. The first two scores focused mainly on the aPL profile, while the most recent one, the Global APS Score or GAPSS , also included other variables such as cardiovascular risk factors and autoimmune profile at the time of implementation of the risk model ( Table 3 ).

Table 3

The global antiphospholipid syndrome score (GAPSS).

Factor	Point value
Anticardiolipin IgG/IgM	5
Anti-β2-glycoprotein IgG/IgM	4
Lupus anticoagulant	4
Anti-prothrombin/phosphatidylserine complex (aPS/PT) IgG/IgM	3
Hyperlipidemia	3
Arterial hypertension	1

The GAPSS scoring system is derived from the combination of independent risk for both thrombosis and pregnancy loss, and accounted for multiple factors, including the patient’s aPL profile, conventional cardiovascular risk factors, autoimmune antibody profile, and thromboprophylactic drug use. The GAPSS can be calculated for each patient by adding the points corresponding to the different risk factors, weighted as shown. GAPSS values ≥10 have demonstrated the best diagnostic accuracy compared with the different thresholds for APS diagnosis.

This score has been internally and externally validated in different large prospective studies of patients with primary and secondary APS . GAPSS appears to be a promising tool in quantifying the risk of thrombosis and obstetric morbidity in patients with APS.

Treatment

Primary thromboprophylaxis

It is still an open question whether prophylactic treatment is needed in subjects with aPLs who have no history of thrombosis. The net benefit of active therapy against placebo has never been clearly proven. However, we recommend a careful thrombotic risk assessment and general measures to control cardiovascular risk factors for all patients with aPLs as part of good clinical practice . The avoidance of smoking and control of body weight, blood pressure, and lipids should be considered a primary management goal in all subjects with aPLs . Estrogen-containing oral contraceptive pills or estrogen replacement therapy should be avoided due to their prothrombotic effect.

Autoimmune conditions, mainly SLE, constitute an additional risk factor for thrombosis. Thus, primary thromboprophylaxis should be considered with LDA (75–100 mg/day) in all patients with an underling systemic autoimmune condition and persistent aPLs at medium–high titers (IgM or IgG phospholipid units >40 IgG phospholipid units (GPL) or IgM antiphospholipid units (MPL) or >99th percentile). In patients with SLE and with persistently positive aPLs, primary thromboprophylaxis including LDA and/or hydroxychloroquine (HCQ; 200–400 mg/day) is strongly recommended . This is based on studies that have shown that HCQ protects against thrombosis in patients with SLE, including those with aPLs . Although no study has specifically investigated whether the addition of anti-platelet agents offers additional protection, LDA is generally considered effective for primary thromboprophylaxis . Thus, given the recommendation of HCQ therapy in all patients with SLE, the decision to add LDA should be determined on an individual basis. Specifically, the addition of LDA may be appropriate in selected cases, such as those with a high-risk aPL profile (e.g., triple positivity for LA, aCL, and anti-β2GPI, and/or other concomitant cardiovascular risk factors, and for SLE patients with a history of obstetric APS).

However, while the Physicians’ Health Study demonstrated no protection against deep venous thrombosis in men with aCL receiving LDA , more recent evidence suggests a protective role of LDA in venous thrombosis, at least in the general population .

In asymptomatic carriers of aPLs without an underlying connective tissue disease, the decision regarding thromboprophylaxis should be based on the aPL profile. LDA is suggested for those with a high-risk profile, such as patients with LA, and particularly for triple-positive individuals .

Recently, a prospective, multicenter, randomized, open, controlled trial conducted in patients positive for aPLs aimed to examine the efficacy and safety of LDA versus LDA plus low-intensity warfarin in the primary prevention of thrombosis in aPL-positive patients with SLE and/or obstetric morbidity . No differences were observed in the number of thromboses between patients treated with LDA versus those treated with LDA + low-intensity warfarin. In the LDA + warfarin group, more episodes of bleeding were detected. The authors concluded that the LDA + warfarin regimen was significantly less safe than LDA alone.

Prevention of recurrent thrombosis

At present, the management of APS patients with a history of thrombosis is based on long-term antithrombotic therapy, due to the high rate of recurrent thrombosis (29% per year without treatment) . Whether patients with APS should receive the same therapy as the general population with similar manifestations and whether arterial and venous events should be treated differently remain to be answered.

Two randomized, controlled trials have compared high (target international normalized ratio, INR, 3.0–4.0) with standard (target INR 2.0–3.0) intensity of anticoagulation for secondary thromboprophylaxis in patients with APS . In these trials, no significant differences in terms of efficacy or safety between the regimens were observed. However, both suffered from bias due to overrepresentation of patients with their first VTE. Thus, we recommend indefinite anticoagulant therapy with a vitamin K antagonist (VKA) to a target INR of 2.0–3.0 for patients with APS and the first venous event. A reduction in the duration of treatment with VKA can be considered only in patients with a low-risk aPL profile and clear provoking factors (e.g., surgery and prolonged immobilization) at the time of the thrombosis.

Instead, management of arterial events is more controversial and debate persists . The APS and Stroke Study concluded that patients with stroke and aPL not fulfilling classification criteria should be best treated as the general population, with low-dose aspirin . However, at present we use a more aggressive approach for patients with definite APS with arterial disease and/or recurrent events, which might include VKA with a target INR of 3.0–4.0. Occasionally, combined anticoagulant–antiaggregant therapy may also be considered . In fact, recurrences are very infrequent (0.016–0.031 events per patient per year) among patients receiving effective oral anticoagulation targeting an INR of 3.0–4.0 . However, the physician has to be aware that high-intensity oral anticoagulation therapy carries a risk of serious hemorrhage, although this risk does not appear higher than that observed in other thrombotic conditions warranting oral anticoagulation .

However, the management of VTE is changing rapidly as the new direct oral anticoagulants (DOACs) have been shown to be effective in the management of VTE and they do not require laboratory monitoring . A direct thrombin inhibitor (dabigatran etexilate) and direct anti-Xa inhibitors (rivaroxaban, apixaban, and edoxaban) are currently available. The use of these agents may represent a major step forward as, unlike VKA, they have few reported drug interactions and they do not interact with food or alcohol intake, thereby resulting in more stable anticoagulant intensity. Early evidence for the use of DOACs for secondary thromboprophylaxis for APS patients with a history of VTE is promising, but until data from ongoing clinical trials are available , there is not enough evidence to use DOACs in patients with APS and a history of arterial events. However, it is worth noting that LA testing in patients receiving DOACs may be unreliable .

Alternative therapies for refractory and difficult cases

Long-term management of APS patients with recurrent thrombosis may be complicated by fluctuating INR levels, major bleeding, or a high risk of major bleeding. For these reasons, further therapeutic options other than VKA might be considered in selected cases.

Long-term low-molecular-weight heparin (LMWH; e.g., subcutaneous enoxaparin 1 mg/kg every 12 h or 1.5 mg/kg/day or subcutaneous dalteparin 100 IU/kg every 12 h or 200 IU/kg/day), HCQ (200–400 mg/day), or statins have been suggested in these selected cases .

Recently, rituximab, an anti-CD20 monoclonal antibody, has been shown to be effective in life-threatening catastrophic APS, although it has been investigated in a small number of cases . B-cell depletion with rituximab may also be used successfully in patients with aPLs and autoimmune-mediated thrombocytopenia and hemolytic anemia . A pilot open-label phase II trial of B-cell depletion with rituximab for non-criteria manifestations of APS concluded that rituximab may represent a safe option in APS especially in the case of thrombocytopenia or skin ulcers, although it may not be effective for all non-criteria manifestations (e.g., cardiac valve disease and aPL nephropathy) .

There is limited evidence for the use of steroids, other immunosuppressive drugs, or intravenous immunoglobulin (IVIg) in the treatment of APS patients with thrombosis . These drugs should be considered only as rescue therapy in patients with repeated episodes of thrombosis despite adequate anticoagulant therapy, or in catastrophic APS. Anecdotal cases of the use of eculizumab, which targets the complement protein C5, have also been reported .

Other options can be considered on an individual basis, such as the use of intra-arterial fibrinolysis in patients with acute myocardial infarction associated with APS or prostacyclin analogs in patients with severe ischemic necrotic toes associated with APS.

Newer therapeutic agents targeting pathways involved in the development of aPL-mediated clinical manifestations are under investigation. These include blocking of aPL/β2GPI receptors on target cells, complement and nuclear factor-κB, and P38 mitogen-activated kinase inhibitors. However, the multifactorial mechanisms underlying thrombosis and pregnancy morbidity in APS are still not fully understood and this might limit the development of new targeted therapies for APS. Potentially, the current “antithrombotic” approach to APS patients will be replaced in the future by an “immunomodulatory” approach as our understanding of the mechanisms of aPL-mediated clinical manifestations improves .