16th International Congress on Antiphospholipid Antibodies Task Force Report on Antiphospholipid Syndrome Treatment Trends

Abstract

Antiphospholipid syndrome (APS), an acquired autoimmune thrombophilia, is characterised by thrombosis and/or pregnancy morbidity in association with persistent antiphospholipid antibodies. The 16th International Congress on Antiphospholipid Antibodies Task Force on APS Treatment Trends reviewed the current status with regard to existing and novel treatment trends for APS, which is the focus of this Task Force report. The report addresses current treatments and developments since the last report, on the use of direct oral anticoagulants in patients with APS, antiplatelet agents, adjunctive therapies (hydroxychloroquine, statins and vitamin D), targeted treatment including rituximab, belimumab, and anti-TNF agents, complement inhibition and drugs based on peptides of beta-2-glycoprotein I. In addition, the report summarises potential new players, including coenzyme Q10, adenosine receptor agonists and adenosine potentiation. In each case, the report provides recommendations for clinicians, based on the current state of the art, and suggests a clinical research agenda. The initiation and development of appropriate clinical studies requires a focus on devising suitable outcome measures, including a disease activity index, an optimal damage index, and a specific quality of life index.

Introduction

Antiphospholipid syndrome (APS), an acquired autoimmune thrombophilia, is characterised by thrombosis and/or pregnancy morbidity in association with persistent antiphospholipid antibodies (aPL; lupus anticoagulant [LA], and IgG/IgM anticardiolipin [aCL] and anti-beta-2-glycoprotein I [aβ₂GPI]).¹ Triple aPL-positive denotes the presence of all three aPL, i.e. LA, aCL and aβ₂GPI. The overall prevalence of APS has been estimated at 50 per 100,000 people,² with a female-to-male ratio of approximately 5:1.³ Thrombosis, a cardinal disease manifestation, may be venous, arterial, or microvascular. APS-associated pregnancy morbidity includes recurrent early miscarriages, fetal death after 10 weeks’ gestation, and premature delivery before 34 weeks’ gestation because of pre-eclampsia/eclampsia or placental insufficiency, which leads to fetal growth restriction.¹

Non-criteria manifestations, that are usually refractory to standard APS treatment of anticoagulation with a vitamin K antagonist (VKA), include livedo reticularis, thrombocytopenia, hemolytic anemia, aPL-related cardiac valve disease and nephropathy, skin ulcers, and cognitive dysfunction.¹ Catastrophic APS (CAPS), the most severe form of APS with a high overall mortality rate of 37%, is associated with multiple small vessel thromboses.⁴ Although all these clinical manifestations are grouped as a single entity of APS, there may be individual differences in disease pathogenesis. Patients with CAPS who receive anticoagulation in combination with glucocorticoid plus plasma exchange and/or intravenous immunoglobulin, have the highest survival rate (mortality rate 28.6%).⁴ In other APS patients with small vessel thrombosis, anticoagulation is widely used, although without any strong supporting evidence and further approaches, including immunosuppression, may be required.

This Task Force Report reviews and updates “APS Treatment Trends” that have been discussed during the 16th International Congress on aPL, convened in Manchester, United Kingdom, in September 2019. It represents a continuation of the work of the 14th and 15th International Congress on aPL Task Force Reports.^5,6

Brief overview of the pathogenesis of antiphospholipid syndrome

Evidence suggests that prothrombotic, proinflammatory and angiogenic pathways are involved in the pathogenesis of aPL-related thrombosis, in turn suggesting why antithrombotic treatment alone may not suffice. A key initiating pathogenic process in cell activation is binding of β₂GPI to exposed, negatively charged phospholipids on the surface of endothelial cells, monocytes and platelets, which may all be involved through the shedding of prothrombotic microparticles. Cell activation likely involves binding of aβ₂GPI/β₂GPI complexes to toll-like receptor 4 (TLR4), annexin A2 or low density lipoprotein receptor-related protein 8 (LRP8) and activation of their intracellular signal transduction pathway, with induction of P38/mitogen‐activated protein kinase (P38/MAPK) and nuclear factor kappa-B (NFkB)-dependent genes, resulting in a prothrombotic and proinflammatory phenotype.⁷ Expression of tissue factor (TF), a key initiator of in vivo coagulation, and vascular endothelial growth factor are elevated in patients with aPL.^8–10 Thrombotic APS patients have raised levels of complement activation markers^11,12 and they are recognized as amplifiers of the inflammatory milieu. Increasing evidence suggests that aPL-related thrombosis is mediated by neutrophil activation, leading to release of extracellular chromatin-based structures, termed neutrophil extracellular traps (NETs) through a process known as NETosis.¹³ The key cellular and humoral molecular interactions of aPL leading to thrombosis can be both precipitated by (the “two-hit” hypothesis) and propagate inflammation.¹⁴

Direct oral anticoagulants (DOACs)

Vitamin K antagonists (VKAs), notably warfarin, are the standard treatment for thrombotic APS.^5,6,15,16 The primacy of VKAs for the anticoagulation of APS patients has been challenged by the introduction of direct oral anticoagulants (DOACs). The advantages of DOACs compared to warfarin include prescription of a fixed dose with predictable anticoagulant effect and no routine anticoagulation monitoring, having fewer drug interactions, and no alimentary interactions. These characteristics are appealing for thrombotic APS patients who generally require life-long anticoagulation.

Randomized controlled trial (RCT) evidence

RAPS: Rivaroxaban in antiphospholipid syndrome

This phase 2/3 randomised controlled trial (RCT) compared rivaroxaban 20 mg once daily versus standard-intensity warfarin, target INR 2.5 (range 2.0-3.0) in 116 patients with a first episode of venous thromboembolism (VTE), or recurrence while on subtherapeutic or no anticoagulation. Twenty-eight percent of patients overall (24.6% [14/57] on rivaroxaban, 32.2% [19/59] on warfarin) were triple-positive for aPL, i.e. LA, aCL IgG/M, and aβ₂GPI IgG/M. Patients with previous APS-related arterial thrombosis were excluded. The primary outcome, percentage change in endogenous thrombin potential (ETP) for rivaroxaban, did not reach the non-inferiority threshold. However, peak thrombin was significantly lower on rivaroxaban versus warfarin and the authors concluded that the overall thrombin generation curve, in which the higher ETP reflects the altered reaction kinetics with rivaroxaban, was not indicative of increased thrombotic risk. Although the trial was not powered for clinical outcomes, there were no thrombotic events during seven months of follow-up.¹⁷

TRAPS: Rivaroxaban in thrombotic APS

This phase 3 RCT designed to enroll 536 patients compared rivaroxaban 20 mg once daily (or 15mg once daily if renally impaired [2/59 patients]) versus warfarin target INR 2.5, and recruited 120 triple aPL-positive thrombotic APS patients. This trial was terminated prematurely at the recommendation of the safety committee after a mean follow-up of 1.6 years. Thromboembolic events occurred in seven of 59 patients (annualised thrombosis rate 7.5%) randomised to rivaroxaban (four ischemic strokes and three myocardial infarctions), compared to none on warfarin. Nineteen percent (11/59) of patients on rivaroxaban had previous arterial thrombosis and comprised 57% (4/7) of those with recurrent thrombosis.¹⁸

Rivaroxaban versus VKA in APS: a noninferiority trial

This phase 3 trial randomised 190 patients with thrombotic APS, approximately 60% triple aPL-positive, to rivaroxaban 20 mg once daily (or 15 mg once daily if renally impaired [5/95 patients]) versus VKA, target INR 2.0-3.0 (or target INR 3.1-4.0 among those with a history of recurrent thrombosis). The annualised recurrent thrombosis rate after three years follow-up was 3.9% on rivaroxaban versus 2.1% in the VKA group. Stroke occurred more commonly in patients receiving rivaroxaban (nine events) than in those receiving VKA (0 events) (corrected relative risk (RR), 19.00 [confidence interval (CI), 1.12 to 321.9]). Post hoc analysis suggested an increased risk of recurrent thrombosis in rivaroxaban-treated patients with previous arterial thrombosis, livedo racemosa, or APS-related cardiac valvular disease.¹⁹

ASTRO-APS: Apixaban for the secondary prevention of thrombosis in APS

The ASTRO-APS trial protocol (apixaban 2.5 mg twice daily versus warfarin INR 2.0-3.0 in thrombotic APS patients²⁰ [ClinicalTrials.gov Identifier: NCT02295475]) was modified twice due to a higher rate of thrombosis in patients with a history of arterial thrombosis. The protocol was modified after recruitment of 25 patients, to use apixaban 5 mg twice daily instead of 2.5 mg twice daily. Subsequently, five patients were enrolled. Because of investigator concern for a possibly higher rate of stroke among patients randomized to apixaban, a second protocol modification excluded the subsequent enrollment of APS patients with prior arterial thrombosis, and required MRI of the brain prior to randomization.²¹ Patient enrollment and follow-up is now complete and the investigators hope to publish results in 2020.

Other evidence

A phase 4 pilot study of rivaroxaban 20 mg daily was completed in 82 APS patients with prior VTE. The authors reported recurrent thrombosis in 4.9% (n = 4)²² and concluded that the rate of recurrent thrombosis after at least a year of follow-up was comparable to previous RCTs (annualised recurrent thrombosis rate 1.3-4%)^23,24 among thrombotic APS patients treated with warfarin. A prospective cohort study of 176 patients with 51 months follow-up (82 on DOACs [42 on apixaban, 36 on rivaroxaban and 4 on dabigatran] and 94 on VKA) reported annualized recurrent thrombosis rates of 3.3% (3/10 arterial) and 2.5% (2/12 arterial) for DOACs and warfarin, respectively.²⁵ A retrospective case control study including 18 patients on DOACs (12 on edoxaban, 5 on rivaroxaban and 1 on apixaban) and 36 matched controls on VKA followed for 5 years reported annualized recurrent thrombosis rates of 6.6% for DOACs and 4.4% for VKA.²⁶

A systematic review of 728 APS patients on DOACs (48% triple aPL-positive) reported an annualised recurrent thrombosis rate of 11%. Factors associated with a higher risk for recurrent thrombosis included a higher mean number of prior thrombotic events, a history of combined arterial and venous thrombosis, previous treatment with low-molecular-weight heparin (LMWH), use of immunosuppressant treatment, and patient preference to switch to a DOAC.²⁷ An earlier individual patient data meta-analysis of 447 patients, in which the annualised recurrent thrombosis rate was 11.7%, suggested that additional risk factors for recurrent thrombosis include triple aPL-positivity, a higher number of clinical criteria for APS classification, prior thrombosis while on a VKA and, in patients treated with anti-Xa inhibitors, a history of arterial or small vessel thrombosis. Among the 73/447 patients with recurrent thrombosis, 31 had arterial events; 18 (58%) of these had a prior single VTE, with 10/18 (56%) triple aPL-positive.²⁸

DOACs and VTE among patients with APS

DOACs are established as standard treatment for ‘general population’ patients with a first unprovoked VTE following large phase 3 multicentre international RCTs of DOACs versus standard-intensity warfarin.²⁹ These trials did not focus on APS patients, although post hoc analysis of the RE-COVER®, RE-COVER II™, and RE-MEDY™ RCTs indicated that the efficacy and safety of dabigatran etexilate were not significantly different among patients with at least one positive criteria aPL test and VTE.³⁰ A prospective cohort study in 290 patients with a first unprovoked VTE found that 9% met criteria for APS, showing persistent aPL. Two patients, i.e., 1% of the 191 patients tested for all three aPL, were triple aPL-positive, with persistent triple aPL-positivity proven in one.³¹ A cross-sectional study of 491 patients with a first unprovoked VTE also found that 9% (44/491) of patients met criteria for APS, with 1.4% (7/491) being persistently triple aPL-positive.³² These observations raise the issue of the optimal timing for aPL testing after a first VTE.

Testing for LA in the acute post-thrombotic state may be confounded by acute phase reactants such as factor VIII³³ and C-reactive protein,³⁴ as well as the effects of anticoagulation treatment,³⁵ as false positive and negative results can occur. In contrast, triple aPL-positivity persists in the majority of patients,^36,37 although it occurs in only a minority of patients with a first VTE.^31,32 Guidance regarding the timing of thrombophilia testing including testing for APS exists.³⁸ It may be preferable to defer screening for aPL for most patients with a new VTE in the acute setting. For those patients in whom there is clinical concern for APS, however (e.g., patients with a new VTE and obstetric or non-criteria manifestations of APS), testing can be performed with appropriate interpretation of the laboratory results.

DOACs and APS-related stroke

Standard dosages of DOACs have been established to be effective versus standard-intensity warfarin as the comparator following a first VTE episode in ‘general population’ patients.²⁹ However, whether standard doses of DOACs offer sufficient protection against recurrent thrombosis when high intensity anticoagulation is recommended is an uncertainty.^16,39 There is a lack of consensus regarding appropriate VKA anticoagulation intensity for APS patients in certain circumstances which reflects the lack of conclusive data. The European League against Rheumatism (EULAR) guidelines, based on pooled data from two retrospective studies and two RCTs showing no significant difference in thrombosis recurrences between INR target 3.0–4.0 versus 2.0–3.0 (relative risk 0.46 [0.06–3.52]), recommend treatment with VKA with INR 2.0–3.0 or INR 3.0–4.0 in APS patients with a first arterial thrombosis, considering the individual’s risk of bleeding.^16,40 The RISAPS RCT aims to investigate the use of high-intensity rivaroxaban 15 mg twice daily versus warfarin in APS patients with stroke or other ischaemic brain manifestations: (ClinicalTrials.gov Identifier: NCT03684564).

Other considerations regarding DOAC use in APS

The role of DOACs in APS is not established, and APS, by definition a syndrome, includes a population that is heterogenous in clinical and laboratory manifestations of disease. The TRAPS trial¹⁸ triggered the risk assessment that led to the European Medicines Agency (EMA) statement that DOACs are not recommended for thrombotic APS patients, especially those who are triple positive for aPL. The United States Food and Drug Administration (FDA) has endorsed the EMA recommendations,⁴¹ as have other regulatory authorities worldwide. Of note, DOACs are not contraindicated for APS.^42–45 The TRAPS trial included only triple aPL positive patients, whereas the EMA recommendation extends to all APS patients. Definitive evidence regarding the role of DOAC therapy among patients with thrombotic APS is required. However, concern remains that guidance from regulatory bodies, such as that noted above, will diminish clinician drive and patient willingness to participate in clinical studies necessary to inform optimal care.

Task force recommendations

Recommendations for clinicians

1. DOACs should be avoided in APS patients with arterial thrombosis. For these patients, first line therapy should be a VKA.

2. DOACs should be avoided in thrombotic APS patients with small vessel thrombosis or aPL-related cardiac valvular disease. The first line anticoagulant option should be a VKA.

3. For patients found to have single- or double-positive aPL following a first episode of VTE (in the acute setting or later in their course), we suggest that continuation of the DOAC may be considered, while awaiting confirmation of persistence of aPL, based on testing after at least 12 weeks, and thereafter. Discussion with the patient and shared decision-making regarding the the perceived risks, benefits, and the uncertainties of choice of anticoagulant should be undertaken. Testing for aβ2GPI to distinguish patients with double- rather than triple aPL-positivity should be performed if a DOAC is considered.

4. For triple aPL-positive APS patients, if started on a DOAC upon initial presentation with a first episode of VTE, and upon considering limitations of testing (especially as it pertains to assessment for the presence of LA), we recommend that therapy be switched to warfarin or an alternative VKA. If the patient declines, then the DOAC may be continued, with clinical surveillance. It is suggested that surveillance could include MRI brain imaging to identify ischaemic lesions, which, if present, merit consideration of a switch to alternative anticoagulation, with the first option a VKA.

5. DOACs should not be used in APS patients with recurrent thrombosis while on standard-intensity VKA. Other treatment options may include an increased target INR range, standard treatment dose low-molecular-weight heparin (LMWH), fondaparinux if VKA/LMWH are not suitable, or the addition of antiplatelet therapy.

Clinical research agenda

1. Further studies are required to determine the potential role of DOACs in thrombotic APS. These studies need to take into account that APS is heterogeneous and that thrombotic risk is influenced by both the clinical and laboratory APS phenotype.

2. All cases of DOAC use in APS patients should be reported to the International Society on Thrombosis and Haemostasis-supported international registry, currently being established (ClinicalTrials.gov Identifier: NCT04262492). This Registry will ensure consistency of data collection and provide safety information in APS patients currently on DOACs.

Antiplatelet agents

Low dose aspirin

Aspirin irreversibly inhibits the cyclooxygenase activity of prostaglandin H synthase-1 in platelets, thereby blocking the formation of thromboxane A2, which is a potent vasoconstrictor and facilitates platelet aggregation. Low dose aspirin (LDA), in combination with prophylactic dose LMWH, is standard treatment during pregnancy for obstetric APS.^15,16,46 A meta-analysis of five trials involving 334 patients with recurrent miscarriage, reported live birth rates of 74.3% and 55.9% in women who received a combination of unfractionated heparin/LMWH plus LDA or LDA alone, respectively.⁴⁷ The role of LDA for primary prevention of thrombosis in non-pregnant women with a history of obstetric APS and individuals with persistent aPL is considered below.

Low dose aspirin for primary prevention of thrombosis in patients with persistent antiphospholipid antibodies

A meta-analysis of 10 observational studies, and one RCT (1208 patients and 139 thrombotic events), investigated the utility of LDA in aPL-positive individuals asymptomatic for thrombosis. The majority of the patients had double- or triple-aPL positivity, or persistently high aPL titres. Subgroup analysis showed that the risk of a first venous thromboembolic event (VTE) was significantly decreased by the use of LDA among asymptomatic aPL-positive individuals (7 observational studies; odds ratio (OR): 0.50 [0.25–0.99]); patients with SLE (7 observational studies, 1 RCT; OR: 0.55 [0.31–0.98]); or those with a history of obstetric APS (5 observational studies; OR: 0.25 [0.10–0.62]).⁴⁸

An individual patient meta-analysis of five cohort studies by the same group included 497 subjects with 79 first thrombotic events. After adjustment of cardiovascular risk factors, aPL profiles, and treatment with hydroxychloroquine, a hazard ratio (HR) for the risk of a first thrombosis of any type in aPL carriers treated with LDA versus those not treated with aspirin was 0.43 (95%CI 0.25–0.75). Subgroup analysis showed a protective effect of LDA against arterial (HR: 0.48 [95%CI: 0.28–0.82]) but not venous thrombosis as well as in retrospective (OR: 0.23 [0.13–0.42]) but not prospective studies (OR: 0.91 [0.52–1.59]. After further adjustment on the gender, age, and presence of cardiovascular risk factors, subgroup analysis showed a protective effect of aspirin against arterial thrombosis was observed in patients with SLE (HR: 0.43 [95%CI: 0.20–0.93]) and asymptomatic aPL carriers (HR: 0.21 [95%CI 0.04–0.99). The number of women with obstetric APS was relatively small, limiting conclusions: 15/221 (7%) of patients on LDA and 65/276 (24%) of those not on LDA.⁴⁹ The APLASA (Antiphospholipid Antibody Acetylsalicylic Acid) trial was the only RCT that directly addressed the question of the efficacy of LDA in primary thrombosis prevention in asymptomatic, persistently aPL-positive individuals, but could not confirm the benefit of low dose aspirin for primary prophylaxis in this setting, perhaps due to limited power and/or sample size.⁵⁰ Of note, a prospective study reported a significantly increased rate of VTE and cerebrovascular events, in women with obstetric APS, despite low-dose aspirin primary prophylaxis, compared to women with heritable thrombophilia or with negative thrombophilia screens.⁵¹

It is unclear whether the benefit of LDA outweighs the risk of major bleeding associated with LDA in a low-risk population. The estimated average annual incidence rate of overall VTE in the general population is 0.1–0.18 per 100 patient-years and similar to that of stroke.⁵² The annual incidence of thrombosis in unselected aPL-positive patients is reported to be 0 to 2.8%.⁵³ A more recent review estimated that the annual thrombosis rate among aPL-positive individuals with or without systemic autoimmune disease (SAID) is 0 to 5.3%, probably very low (<1%/year) in those with no other SAID or other thrombosis risk factors. The authors suggest risk stratification, based on aPL profile, age, additional SAIDs and traditional cardiovascular or VTE risk factors.⁵⁴ Compared with the general (non-aspirin treated) population, the risk of major bleeding with LDA for primary cardiovascular disease prevention was reported in a systematic review to be 3.6 per 1000 person-years.⁵⁵ This systematic review reported that LDA increased major gastrointestinal bleeding risk by 58% (OR, 1.58 [95% CI, 1.29 to 1.95]) and haemorrhagic stroke risk by 27% (OR, 1.27 [CI, 0.96 to 1.68]).⁵⁵ SLE is associated with increased thrombotic risk, (both arterial and venous), increased by the presence of aPL.⁵⁶ In the absence of RCTs evaluating prophylactic strategies, recommendations are based on analysis of lower quality studies and expert opinion.

The EULAR guidelines advise that the decision to treat with LDA for primary thromboprophylaxis should be based on stratification of thrombotic risk:¹⁶ thus, in asymptomatic aPL carriers, with a ‘high-risk’ aPL profile (defined as the presence of persistent LA, double- or triple-aPL positivity, or persistently high aPL titres) with or without traditional risk factors and in patients with SLE and a ‘high-risk’ aPL profile, prophylactic treatment with LDA is recommended (2a/B). In individuals with a ‘low-risk’ aPL profile (defined as isolated aCL or aβ₂GPI at low-medium titres, particularly if transiently positive), the EULAR guidelines advise that LDA may be considered (2 b/C). They advise that non-pregnant women with a history of obstetric APS (with or without SLE) as the sole manifestation of APS should receive prophylactic treatment with LDA, after adequate risk/benefit evaluation (2a/B).^16,40

Low dose aspirin for secondary prevention of thrombosis

Optimal antithrombotic therapy among APS patients with prior stroke/transient ischaemic attack remains uncertain. Prospective evidence has suggested that no substantive benefit exists between high-intensity and standard-intensity warfarin, although patients with arterial thrombosis were under-represented.^23,24 Small prospective cohort evidence suggests VKA over LDA for APS-related stroke.⁵⁷ The APASS (Antiphospholipid Antibodies and Stroke Study) in older patients with stroke reported no difference in event recurrences between LDA and warfarin,⁵⁸ but aPL testing did not fulfil the international APS laboratory classification criteria.¹ While no definitive data exist, professional society guidance to elect either high-intensity VKA or standard-intensity VKA, with or without antiplatelet therapy has been recommended.¹⁶ Based mainly on expert consensus, the EULAR guidelines recommend consideration of LDA plus standard-intensity VKA (4/C) as an option, in addition to consideration of standard- versus high-intensity VKA alone, following a first arterial thrombosis; and, in APS patients with recurrent arterial or venous thrombosis on standard-intensity VKA, addition of LDA, increase of INR target to 3.0–4.0 or change to LMWH.^16,40

There is a lack of data about the use of other antiplatelet agents in patients with aPL/thrombotic APS, including clopidogrel, prasugrel, ticagrelor and cangrelor. Dipyridamole, a phosphodiesterase inhibitor, that inhibits platelet function and induces vasodilation,⁵⁹ also appears to inhibit aPL-mediated NETosis (discussed below).

Recommendations for clinicians

1. In asymptomatic aPL carriers, with or without SLE, or in individuals with prior obstetric APS, with persistent LA, double- or triple-aPL positivity, or persistently high aPL titre, LDA should be considered for primary prevention of thrombosis on a case-by-case basis.

2. In asymptomatic aPL carriers, with or without SLE, or in individuals with prior obstetric APS who have any other aPL laboratory phenotypes, LDA may be considered for primary prevention of thrombosis on a case-by-case basis.

3. The risk-benefit analysis should include patient-related factors for arterial thrombosis and VTE. Risk factors for bleeding and upper gastrointestinal reflux disease should also be taken into account.

4. There is insufficient evidence to make strong recommendations about the use of LDA for secondary prevention following a first APS-associated arterial thrombosis. LDA may be considered, in combination with standard-intensity VKA (target INR 2.5, range 2.0–3.0), in APS patients with a first arterial thrombosis, with an alternative option high-intensity VKA.

5. LDA may be considered, in combination with anticoagulation, in APS patients who develop recurrent arterial or venous thrombosis while on standard-intensity VKA.

Clinical research agenda

1. RCTs should be undertaken to define the potential role of LDA or other antiplatelet agents for primary prevention of thrombosis in aPL-positive patients asymptomatic for thrombosis.

2. RCTs are also required to define the role of LDA or other antiplatelet agents, in combination with anticoagulation, in thrombotic APS patients.

Adjunctive therapies

The majority of thrombotic APS patients respond to anticoagulation, but a small proportion continue having clinical events despite anticoagulation. Hydroxychloroquine, statins and vitamin D may have a role as adjunctive treatment in the treatment of thrombotic APS patients.

Hydroxychloroquine

In vitro and in vivo studies show that antithrombotic effects of hydroxychloroquine (HCQ), standard treatment in SLE, include reversal of aPL-induced platelet activation,⁶⁰ reduction of LDL and VDRL cholesterol, and increased HDL cholesterol levels.⁶¹ HCQ also protects the annexin A5 anticoagulant shield from syncytiotrophoblast disruption by aPL.⁶² HCQ reduced clot formation and thrombin generation in a mouse model and in human endothelial cells. It may improve endothelial function and correct the proinflammatory phenotype observed both in vivo and in vitro.⁶¹ A retrospective study suggested that HCQ may also reduce aPL titres in primary APS patients.⁶³ A prospective study showed that long-term HCQ use (average 2.6 years) was associated with a decrease in aPL titers.⁶⁴

Primary prevention of thrombosis

HCQ has been related to a reduction in thrombosis risk in aPL-positive patients with SLE,⁶⁵ although the aforementioned individual patient meta-analysis found no independent protective effect of hydroxychloroquine.⁴⁹ The Antiphospholipid Syndrome Alliance for Clinical Trials and International Networking (APS ACTION) initiated a multicentre, international RCT of HCQ versus standard care in persistently aPL-positive, thrombosis-free patients without systemic autoimmune diseases. The trial was terminated early due to the low recruitment rate (n = 20), exacerbated by the prolonged manufacturing shortage and significant price increase of HCQ in the United States. Thus, no conclusions on the effectiveness of HCQ could be made. The authors concluded that conducting an international RCT without pharmaceutical industry support is extremely challenging.⁶⁶

Hydroxychloroquine: Secondary prevention of thrombosis

Data on the role of HCQ in secondary prevention of thrombosis in primary APS patients are also scarce. A prospective non-randomised study in 40 patients suggested that adding HCQ to oral anticoagulation with VKA reduced VTE in patients with primary APS.⁶⁷ The HIBISCUS project was proposed to study the use of HCQ in secondary prevention of thrombotic and obstetric events in primary APS.⁶⁸ A prospective RCT of HCQ versus placebo aims to assess the role of HCQ in the secondary prevention of thrombotic events (ClinicalTrials.gov Identifier: NCT03540810). A pilot open-label RCT of 50 primary APS patients and 15 asymptomatic aPL carriers showed a significantly lower thrombosis rate (average follow-up 2.6 years) in patients randomised to HCQ plus standard care vs. standard care alone.⁶⁴

Hydroxychloroquine: Obstetric APS

Standard treatment with LDA plus LMWH to improve obstetric outcome fails in 20-30% of APS patients.⁴⁷ A systematic review analysed the limited evidence and found that one study showed that HCQ improved pregnancy outcome, but its effect was not adjusted for the use of other medications (LDA, LMWH, steroids). An expert panel concluded that HCQ use could be considered after failure of standard treatment and in women with previous thrombosis (either arterial and/or venous), and/or with previous ischaemic placenta-mediated complications.⁶⁹ EULAR recommendations for the use of HCQ in pregnancy state that in women with ‘criteria’ obstetric APS with recurrent pregnancy complications despite combination treatment with LDA and heparin at prophylactic dosage, increase of heparin to therapeutic dose, addition of hydroxychloroquine or addition of low-dose prednisolone in the first trimester may be considered.¹⁶ In a retrospective cohort study of 170 pregnancies in 96 women, HCQ treatment was associated with a higher rate of live births (67% in HCQ-treated vs 57% in those who did not receive hydroxychloroquine; P = 0.05) and a lower prevalence of aPL-related pregnancy morbidity (47% vs 63% respectively; P = 0.004).⁷⁰ A multicentre RCT of HCQ versus placebo during pregnancy in women with aPL (HYPATIA; HYdroxychloroquine to Improve Pregnancy Outcome in Women with AnTIphospholipid Antibodies) is assessing the effect of HCQ on adverse pregnancy outcomes.⁷¹ A prospective RCT of HCQ versus placebo aims to assess the role of HCQ in achieving an uncomplicated term pregnancy in primary obstetric APS (ClinicalTrials.gov Identifier: NCT04275778).

Task force recommendations

Recommendations for clinicians

1. The addition of HCQ may be considered as adjunctive to antithrombotic treatment, for anticoagulant-refractory thrombotic APS, in accordance with our previous Task Force recommendations.^5,6

2. The addition of HCQ to standard treatment may be considered in patients with obstetric APS refractory to standard treatment with LDA and LMWH.

Clinical research agenda

1. The potential benefit of HCQ use in non-SLE patients with aPL/thrombotic APS should be explored further in appropriate studies.

2. The results of ongoing studies could inform HCQ use in primary APS.

Statins

Statins inhibit the enzyme HMG-CoA reductase which has a central role in hepatic cholesterol production, but also have pleiotropic effects including anti-inflammatory and antithrombotic actions on endothelial cells and monocytes.⁷²

Statins: Primary prevention of thrombosis

Numerous studies in the general population show that statin use leads to primary⁷³ and secondary⁷⁴ prevention of cardiovascular disease, and statins have been reported to reduce the occurrence of symptomatic VTE among healthy individuals.⁷⁵ However, we are not aware of published studies on the use of statins for primary or secondary prevention of thrombosis in aPL-positive patients. Statins reduce aPL-induced expression of TF and cell adhesion molecules.^76,77 A prospective, open label pilot study concluded that the use of fluvastatin in aPL-positive patients, reduced proinflammatory and prothrombotic biomarkers such as interleukin (IL)-6, IL1β, vascular endothelial growth factor, TNF-α, interferon–α and soluble tissue factor.⁷⁸

Statins: Secondary prevention of thrombosis

There is no evidence supporting statin use in APS patients with normal lipid levels, as also concluded in our previous Task Force reports.^5,6 The importance of hyperlipidaemia for thrombosis prediction in APS patients is emphasised by its inclusion, together with hypertension and aPL status, in the Global AntiphosPholipid Syndrome Score (GAPSS), a well-known, validated score in APS.⁷⁹ The adjusted GAPSS has been applied to asses the risk of recurrent thrombosis in the APS ACTION cohort and for risk stratification in young APS patients with acute myocardial infarction.^80,81

Statins: Obstetric APS

The global prevalence of pre-eclampsia, a leading cause of maternal and fetal mortality, is estimated to be 4.6% (95% CI 2.7%–8.2%) of primigravidae.⁸² Studies in animal models and in humans suggest that pravastatin may prevent pregnancy complications associated with placental dysfunction, particularly preeclampsia.⁸³ A large cohort study showed that statins in the first trimester do not appear to be associated with congenital fetal abnormalities.⁸⁴ In this regard, pravastatin is hydrophilic and may have a limited passage through the placenta.⁸³ A small open label non-randomized study investigated the effect of pravastatin 20 mg od in 21 pregnant women with probable APS with early severe preeclampsia and/or fetal growth restriction (FGR). All women received treatment with LDA plus LMWH and 11/21 received additional pravastatin (20 mg/d). Results suggested that pravastatin improves pregnancy outcomes in women with refractory obstetric APS when taken at the onset of preeclampsia or FGR until the end of pregnancy.⁸⁵ However, a randomised, double-blind, placebo-controlled trial in 62 women with early-onset severe pre-eclampsia (without APS) showed no improvement in pregnancy outcome with pravastatin 40 mg daily.⁸⁶

A systematic review that included 16 clinical studies, noted that although early uncontrolled case series reported congenital anomalies associated with statin use, more recent observational studies did not report an increased risk of congenital anomalies with statin exposure in pregnancy when compared to control groups or the prevalence of congenital anomalies in the non-APS pregnant population. The findings of this systematic review showed no clear relationship of congenital anomalies with statin use in pregnancy, and supported the conclusion that statins are probably not teratogenic.⁸⁷ The United States FDA and other parts of the world still categorise statins as contraindicated in pregnancy, which limits their application during pregnancy.

Task force recommendations

Recommendations for clinicians

1. Statins may be beneficial in the primary and secondary prevention of arterial thrombosis in patients with aPL/APS. However, based on available data, statins cannot be recommended in patients with aPL/APS in the absence of hyperlipidaemia, in accordance with general population guidelines

2. Statins may be considered as adjunctive to antithrombotic treatment in anticoagulant-refractory thrombotic APS patients.

These recommendations accord with our previous Task Force reports.^5,6

Clinical research agenda

1. There is a pressing need for further studies on statins, to define clinical and laboratory indications/biomarkers to inform RCTs in individuals with aPL/thrombotic APS.

2. Studies should be performed to define the potential utility of statins in patients with thrombotic APS.

3. The results of active largescale RCTs investigating whether statins can improve pregnancy outcomes in women at high risk of pre-eclampsia, should guide the future use of statins during pregnancy.

Vitamin D

Low vitamin D levels correlate with venous and arterial thrombotic manifestations in APS patients. Vitamin D insufficiency (<30 ng/mL) occurs in up to 70% of APS patients, while the prevalence of vitamin D deficiency (<10ng/mL) ranges between 11-50%.^88–90 Notably, not only were vitamin D levels shown to be lower in APS patients compared to controls, but values in APS patients with thrombotic manifestations were significantly lower than in APS patients with only obstetric manifestations.^89,90 A retrospective cohort study and meta-analysis of four case-control studies confirmed that the combined mean difference in serum vitamin D levels between APS and controls was −3.605 (p < 0.001) and that APS patients had an approximately 3-times increased frequency of vitamin D deficiency.⁹¹ These studies suggest a possible role of vitamin D in ameliorating the development of thrombotic complications in aPL-positive individuals.

The central role of inflammation in aPL-mediated thrombosis underpins the potential of vitamin D in treating APS patients since it possesses numerous immunomodulatory properties.⁹² Among its many effects, the ability of vitamin D to inhibit TLR4/MyD88 signaling,^93,94 TF expression⁸⁸ endothelial activation, and inflammation and cell perturbation play crucial roles in obstetric as well as thrombotic APS.^95,96 An in-vitro study demonstrated that vitamin D modulates signalling through TF/PAR-2, indicating that it could potentially limit TF/PAR-2 mediated placental inflammation and subsequent adverse outcomes in APS pregnancies.⁹⁷ Studies evaluating pravastatin provide further evidence for vitamin D treatment in obstetric APS since this drug has been shown to prevent adverse pregnancy outcomes in APS patients and mouse models. Pravastatin blocks TF and PAR2 expression on neutrophils and increases vitamin D levels in animal models,^84,98 suggesting that these effects could underlie its potential efficacy in treating pregnant patients with APS. In the early stages of pregnancy, trophoblasts respond to and produce vitamin D, which promotes an anti-inflammatory environment and induces decidualization for successful obstetric outcomes.^99,100 In the general population, vitamin D deficiency in pregnant women is associated with an increased risk of obstetric complications including pre-eclampsia and fetal growth restriction.¹⁰¹ A retrospective cross-sectional study of women with recurrent pregnancy losses highlighted an association between the presence of aPL and low vitamin D levels.¹⁰² Low vitamin D levels were associated with complement activation, placental insufficiency and pre-eclampsia in this study.

In a small observational study evaluating vitamin D deficiency in primary APS patients, vitamin D supplementation in a subgroup of these patients (average 400 IU daily) was ineffective in raising levels above 30 ng/mL in approximately 60% of patients.⁹⁰ In contrast, a more recent interventional study of 16 APS patients receiving supplementation with 1000 IU vitamin D daily for 3 months reported an almost doubling of median 25-hydroxy vitamin D3 levels.¹⁰³ Interventional studies of vitamin D supplementation in the general population and in other chronic prothrombotic conditions have utilized doses as high as 100,000 IU monthly, although the antithrombotic effect of vitamin D treatment in these studies was not conclusively demonstrated.¹⁰⁴

Task force recommendations

Recommendations for clinicians

1. Vitamin D deficiency and insufficiency should be corrected in all aPL-positive patients based on the general population guidelines, as also recommended in our previous Task Force reports.^5,6

Clinical research agenda

1. The prognostic role of vitamin D deficiency and therapeutic value of supplementation (including the dosage and definition of treatment goals) in aPL-positive patients should be clarified with prospective studies that include appropriate control groups and standardized definitions of vitamin D deficiency.

2. In addition to clarifying the role of vitamin D treatment in ameliorating thrombotic and obstetric pathology in APS amid dosage considerations, further studies are needed to determine if vitamin D deficiency observed in APS patients occurs as a part of disease pathogenesis, and/or a consequence of disease activity and/or an incidental disease modifying factor.

Biologics: Rituximab, belimumab, and anti-TNF therapies

Autoantibodies have a central role in the pathogenesis of APS,¹⁰⁵ thus, B-cells are a potential therapeutic target. Currently, belimumab and rituximab are the two most commonly used B-cell modulating agents in the treatment of autoimmune diseases.

Rituximab

Rituximab is a chimeric monoclonal antibody that blocks the CD20 molecule on many B-cell precursors and is currently approved in several countries for the treatment of rheumatoid arthritis and anti-neutrophil cytoplasmic antibodies (ANCA) vasculitis. Clinical trials in lupus failed to reach their primary endpoints,^106,107 though rituximab is widely used in clinical practice and has been recommended in the guidelines of the American College of Rheumatology (ACR)¹⁰⁸ and EULAR¹⁰⁹ for the treatment of lupus nephritis and by NHS England for more general use in lupus.¹¹⁰

The evidence for the use of rituximab in APS derives from multiple case reports, case series and an open label clinical trial. However, reports of its use for recurrent thrombosis are scarce. The largest reported case series includes five patients with SLE/APS patients who had recurrent thrombosis despite appropriate anticoagulation with warfarin. Four of these patients had no further thrombotic events after its use.¹¹¹ In primary APS, only sporadic reports of successful treatment for thrombotic events are available.^112,113 The data seem to be more promising for some non-criteria manifestations. An open-label phase II study of rituximab for non-criteria manifestations of APS demonstrated that the safety profile of rituximab in the disease was very similar to that reported in other autoimmune diseases. More than half (13/19) of the patients enrolled had partial or complete response in treating aPL-related thrombocytopenia, cognitive dysfunction, aPL-related nephropathy or skin ulcers, despite no substantial change in aPL profiles.¹¹⁴ In contrast, a small observational study showed significant reductions of aCL levels in seven SLE patients (mean baseline IgG aCL level: 20.6 standardized IgG antiphospholipid units (GPLU) (range (SD) 10–32, (10.1), normal level <5) at 6-9 months post B-cell depletion with rituximab.¹¹⁵ Numerous cases have been reported describing the successful use of rituximab, in particular, for thrombocytopenia¹¹⁶ and skin ulcers.¹¹⁷ A growing number of case reports and series report on the use of rituximab for diffuse alveolar hemorrhage. Approximately 13 cases have been reported, with 50% of patients achieving remission using rituximab monotherapy or in combination with cyclophosphamide or mycophenolate mofetil.^118–121

Rituximab has been reported in the treatment of CAPS. Based on the CAPS international registry, its use has been proposed for those patients who have had CAPS refractory to triple therapy (steroids, anticoagulation and plasma exchange).^122–124 Finally, for obstetric manifestations of the disease, there is one case report of unsuccessful treatment in SLE/APS.¹²⁵ The main caveat about the evidence available is that is predominantly based on case reports or series. This type of evidence is often biased as unsuccessfully treated patients are less frequently reported.

Belimumab

Belimumab is a monoclonal antibody directed against BAFF, a B cell activating molecule.¹²⁶ This drug has been approved for use in SLE in several parts of the world. Several reports have shown that aPL titers in SLE patients treated with belimumab decrease. This effect was observed for both aCL and aβ₂GPI, but no change in LA status was noted.^127,128 A post-hoc analysis of two randomised placebo-controlled trials in SLE did not show any significant effect of belimumab on IgG or IgM aCL titres, but only on IgA, notably in patients on concomitant antimalarials: median titre (interquartile range [IQR]) IgA aCL at baseline; placebo 22 (18; 30); belimumab 10 mg/kg 24 (19; 37); median change [IQR] at 12 months: placebo −7 (−12; 2); belimumab 10 mg/kg: −10 (−15; −7), p < 0.0007.¹²⁹ Only two APS cases with clinical outcomes have been published; one with refractory lower extremity skin ulcers and one with thrombocytopenia. Both had a partial response and were able to taper glucocorticoids after the introduction of belimumab to the therapeutic regimen.¹³⁰

Anti-TNF therapy

The PROMISSE (Predictors of Pregnancy Outcome: Biomarkers in Antiphospholipid Antibody Syndrome and Systemic Lupus Erythematosus), a longitudinal, prospective, multicenter observational study that enrolled pregnant women with aPL/APS and SLE (ClinicalTrials.gov Identifier: NCT00198068), has led to new insights about the role of TNF-α in pregnancy morbidity. In the PROMISSE study, 39% of pregnancies in women with APS and LA resulted in adverse pregnancy outcomes (APO) despite treatment with LMWH/unfractionated heparin and LDA.¹³¹ Angiogenic dysregulation early in pregnancy predicted APO, most of which were due to failure of adequate vascularization of the developing placenta and underperfusion of the fetus.¹³² Mouse models show that aPL causes elevation of TNF-α levels in the placental tissues, that TNF-α levels are associated with pregnancy loss, and its blockade rescues pregnancies.¹³³ In a recent case series, 18 aPL-positive women with obstetric APS refractory to LMWH, aspirin and hydroxychloroquine were treated with adalimumab or certolizumab. Seventy percent of the patients had a live birth.¹³⁴ Currently the IMProve Pregnancy in APS with Certolizumab Therapy (IMPACT: ClinicalTrials.gov Identifier: NCT03152058) clinical trial is ongoing in patients with obstetric APS. This trial should clarify the role of anti-TNF therapy for this APS manifestation. It is important to note that there are reported cases of the development^135,136 or exacerbation¹³⁷ of APS while using anti TNF therapy. These drugs currently have no role in other disease manifestations and may have deleterious effects.¹³⁸

Task force recommendations

Recommendations for clinicians

1. Rituximab may have a role in the treatment of some aPL-related non-criteria manifestations, such as thrombocytopenia, diffuse alveolar hemorrhage, aPL-related nephropathy and microvascular skin ulcers. Rituximab may also have a role in refractory CAPS. There is a paucity of evidence to inform the use of rituximab for anticoagulant-refractory thrombotic APS.

2. There are limited data currently, to support the use of belimumab in APS.

Clinical research agenda

1. The introduction of fully humanised anti-CD20 monoclonals, e.g. obinutuzumab, offers the chance to undertake larger studies, without the relatively frequent allergic responses that accompany the use of rituximab.¹³⁹

2. As belimumab becomes more widely available internationally, it should become easier to assess its true potential in both the thrombotic and non-criteria APS manifestations.

3. The results of the ongoing clinical trials using TNF-α blockade might provide justification for its use in obstetric APS.

Complement inhibition

Complement activation contributes to thrombosis and pregnancy complications in animal models of APS.^140,141 Extensive evidence indicates that aPL triggers complement activation (generation of C5a) and that specific complement inhibition (anti-C5a) reduces fetal loss.¹⁴¹ Inhibition of complement has also proven to be an effective therapeutic intervention for treating microvascular, as well as large vessel, thrombotic disease in paroxysmal nocturnal haemoglobinuria.¹⁴²

Passive transfer of human aPL induced complement activation in pregnancy animal models.¹⁴¹ C5a, a potent anaphylotoxin and chemotactic protein, promotes vascular inflammation and thrombosis by activation of inflammatory cells, endothelial cells and platelets.^141,143 The inflammatory and prothrombotic milieu is amplified by C5a-C5aR interaction with increased recruitment of neutrophils and monocytes, expression of adhesion molecules and tissue factor by neutrophils, resulting in trophoblast damage and angiogenic factor imbalance.^144,145 The importance of C5a and C5aR in rescuing aPL-mediated thrombosis phenotype has been confirmed in experimental models: C5a monoclonal antibodies (mAb), C5aR antagonist peptides and anti-C5aR mAb have been able to reverse the pathogenic effect of complement-induced fetal injury.¹⁴⁶

Heparin, an anticoagulant used to treat thrombotic events and to prevent miscarriages, has been successfully used as an anti-inflammatory drug due to its anti-complement properties.¹⁴⁷ Although complement has several implications in microvascular thrombotic diseases, few studies have demonstrated complement consumption in APS. One possible explanation is that complement activation products are unstable in vitro leading to collection artifacts and methodologic issues. Another possibility is that complement activation is local and not detectable in the circulation. Blood cell bound C4d has been reported to be a more sensitive indicator of complement activation in SLE and APS than serum levels of complement C3 and C4.¹⁴⁸ A novel 2-stage approach to measure aPL-induced complement components has proven to be more sensitive means of detection than traditional methods. When patient’s plasma was preincubated with phospholipid vesicles. C5a and soluble C5b-9 levels were significantly increased in APS patients compared to controls.¹⁴⁹ Further validation of these methodologies should be explored in larger prospective aPL/APS cohorts.

Outcomes in complement-mediated thrombotic microangiopathies (TMAs), such as paroxsysmal nocturnal hemoglobinuria (PNH) and atypical haemolytic uremic syndrome (aHUS), were improved with anti-complement therapy (anti-C5). Eculizumab, a monoclonal anti-C5 drug, has been utilised as salvage therapy in refractory CAPS in case reports.^150–157 A recent study showed that CAPS patients experience a higher frequency of rare germline mutations of complement regulatory proteins, that might make them more susceptible to thrombosis.¹⁵⁸ There is a substantial body of literature endorsing the use of eculizumab for preventing aPL-related nephropathy recurrence following renal transplantation. Eculizumab improved creatinine levels after infusion in three patients but failed to prevent chronic vascular changes, suggesting multiple mediators of disease pathogenesis.¹⁵³ Anti-β₂GPI IgA levels appear to be an independent risk factors for early graft loss after renal transplantation in a retrospective cohort in Spain, but complement activation mediated by IgA needs evaluation.¹⁵⁹ An association between complement activation and recurrent thrombosis in APS patients has been demonstrated using a functional modified HAM (mHAM) assay and patient-derived aβ₂GPI also increased C5b-9 deposition on the cell surface. These observations suggest a basis for the use of complement inhibition in patients with refractory thrombotic APS.¹⁵⁸

The successful use of Eculizumab was reported in pregnancy, in a triple-positive APS patient with previous recurrent arterial events despite anticoagulation.¹⁶⁰ Of note, an in vitro assay for eculizumab/C5 complexes demonstrated negligible drug concentration in the infant’s serum. Results from the PROMISSE Study demonstrated that the 20.5% of SLE patients and/or aPL with APO during pregnancy presented higher levels of complement split product Bb and sC5b-9 as early as 12–15 weeks. Bb and sC5b9 at 12–15 weeks were significantly associated with APO, after controlling for demographic and clinical risk factors for APO. The adjusted OR for APOs with alternative complement pathway activation, measured by circulating Bb at 12–15 weeks of gestation, was increased in patients who were LA-positive or had a history of thrombosis.¹⁶¹ Safety data in pregnant women have been published^162,163 and although numbers are small, eculizumab may eventually be an alternative for treatment during pregnancy in selected APS patients who are at extremely high risk.

Ongoing interventional clinical studies in antiphospholipid antibody-positive patients

Prospective validation for complement blockade findings from case reports is necessary. New treatment targets from the complement cascade are an alternative that should be further explored. Clinical trials of varying status are detailed below:

1. Terminated due to slow patient enrollment: A Phase IIa for the Treatment of Non-Criteria Manifestations of Antiphospholipid Syndrome: Nephropathy, Thrombocytopenia and Skin ulcers was designed to evaluate safety of an intravenous C5a inhibitor (ClinicalTrials.gov Identifier: #NCT02128269).

2. Active not recruiting: Phase II Study of the Use of Eculizumab to Prevent Thrombosis after Renal Transplantation in Patients With a History of Catastrophic Antiphospholipid Antibody Syndrome (CAPS) (ClinicalTrials.gov Identifier: NCT1029587).

3. Completed: Phase II open label study recruiting patients entitled: Eculizumab for Prevention and Treatment of Kidney Graft Reperfusion Injury (ClinicalTrials.gov Identifier: NCT01756508)

Task force recommendations

Recommendations for clinicians

Anti-complement therapy should be considered in cases of CAPS and refractory microangiopathic disease. Costs are an important limitation that should be taken into account in the decision-making process. Larger clinical trials are required to confirm the findings reported from anecdotal reports and case series. In the meantime, targeting complement is a reasonable alternative that should be considered in selected cases.

Clinical research agenda

Complement inhibition therapy is a promising strategy for patients with microangiopathic APS manifestations. New therapies for targeting complement, including alternative pathway inhibitors, inhibitors of C3, mAbs or small molecules that block C5aR or C5 cleavage, and inhibitors that can be localized to areas of inflammation, may provide alternative therapies for APS.

Treatments based on peptides of beta-2-glycoprotein I

Beta-2-GPI (β2GPI) is an antithrombotic plasma protein that has five domains. Proposed new therapies are based either on the C-terminal Domain V or the N-terminal Domain I. This is because the currently accepted theory for pathogenesis of APS proposes that most pathogenic aPL bind β2GPI via Domain I and the aPL-β2GPI complex then binds to phospholipids on cell surfaces via Domain V leading to thrombosis.¹⁶⁴

Compared to many other potential novel therapeutic agents for APS (e.g. direct acting anticoagulants and anti-complement drugs), these β2GPI-targeted peptides are less likely to be developed as therapies for other medical conditions. Thus, progress in this field is likely to be a little slower than for more widely applicable agents. There are three major challenges to be overcome in showing that peptide-based therapies have potential utility in APS. These challenges are:

• Prove that the peptide blocks binding in-vitro

• Prove that it blocks pathogenic effects of aPL from patients with APS in-vivo using animal models

• Overcome the issue of short in-vivo half-life of peptide agents.

None of the agents tested so far has fully overcome all these challenges.

Peptides that target domain V

The aim of these agents is to block Domain V from binding to cell surfaces. Early work utilised the peptide TIFILFCCSKEKRKKKQAAT, which is derived from cytomegalovirus and has homology to Domain V. In comparison to a control peptide, TIFI reduced binding of β₂GPI to endothelial cells¹⁶⁵ and of IgG from patients with APS (APS-IgG) to human trophoblast in-vitro.¹⁶⁶ Furthermore, TIFI reduced the ability of human aPL to stimulate thrombosis¹⁶⁵ or fetal loss¹⁶⁶ in mouse models. However, no new work on TIFI has been presented for some time.

Kolyada and colleagues have developed a dimer of the A1 domain of the apolipoprotein E receptor (ApoER2). This domain binds Domain V of β₂GPI, so that the dimer acts as an inhibitor preventing β₂GPI from binding to cells. In a series of papers this group showed that A1-A1 blocks binding of β₂GPI-anti-β₂GPI complexes to cardiolipin,¹⁶⁷ and inhibits the induction of thrombosis by laser both in autoimmune (NZW x BXSB) F1 mice and non-autoimmune BALB/c mice infused with human APS-IgG.¹⁶⁸ In the latter experiment 84% of A1-A1 was lost from the blood of the mice within an hour. More recently, the group achieved longer-term delivery of A1-A1 in (NZW x BXSB) F1 mice by implanting a subcutaneous osmotic pump.¹⁶⁹ This led to a reduction in blood pressure in the mice – though this method may not be acceptable to patients. Unlike TIFI, there is currently no evidence that A1-A1 inhibits aPL-induced fetal loss in mice.

Peptides that target domain I

A group at University College London, UK have used recombinant Domain I produced by bacterial expression, rather than a peptide. Initially wild-type Domain I and a series of point mutants were expressed.¹⁷⁰ These different products varied in ability to block binding of APS-IgG to β₂GPI. In the same mouse model previously used in the TIFI experiments, wild-type Domain I and a high-binding mutant inhibited the ability of APS-IgG to promote thrombosis whereas a mutant with no binding had no such inhibitory effect.¹⁷¹ Recombinant Domain I has now been PEGylated in an effort to increase its in-vivo half-life and reduce potential immunogenicity.¹⁷² A potential disadvantage of PEGylation is reduction of the biological activity of the PEGylated molecule. McDonnell et al, however, have shown that PEGylated Domain I retains the ability of non-PEGylated Domain I to block binding of APS-IgG to β₂GPI in-vitro and the ability to block thrombosis induced by APS-IgG in-vivo.¹⁷² The PEGylated product can be produced at 95% purity, but its half-life in-vivo has not yet been demonstrated and there have been no experiments to investigate whether either PEGylated or non-PEGylated Domain I blocks APS-IgG-induced fetal loss. Table 1 summarises and compares evidence for peptide therapies based on Domain I and Domain V.

Table 1.

Summary and comparison of evidence for peptide therapies based on Domain I and Domain V.

	Domain I-based treatment	Domain V-based treatments
Agents tested	Only recombinant Domain I	Both TIFI and A1-A1 dimer
Blocks binding in vitro	Yes – blocks binding of APS-IgG to β2GPI	Yes – blocks binding of β2GPI to cells or cardiolipin.
Blocks thrombosis induced by APS-IgG in vivo	Yes	Yes
Blocks fetal loss induced by APS-IgG in vivo	No	Yes (TIFI only)
Measures to increase half-life in vivo	Yes – by PEGylation	Not for TIFI. A1-A1 has not been modified chemically, but has been administered by a subcutaneous osmotic pump

Task force recommendations

Recommendations for clinicians

It is premature to make recommendations regarding the use of peptides of β2GPI for APS.

Clinical research agenda

The key research aims are to take one or more of these agents forward to formal pharmacokinetic and toxicology studies, then to a first-in-man study.

Potential new players

In pursuit of therapies beyond anticoagulants, APS investigators are increasingly pursuing preclinical and clinical studies with anti-inflammatory and immunomodulatory agents.

Coenzyme Q10

An example of such an approach is a recent APS clinical trial utilizing coenzyme Q10 (CoQ10).¹⁷³ CoQ10 participates as an electron carrier in mitochondrial and other membranes, with adequate CoQ10 levels seemingly protecting cells from protein oxidation and lipid peroxidation. In the general population, CoQ10 supplementation decreases the production of proinflammatory cytokines in the context of heart failure and coronary disease.¹⁷⁴ The APS CoQ10 trial arose from an earlier preclinical study in which the same team demonstrated anti-inflammatory effects (less oxidative stress, less mitochondrial dysfunction) when CoQ10 was added ex vivo to peripheral blood cells of patients with APS.¹⁷⁵ For the clinical trial, 36 patients with APS were randomized to receive ubiquinol (reduced CoQ10, 200 mg/day) or placebo for one month; approximately 90% of subjects completed the study.¹⁷³ Among other positive effects, ubiquinol improved endothelial function and decreased monocyte expression of prothrombotic mediators.¹⁷³ Furthermore, ubiquinol ameliorated NET release by neutrophils, while also downregulating neutrophil peroxides.¹⁷³ The authors pointed out that in the absence of clinically significant side effects, and given potential therapeutic benefits, ubiquinol might act as a safe adjunct to standard therapies in APS.¹⁷³

Adenosine receptor agonists

It has recently been reported that the neutrophils of APS patients have a reduced threshold for the release of NETs (prothrombotic tangles of DNA, histones, and granule-derived proteins expelled from dying neutrophils).¹⁷⁶ NETs are required for APS-potentiated thrombosis in at least one human/mouse chimeric model of APS.¹³ Furthermore, profiling of APS neutrophils has identified novel therapeutic targets in APS,¹⁷⁷ including surface adhesion molecules.¹⁷⁸ Given recent evidence suggesting that the second messenger cyclic AMP (cAMP) may suppress NET release in some contexts,^179,180 a preclinical study hypothesized that activation of surface adenosine receptors (which trigger cAMP formation in neutrophils) might mitigate the thrombotic manifestations of APS.¹⁸¹ Indeed, selective agonism of the adenosine A_2A receptor (with CGS21680) appeared to be highly effective in suppressing antiphospholipid antibody-mediated NET release from control neutrophils, as well as spontaneous NET release from APS neutrophils.¹⁸¹ In vivo, CGS21680 reduced thrombosis in the inferior vena cavae of both control mice and mice administered aPL.¹⁸¹

Adenosine potentiation

The antithrombotic medication dipyridamole is known to potentiate adenosine signaling by increasing extracellular concentrations of adenosine and also by interfering with the breakdown of cAMP. Similar to CGS21680, dipyridamole appeared to suppress aPL-mediated NETosis (in adenosine A_2A receptor-dependent fashion) and to mitigate venous thrombosis in mice.¹⁸¹ While dipyridamole has never been systematically studied in patients with APS, drugs with similar adenosine-amplifying properties such as defibrotide¹⁸² and dilazep¹⁸³ have been reported as effective in case reports and preclinical models.

Another recent study involved preclinical work in the area of aPL-induced pregnancy morbidity.¹⁸⁴ Catabolism of extracellular ATP to adenosine by the cell-surface enzymes CD39 and CD73 has anti-inflammatory and antithrombotic effects in a number of contexts.^185,186 In a model involving passive transfer of human antiphospholipid antibodies into pregnant mice, pregnancy morbidity was exacerbated in mice with reduced ability to generate extracellular adenosine due to deficiency of either CD39 or CD73.¹⁸⁴ In the absence of efficient adenosine generation, the placental decidua demonstrated increased tissue factor expression and complement deposition, as well as elevated oxidative stress and inflammatory cytokines.¹⁸⁴ The potential for adenosine-mediated signaling to mitigate both thrombosis and pregnancy morbidity in APS-relevant models is intriguing and seems worthy of further investigation.

Other novel approaches

What other approaches may be on the horizon? One potential area of interest is targeting antibody-producing cells such as plasmablasts and longer-lived plasma cells. While agents that impact B cells via CD20 or the BAFF/BLyS receptor have not consistently demonstrated reduction in circulating aPL,^{113,114,128,129} it must be noted that these strategies do not directly impact antibody-producing cells. The potential utility of direct plasma cell agents in APS (for example, anti-CD38, as is currently employed for multiple myeloma) was recently emphasized by a preclinical study characterizing lymphocyte subsets of patients with primary APS.¹⁸⁷ Elegant ex vivo experiments revealed that aPL were still robustly produced by peripheral-blood cells depleted of CD20+ B cells, but not when CD20- CD19+ B cells (i.e., CD38-positive plasmablasts) were depleted.¹⁸⁷

Another target area is the development of anti-interferon therapies, as are being pursued for treatment of SLE. While the so-called interferon signature is classically associated with certain rheumatic diseases such as lupus and dermatomyositis, a number of groups have recently detected elevated levels of type I interferons in primary APS,^188,189 including potential associations with triple positivity and pregnancy morbidity.¹⁹⁰ Whether neutralization of interferons might mitigate any of the thrombotic -- or perhaps more likely non-thrombotic -- manifestations of APS awaits further study.

A further area of potential interest is anti-FcRn targeted therapies that are being used in various autoimmune IgG driven diseases. Efgartigimod is a human IgG1 antibody Fc-fragment, a natural ligand of the neonatal Fc receptor (FcRn) that blocks FcRn, preventing IgG recycling, and causing targeted IgG degradation. In a Phase 2 study in 38 patients with primary immune thrombocytopenia predominantly refractory to previous lines of therapy, efgartigimod was well tolerated and had a favorable safety profile. It induced a rapid reduction of total IgG levels (mean change from baseline up to 63.7%), associated with clinically relevant increases in platelet counts: 46% patients on efgartigimod vs 25% on placebo achieved a platelet count of ≥50 × 10⁹/L on at least two occasions, and 38% vs 0% achieved ≥50 × 10⁹/L for at least 10 cumulative days. There was also a reduced proportion of patients with bleeding. The authors concluded that FcRn antagonism warrants further evaluation as a novel therapeutic approach in ITP.¹⁹¹ This approach might be potentially useful in APS.

Task force recommendations

Recommendations for clinicians

None of the agents discussed in this section should be formally recommended at this time.

Clinical research agenda

The Task Force strongly supports continued preclinical and clinical studies that leverage mechanistic endpoints such as inflammatory biomarkers and aPL levels. The most promising agents should then be considered for large-scale multicenter trials.

Antiphospholipid syndrome treatment trends task force conclusions

The management of APS is complex and challenging. The importance of the goal to provide optimal care is highlighted by the potentially severe and life-threatening complications that APS patients can experience, as a result of thrombotic, obstetric and non-criteria manifestations. The lack of clinical trials in APS patients necessitates empirical approaches to try to manage the multiple manifestations of this disorder. Continued work on improving understanding of the pathophysiology of APS is an essential prerequisite to providing a basis and rationale for the development of optimal therapeutic approaches. For meaningful advances in clinical management, a focus on international registries (such as that of APS ACTION), prospective cohort studies and RCTs is essential. The Task Force does not recommend the use of statins in pregnancy in view of their current regulatory status, and the lack of RCTs. APS studies, which need to be appropriately designed and powered, and capture the clinical and laboratory heterogeneity of the syndrome, could eventually provide sufficient high-quality data to underpin evidence-based management. To enable the initiation and development of appropriate clinical studies, there needs to be a focus on devising suitable outcome measures, including a disease activity index, an optimal damage index, and a specific quality of life index.

ORCID iDs

Hannah Cohen https://orcid.org/0000-0003-2032-390X

Ali Duarte-Garcia https://orcid.org/0000-0003-1749-5719

David A Isenberg https://orcid.org/0000-0001-9514-2455

Maria G Tektonidou https://orcid.org/0000-0003-2238-0975

Scott C Woller https://orcid.org/0000-0002-2522-2705