Antiphospholipid Syndrome Nephropathy and Other Thrombotic Microangiopathies Among Patients With Systemic Lupus Erythematosus

Abstract

Antiphospholipid syndrome (APS) and other causes of thrombotic microangiopathy (TMA) negatively impact the renal outcomes of patients with systemic lupus erythematosus (SLE) and lupus nephritis. Here we review the diagnosis and management of occlusive renal vascular lesions due to APS and other TMAs, with a focus on patients with SLE and lupus nephritis. The presence of a thrombotic event, unexplained hypertension, thrombocytopenia, or haemolytic anemia should prompt consideration for TMA syndromes. The differential diagnosis of a TMA in a patient with SLE includes APS, thrombocytopenic purpura, complement-mediated or infection-associated haemolytic uremic syndrome, drug-mediated TMA (particularly due to calcineurin inhibitor toxicity), and malignant hypertension. Treatment of APS with a documented thrombotic event focuses on anticoagulation to reduce the risk for further thrombotic events. Treatment of classic presentations of thrombocytopenic purpura and hemolytic uremic syndrome in the SLE population is the same as in patients without SLE. Treatment of APS nephropathy or TMA when it is diagnosed by biopsy with concomitant lupus nephritis presents a challenge to clinicians because there is no clear standard of care. Small and retrospective studies suggest potential benefit of complement inhibition, mammalian target of rapamycin (mTOR) inhibition, B cell depleting therapy, and plasma exchange therapy for patients with lupus nephritis and TMA, and prospective investigation of these therapies should be a research priority.

Nephrologists routinely care for patients with systemic lupus erythematosus (SLE). In addition to expertise in managing lupus nephritis, it is important to be aware of the diagnosis and management of occlusive renal vascular lesions due to antiphospholipid syndrome (APS) and other thrombotic microangiopathies (TMAs), which may occur concurrently with or independently from lupus nephritis (Fig 1 and Table 1). Among patients with lupus nephritis, the co-occurrence of APS nephropathy or other TMA impacts patient prognosis and may prompt a change in treatment. Further investigation into the pathogenesis of APS nephropathy as well as how SLE intersects with the spectrum of TMA syndromes may allow for tailored therapy to improve patient outcomes.

Figure 1.

Spectrum of the clinical syndromes of TMA and APS. Clinical thrombotic microangiopathy: microangiopathic hemolytic anemia with thrombocytopenia and organ dysfunction. Of note, CNI toxicity and malignant hypertension may occasionally co-occur with SLE, which is not represented in this figure. Abbreviations: APS, antiphospholipid syndrome; CM-HUS, complement-mediated hemolytic uremic syndrome; CNI, calcineurin inhibitor; IA-HUS, infection-associated hemolytic uremic syndrome; SLE, systemic lupus erythematosus; TMA, thrombotic microangiopathy; TTP, thrombotic thrombocytopenic purpura.

Table 1.

Clinical Syndromes of TMA in Accordance With 2017 Consensus Terminology

Disorder	Overview
TTP	• MAHA with moderate to severe thrombocytopenia and organ dysfunction • Confirmed by <10% ADAMTS-13 activity • Kidney injury is mild to moderate
Congenital TTP	• TTP with no inhibitory autoantibody • Confirmed with sequencing of the ADAMTS-13 gene
Primary immune-mediated TTP (primary iTTP)	• TTP with ADAMTS-13 autoantibody without an identified risk factor • Accounts for the majority of cases with TTP
Secondary immune-mediated TTP (secondary iTTP)	• TTP with ADAMTS-13 autoantibody and a risk factor (SLE; connective tissue disease; HIV or CMV infection; immune-mediated drug reaction)
HUS	• MAHA with mild to moderate thrombocytopenia (or thrombocytopenia may be absent), with organ dysfunction. ADAMTS-13 may be normal or moderately low, but not <10%. • Kidney injury is moderate to severe, often dialysis dependent • Hypertension is common
IA-HUS	• HUS with identified infectious precipitant, including Escherichia coli O157:H7, Salmonella, Shigella, and Campylobacter
CM-HUS	• HUS caused by a defect in regulation of the alternative complement pathway
Other TMA syndromes
Drug-mediated TMA	• Immune reaction: Triggers a secondary iTTP with presence of an ADAMTS-13 autoantibody. The best evidence exists for quinine although numerous others implicated • Dose-related reaction: No autoantibody is present. Often indolent, often kidney-limited TMA. Implicated drugs include calcineurin inhibitors, vascular endothelial growth factor inhibitors, and oxymorphone (in IV abuse)
Pregnancy-associated TMA	• Occurs in the contextof pre-eclampsia and hemolysis with elevated liver enzymes and low platelets
Pneumococcal-associated HUS	• Occurs in the context of streptococcal pneumonia infection, usually pneumonia, and typically occurs in children • Patients have Coombs-positive hemolytic anemia
Very rare genetic causes of TMA	• Cobalamin C disease and coagulation-mediated TMA

Abbreviations: CM-HUS, complement-mediated hemolytic uremic syndrome; CMV, cytomegalovirus; HUS, hemolytic uremic syndrome; IA-HUS, infection-associated hemolytic uremic syndrome; IV, intravenous; MAHA, microangiopathic hemolytic anemia; SLE, systemic lupus erythematosus; TMA, thrombotic microangiopathy; TTP, thrombotic thrombocytopenic purpura.

Scully and colleagues 2017¹; George and Nester 2014².

ANTIPHOSPHOLIPID SYNDROME IN SYSTEMIC LUPUS ERYTHEMATOSUS

Definitions and Clinical Presentation

APS refers to a syndrome of vascular thrombosis and/or pregnancy morbidity, with persistently present antiphospholipid antibodies. Antiphospholipid antibodies refer to lupus anticoagulant, anticardiolipin antibody (aCL), and anti-B2 glycoprotein-1 antibody (anti-B2GPI) (see Fig 2), and to be considered persistent they must be present on 2 or more occasions at least 12 weeks apart. Pregnancy morbidity refers to 1 or more unexplained fetal deaths at or beyond the 10th week of gestation; 1 or more premature births before 34 weeks of gestation due to pre-eclampsia or placental insufficiency; or 3 or more unexplained consecutive abortions before the 10th week of gestation.³ Vascular thrombosis in APS is present without significant inflammation in the vessel wall. The thromboses may occur in arteries or veins and may occur in large vessels (deep vein thrombosis, pulmonary embolism, renal vein thrombosis, and renal artery stenosis) or in small vessels in any organ. Frequently involved organs include skin (livedo reticularis), central nervous system (transient ischemic attack, stroke, and cognitive dysfunction), heart (heart valve lesions, rarely coronary artery disease), and kidney (see manifestations below).³ In addition to the organ-specific clinical findings, patients may present with thrombocytopenia, hemolytic anemia, or unexplained prolongation of the activated partial thromboplastin time.⁴ Additionally, hypertension is a frequent finding.⁵

Figure 2.

Antiphospholipid antibody testing. Units for IgG tests are GPL, and for IgM tests are MPL. Abbreviations: aPTT, activated partial thromboplastin time; drVVT, dilute Russell’s viper venom time; ELISA, enzyme-linked immunosorbent assay. Sources: Garcia 2018⁵ and Pengo 2009.⁶⁴

Catastrophic APS refers to rapidly progressing APS with development of thrombotic events in 3 different organs within 1 week.⁶ This occurs in <1% of patients with APS but has a high mortality rate of 50%. Forty percent of patients with catastrophic APS have SLE. Renal manifestations are present in 71% of cases of catastrophic APS.⁷

Kidney Manifestations of Antiphospholipid Syndrome

The most common kidney manifestation of APS is APS-associated nephropathy, referring to noninflammatory occlusion of the small vessels of the kidney; this occurs in 63%−67% of patients with SLE and APS.^8,9 APS nephropathy was first described in 1999 in patients with primary APS,¹⁰ and in 2002 was described in patients with SLE.⁹ Renal pathologic findings are described below. Patients may be asymptomatic with only mild proteinuria, or may present with hypertension, nephrotic-range proteinuria, or an active urinary sediment.⁵ Biopsy is required for diagnosis of APS nephropathy, as it cannot be reliably differentiated from lupus nephritis without tissue diagnosis.

Other kidney manifestations of APS in patients with lupus nephritis include renal vein thrombosis, renal artery infarction, and renal artery stenosis.^5,11 Renal vein thrombosis typically presents with nephrotic-range proteinuria and edema.¹² Renal artery infarction typically presents with severe lumbar back pain accompanied by acute kidney injury and hypertension.¹¹ Renal artery stenosis presents with severe hypertension; the location of the stenosis within the main renal artery is classically distal although it may be proximal in comorbid atherosclerotic disease.^11,13

Impact on Renal Outcomes in Lupus Nephritis

Compared with patients with lupus nephritis alone, Gerhardsson and colleagues¹⁴ found that patients who have renal biopsy findings consistent with APS nephropathy concomitant with lupus nephritis are more likely to progress to end-stage kidney disease (ESKD) (odds ratio [OR] 5.8, 95% confidence interval [CI] 1.7–19.7).

The impact of positive serum antiphospholipid antibodies alone (absent a histologic diagnosis of APS nephropathy) on renal outcomes is unclear. In a series of 111 patients with lupus nephritis (inclusive of a subset who likely had APS nephropathy), Moroni and colleagues¹⁵ found that the presence of persistently positive antiphospholipid antibodies was an independent predictor of chronic renal insufficiency (RR 2.2, 95% CI 1.06–4.48). However, Parodis and colleagues¹⁶ examined a cohort of 63 patients with biopsy-proven lupus nephritis without APS nephropathy on biopsy, and did not find a difference in long-term glomerular filtration rate (GFR) or development of CKD between patients with and without antiphospholipid antibodies.

Epidemiology

Among patients with SLE, 20%−30% have persistent anti-phospholipid antibodies,¹⁷ and within that group 70% meet diagnostic criteria for APS.¹⁸ Of those with APS in the context of SLE, 22%−32% have APS nephropathy.^8,9 Antiphospholipid antibody positivity does not appear to differ between patients with lupus nephritis compared to patients with nonrenal SLE.¹⁶ In patients with an autoimmune disease including SLE who meet laboratory criteria for APS but do not have a known thrombosis, the annual risk of a first thrombosis is up to 5%.¹⁹ In several biopsy series of patients with lupus nephritis, prevalence of coexisting APS nephropathy is reported between 1% and 14%.^9,14,20,21

Unlu and colleagues¹⁸ described an international cohort of patients with both SLE and persistently positive anti-phospholipid antibodies (not all of these patients had a thrombotic or pregnancy event meeting diagnostic criteria for APS); 77% were female, 71% white, 9% Latin-American, 12% Asian, and 6% black.

Renal Pathology

Histologically, renal lesions in APS with or without SLE are identical. Early lesions display features of TMA with frequent fibrin thrombi in glomeruli and arterioles. Late lesions display fibrointimal hyperplasia with luminal obliteration and organized thrombi with recanalization in small caliber arteries (Fig 3), focal cortical atrophy, and tubular thyroidization.^3,9

Figure 3.

An interlobular artery with organizing thrombus and recanalization (arrows), Masson’s trichrome stain, and magnification, ×400.

Renal Transplant and Antiphospholipid Syndrome

Graft thrombosis is the main risk for patients with APS after a renal transplant.^22,23 A recent meta-analysis of 22 cohort studies found a higher prevalence of thrombosis (10.4% vs 1.7%) and TMA (10.2% vs 0%) in renal allograft in patients with the presence of antiphospholipid antibodies compared to none.²³ They also noted an association with graft loss, but not rejection. Patients treated with anticoagulation were less likely to develop thrombosis.²³ Gołębiewska and colleagues²⁴ reported that in their cohort of 19 patient with ESRD due to lupus nephritis, patients with APS and lupus nephritis had worse outcomes after transplant compared to patients with lupus nephritis alone, in terms of allograft loss, delayed graft function, and acute rejection.

Presence of aCL antibodies without APS may be associated with estimated GFR decline in the year after transplant.²⁵ Recent studies have also shown a possible association between IgA anti-B2GP1 antibodies with graft thrombosis and loss.²⁶ Serrano and colleagues²⁷ suggest that it may be the presence of circulating immune complexes of IgA bound to B2GP1 and not just the antibody presence that is a predictor of graft thrombosis. Of note, IgA anti-B2GP1 is more common in patients with SLE and APS than in patients with isolated APS.¹⁸

Very promisingly, treatment with mammalian target of rapamycin (mTOR) inhibitors as part of the immunosuppressive regimen appears to be beneficial for allograft survival; Canaud and colleagues²⁸ found that graft survival at 12 years was 70% in those treated with sirolimus vs 11% in those not on mTOR inhibitors.

Pathogenesis

The pathogenesis of APS is complex and emerging evidence is shedding new light which may yield therapeutic targets.

A prime target of autoantibodies in APS is B2GPI, a protein which circulates in plasma and upregulates coagulation when activated. When autoantibodies bind to and activate B2GPI, the result is increased binding of B2GPI to phospholipid cell surfaces including endothelial cells, monocytes, and neutrophils, leading to increased expression of prothrombotic factors (particularly tissue factor, an activator of the extrinsic coagulation pathway, and E-selectin, an adhesion molecule).^3,29 B2GPI also interacts with von Willebrand factor, leading to platelet activation.²⁹ Finally, increased B2GPI activity also inhibits regulatory checks (including tissue factor pathway inhibitor; annexin A5, which is stabilized by hydroxychloroquine; and potentially tissue plasminogen activator).^3,29,30

Of the clinically measured antiphospholipid antibodies, anti-B2GP1 has a direct interaction with B2GPI, whereas aCL interacts with B2GPI in a less clear way and may require a cofactor.³¹ There is debate about the utility of using the aCL in the diagnostic criteria for this reason.²⁹

In APS, the presence of the autoantibodies does not consistently cause thrombotic events, and some type of endothelial cell priming seems to be necessary for the binding of the autoantibodies to result in thrombosis—this is referred to as the second hit. The second hit may occur via endothelial cell injury (infection, surgery, immobilization) or by increased oxidative stress which increases binding of the autoantibodies to targets on B2GPI (either endogenous oxidative stress due to genetic factors, or exogenous such as cigarette smoking).²⁹

Beyond the activation of B2GP1, 2 other pathogenic concepts have been recently described. Canaud and colleagues²⁸ demonstrated that antiphospholipid autoantibodies isolated from patients with known APS activate the mTOR pathway in cultured vascular endothelial cells, and supported their hypothesis that this pathogenic mechanism may be involved in thrombotic events by demonstrating that kidney transplant recipients who were treated with mTOR inhibitors had dramatically better graft survival than patients not treated with mTOR inhibitors. An additional promising and emerging body of knowledge concerns the role for activation of the classical complement pathway in APS thrombotic events; in mouse models the classical complement pathway has been demonstrated to be a necessary intermediary for thrombosis to occur.^29,32,33

Management of Antiphospholipid Syndrome

The 2019 guidelines issued by the European League Against Rheumatism (EULAR) are an excellent evidence-based resource for clinicians treating patients with APS. The mainstay of treatment of APS with a diagnosed thrombotic event is anticoagulation. In the acute setting, patients are treated with unfractionated or low molecular weight heparin, then transitioned to long-term therapy with warfarin with a target INR (international normalized ratio) of 2–3.³ High-intensity warfarin therapy (INR goal 3–4) was not superior to moderate-intensity warfarin (INR goal 2–3) in 2 randomized trials.^34,35 Typically, warfarin is continued indefinitely for unprovoked venous thromboembolism (VTE), as the risk of recurrent thrombosis is 29% per year without treatment, although the risk of recurrent thrombosis may be smaller in patients with provoked VTE or with normalizing antiphospholipid antibody titers.^3,36 Patients with provoked VTE and high-risk antiphospholipid antibody profile or other risk factors for recurrence may be considered for indefinite anticoagulation as well, but lower risk patients with provoked VTE may be treated with a duration of 3–6 months of anticoagulation.³⁷ Low molecular weight heparin is an alternative to warfarin for long-term anticoagulation. There is currently limited but emerging data regarding the use of direct oral anticoagulants in APS,³ and EULAR recommends using a direct oral anticoagulant only for patients who have trouble achieving INR goal; additionally, patients with triple-positive antiphospholipid antibodies should not use rivaroxaban because it has been associated with increased risk of recurrent thrombotic events in that population.³⁷ Thrombocytopenia is not protective against recurrent thrombosis in APS and should not prompt withdrawal of anticoagulation in cases of mild to moderate thrombocytopenia.³⁸

In addition to anticoagulation, all patients with APS should avoid other risk factors for thrombosis, including immobility, smoking, and estrogen-containing contraceptives.³

Patients with concomitant lupus nephritis should receive hydroxychloroquine and immunosuppressive management as warranted for the lupus nephritis.³⁹

For patients who experience recurrent thrombosis despite warfarin, add-on therapies include hydroxychloroquine, statins, low dose aspirin, or a change in therapy from warfarin to heparin; none of these strategies has high-quality evidence and further investigation is needed.³ Investigational approaches include adenosine 2A receptor agonist and hydroxychloroquine.³ Apart from reducing platelet activation, hydroxychloroquine potentially reduces recurrent thrombosis when used as an add-on therapy to warfarin through a variety of mechanisms.⁴⁰ A phase III clinical trial on the effect of hydroxychloroquine in obstetric APS is currently underway.⁴⁰

Interestingly, when heparin is used in APS it exerts its effects not only through potentiation of antithrombin III, but also in an additional disease-specific mechanism of action. Guerin and colleagues⁴¹ showed that heparin binding with B2GPI resulted in a cleaved form which was ineffective in recognizing phospholipids. This resulted in a decrease in the thrombotic activity of anti-B2GPI antibodies in APS.

The benefit of aspirin as primary prevention against thrombosis in patients with antiphospholipid antibodies is unclear. However, a meta-analysis did show a potential benefit from aspirin in a subgroup analysis patients with SLE to protect against first thrombosis (OR 0.55, 95% CI 0.31–0.98).⁴² EULAR does recommend that patients with SLE with a high-risk antiphospholipid antibody profile receive low-dose aspirin for primary prevention, and that low-dose aspirin may be considered in patients with SLE and a low-risk antiphospholipid antibody profile.³⁷

According to the American College of Obstetricians and Gynecologists, pregnant patients with a history of a thrombotic event and APS should be treated with prophylactic heparin during pregnancy and 6 weeks postpartum. The management of a pregnant patient with obstetric-only APS and no history of thrombosis is unclear. Although some suggest low dose aspirin use only, the American College of Obstetricians and Gynecologists recommendation is prophylactic heparin with the low-dose aspirin, especially if there is a history of pregnancy loss.⁴³

Treatment of catastrophic APS usually involves a combination of corticosteroids, intravenous immune globulin, plasma exchange, and anticoagulation.³⁷ Rituximab or eculizumab may be considered for refractory catastrophic APS.³⁷ Early treatment and supportive therapies such as dialysis and mechanical ventilation are critical. Only case studies exist regarding management of catastrophic APS and there are no randomized controlled trials due to rarity of the condition.

OTHER THROMBOTIC MICROANGIOPATHIES IN SYSTEMIC LUPUS ERYTHEMATOSUS

Definitions and Clinical Presentation

TMA refers to both a clinical syndrome and a histopathologic term. The clinical syndrome refers to microangiopathic hemolytic anemia with thrombocytopenia and organ injury, whereas the pathologic term refers to occlusive microvascular or macrovascular disease with characteristic histopathologic findings.^1,2 A tissue diagnosis of TMA may or may not be accompanied by microangiopathic hemolytic anemia with thrombocytopenia; we will use the term kidney-limited TMA to refer to a kidney biopsy showing TMA without the presence of microangiopathic hemolytic anemia with thrombocytopenia. The TMA syndromes are summarized in Table 1. Among various causes of TMA, the literature describes patients with SLE who are affected by APS, immune-mediated thrombocytopenic purpura (iTTP, this is often called acquired TTP and is differentiated from hereditary TTP which is much less common), complement-mediated hemolytic uremic syndrome (CM-HUS, often called atypical HUS), drug-mediated TMA (particularly due to calcineurin inhibitor toxicity), and malignant hypertension. Among these, iTTP tends to present with milder kidney injury and more prominent thrombocytopenia, whereas CM-HUS tends to present with more severe kidney injury and hypertension with milder thrombocytopenia.² Drug-mediated TMA due to calcineurin inhibitor toxicity typically has a more indolent presentation with gradual onset of hypertension and gradual decline in kidney function; it may often present as kidney-limited TMA.

Unlike TTP and HUS, which are frequently diagnosed via noninvasive evaluation described below, kidney-limited TMA requires kidney biopsy for diagnosis.

When TTP is suspected, both in patients with SLE and in the general population, initial evaluation should include an assay of ADAMTS-13 from a sample taken prior to any plasma therapy; this assay should include measurement of ADAMTS-13 activity, mixing studies to identify a functional inhibitor, and anti-ADAMTS-13 IgG.¹ Depending on the availability of the ADAMTS-13 activity level and the time to result, treatment with plasmapheresis often needs to be initiated based on clinical suspicion prior to the results becoming available. Infection-associated HUS (IA-HUS) can be excluded with PCR stool testing for Shiga toxin; diarrhea alone does not differentiate between IA-HUS and CM-HUS.² If concern exists for CM-HUS, testing for abnormalities in the alternative complement pathway can be undertaken. This may include, often in a tiered fashion, mutational analysis of described mutations (CFH, CFI, MCP, C3, CFB, CFHR1–5, DGKE, and copy number variants in the CHF-CFHR region), functional assays of complement activity, testing for autoantibodies to Factor H and Factor I, and testing for complement protein deficiencies.⁴⁴ In the acute setting, the diagnosis of CM-HUS must be made by excluding alternate causes of TMA and treatment should commence immediately, both because assessment of the alternative complement pathway may take weeks to yield results and because a mutation is not identified in every affected patient; however, the mutational analysis may help guide long-term patient management.⁴⁵

Impact on Renal Outcomes in Lupus Nephritis

Patients with TMA in addition to lupus nephritis have worse renal prognosis than patients with isolated lupus nephritis. Wu and colleagues⁴⁶ found that 33% of patients with concurrent TMA and lupus nephritis experienced doubling of serum creatinine or progression to ESKD compared to 3% with isolated lupus nephritis. Song and colleagues²⁰ found an hazard ratio of 2.77 (95% CI 1.01–7.62) for the same composite outcome comparing presence vs absence of TMA in lupus nephritis (the renal TMA group included 2/36 patients with APS nephropathy).

Epidemiology

Among biopsy cohorts of renal biopsies of patients with a diagnosis of lupus nephritis, the presence of concomitant TMA varied between 1% and 24% of patients.^20,21,46,47 In several series of patients with both lupus nephritis and histopathologic TMA, 79%−81% had isolated kidney TMA, whereas 1%−9% met criteria for TTP-HUS (these series did not distinguish among the TTP and HUS syndromes) and 1%−13% had APS.^20,21,48

Features of TMA may be superimposed on any class of lupus nephritis although it is most commonly noted with Class IV, with 83% of cases occurring in Class IV or Class IV + V lupus nephritis.^20,21

Letchumanan and colleagues⁴⁹ found that when compared to patients with TTP without SLE, patients with TTP associated with SLE had higher mortality (63% vs 50%), despite being younger and being treated more aggressively.

Renal Pathology

The findings of TMA in a renal biopsy can be histologically divided into early and late (chronic) lesions.

Early (Acute) Lesions.

Glomeruli display fibrin or platelet thrombi affecting predominantly the small vessels including the preglomerular arterioles (Fig 4). Mesangial changes with mesangiolysis, fragmented erythrocytes, and fibrinoid necrosis are also noted (Fig 5). Electron microscopy characteristically shows a wide subendothelial lucent expansion with electron lucent material and loss of endothelial fenestrations. Arteries and arterioles show mucoid intimal expansion, fibrinoid necrosis, and fragmented erythrocytes (Fig 6). Immunofluorescence studies show fibrin deposition in vessels and glomeruli reflecting the presence of fibrinoid necrosis or fibrin thrombi. Nonspecific IgM and C3 deposits are also sometimes observed due to insudation of plasma proteins. Concomitantly, lupus nephritis shows the presence of immune deposits with a full house pattern. Positive C4d staining in the glomeruli and vessels has been suggested to be a predictive marker of a concomitant/developing TMA.^50,51 Further studies are warranted for validating the prognostic utility of C4d in TMA.

Figure 4.

Glomerulus with the hilar vessel displaying fibrin thrombus formation accompanied by endocapillary cell proliferation and a cellular crescent, Masson’s trichrome stain, and magnification, ×200.

Figure 5.

Glomerulus with fibrinoid necrosis and fragmented erythrocytes. There is concomitant fibrinoid necrosis of the hilar vessel (arrow) with fragmented erythrocytes in the wall, hematoxylin-eosin stain, and magnification, ×200.

Figure 6.

Arteriole displaying fragmented erythrocytes, hematoxylin-eosin stain, and magnification, ×400.

Late (Chronic) Lesions.

Light microscopy shows reduplication of glomerular capillary basement membranes with segments of tuft sclerosis. The arteries and arterioles display fibrous intimal hyperplasia with obliteration of lumina with or without recanalization of thrombi. There is accompanying interstitial fibrosis and tubular atrophy.

Although it is difficult histologically to isolate a particular etiology for a TMA found on renal biopsy, a few clues in the biopsy may suggest an association. HUS primarily affects the glomerular compartment, whereas TTP mostly presents with minimal glomerular involvement. Patients with associated scleroderma renal crisis or malignant hypertension show a vascular predominance with onion skinning and myointimal proliferation of blood vessels.

The histological differential diagnosis of lupus with TMA is primarily a lupus vasculopathy. Lupus vasculopathy sometimes shows similar changes in arterioles with the presence of fibrinoid necrosis; however, the presence of immune complex deposits in the vessels excludes a TMA.

Kidney Transplant and Thrombotic Microangiopathy

TMA in renal allograft has an incidence of 5.6 episodes per 1000 person-years with the highest risk in the first 3 months.⁵² De novo TMA is more common than recurrence of TMA associated with a native kidney disease.⁵³ The most common causes of de novo TMA are antibody-mediated rejection, medications, and viral infections. Recurrent TMA can occur due to TTP, HUS, SLE with or without APS, among others.⁵³ Of note, while mTOR inhibitors are being studied for treatment of APS, they have been associated with the development of TMA. Kidney biopsy is not only essential for the diagnosis of TMA, especially kidney-limited TMA, but can be helpful in determining the etiology of TMA in transplanted kidneys. Treatment is usually directed toward the cause of TMA. For calcineurin inhibitor toxicity a dose reduction may be sufficient.²

Graft and patient outcomes are poor in kidney transplant recipients with HUS as the cause of ESKD. In a cohort of patients with ESKD due to HUS transplanted between 1998 and 2000, 29% of patients had recurrent TMA after transplantation, often within the first 3 months, and survival after recurrent TMA was 50% at 3 years.⁵² Notably, this study was prior to the availability of anticomplement therapy.

A Polish study comparing rheumatic and nonrheumatic diseases found worse long-term renal allograft and recipient survival in patients with rheumatic diseases.⁵⁴

Pathogenesis

The heterogeneity of TMA syndromes precludes a comprehensive review of their pathophysiology here. The pathogenesis of primary immune-mediated TTP has been recently reviewed by Saha and colleagues, and that of complement-mediated HUS was recently reviewed by George and Nester.^45,55 Pertinent to the lupus population, at least 3 mechanisms have been proposed in the literature: production of autoantibodies to ADAMTS-13, activation of the classical complement pathway, and activation of the alternative complement pathway.

Regarding the role of ADAMTS-13 autoantibodies, Yu and colleagues presented a group of 7 patients with lupus nephritis who presented with microangiopathic hemolytic anemia, thrombocytopenia, and had TMA on renal biopsy. They had ADAMTS13 levels at presentation which were moderately low but not <10% as required for a diagnosis of TTP (mean 40% in lupus nephritis with TMA with a range of 34%−55%, mean 69% in lupus nephritis alone, and mean 81% in healthy controls). The presence of anti-ADAMTS-13 autoantibodies was increased in both groups with lupus nephritis (86% in lupus nephritis plus TMA, 18% in lupus nephritis alone, vs 0% in healthy controls). After resolution of anemia and thrombocytopenia following treatment, the ADAMTS-13 levels increased from 40% to 69% in the 7 patients with lupus nephritis plus TMA, and the anti-ADAMTS-13 autoantibodies resolved in 5/6 patients with initial presence of autoantibodies (5/7 of these patients were treated with plasmapheresis). The authors posited that anti-ADAMTS-13 antibodies play a role in TMA syndromes associated with lupus nephritis.⁴⁸

Numerous studies have focused on the role of complement dysregulation in SLE and how this may predispose to TMA physiologies. Evidence for dysregulation of the classical complement pathway in TMA includes a correlation between the intensity of C4d staining in glomeruli with the presence of renal microthrombi and deposition of C5b-9.^50,51 C4d is a stable split product which is present in both classical and lectin pathway activation, but is not present in alternative pathway activation. Further staining is used to differentiate activation of the classical pathway (C1q) and lectin pathway (MBL). Chua and colleagues evaluated markers of complement activation in renal biopsies of a cohort of 42 patients with TMA and found that 90% of the patients with SLE (with and without APS) had markers of classical pathway activation, prominently in the glomeruli. This group found that 88% of all TMAs in the heterogenous cohort showed C4d staining, and the staining pattern differed among etiologies, with glomerular C4d staining being prominent in SLE and IgA nephropathy, and arteriolar staining being prominent in CM-HUS. Notably, staining for terminal complement activation (C5b-9) was present in 75% of the patients with SLE and TMA, suggestive of a potential role for terminal complement inhibition in the management of TMAs in the setting of SLE.⁵¹

Regarding the role of the alternative complement pathway, Park and colleagues⁵⁶ described a case series of 11 patients with both lupus nephritis and CM-HUS; a mutation known to be associated with CM-HUS was identified in 6 of these patients.

SLE is broadly understood to involve widespread dysregulation of both the alternative and classical complement pathway, as evidenced by frequently low serum C3 and C4 levels.⁵⁶ One interesting postulation from Park and colleagues⁵⁶ is that the presence of a complement-amplifying condition such as SLE might be enough to cause a patient with an underlying risk factor, such as a subtle mutation in the alternative complement pathway, to be more likely to develop a disease state of CM-HUS compared to a patient with the same mutation without SLE.

Management of Thrombotic Microangiopathy

Management of TMA varies by clinical syndrome. For patients with drug-induced TMA, withdrawal of the agent or dose reduction in the case of calcineurin inhibitors is the typical treatment.² The acute management of patients with iTTP requires immediate plasmapheresis to clear ADAMTS-13 autoantibodies. This may be followed by immunosuppression to eliminate antibody production; frequently used agents are corticosteroids and rituximab, among others. An important investigational therapy in iTTP is the potential for recombinant ADAMTS-13.^2,55 One case report of a patient with SLE and iTTP refractory to plasmapheresis found benefit from rituximab.⁵⁷ The mainstay of therapy in CM-HUS is anticomplement therapy in the form of eculizumab, a humanized monoclonal antibody which binds to C5 and prevents the formation of the terminal complement complex, which was approved for use in CM-HUS in 2011 and has revolutionized therapy for this disease.

A special challenge for nephrologists is management of lupus nephritis with TMA identified on kidney biopsy in the absence of a clear TTP or CM-HUS syndrome. It is clear that concomitant TMA does have negative effects on the outcomes for patients with lupus nephritis, as discussed above, but there is no clear standard of care for how to manage these patients, both in native kidney disease and in post-transplant TMA. Patients who have predominant lupus nephritis on a kidney biopsy should be treated with immunosuppression, such as corticosteroids and mycophenolate or cyclophosphamide.

A retrospective, nonrandomized cohort study of 70 patients with concomitant lupus nephritis and TMA on kidney biopsy evaluated the role of plasmapheresis in addition to standard treatment for lupus nephritis. Propensity-matched patients who received plasmapheresis in addition to standard immunosuppression for lupus nephritis, when compared to patients who simply received therapy for lupus nephritis, had a higher remission rate (78% vs 11%), a lower treatment failure rate (22% vs 89%), and were less likely to reach a composite endpoint of death, ESKD, doubling of serum creatinine, or treatment failure.⁵⁸

Based on the hypothesis that complement activation is associated with TMA in patients with SLE, eculizumab has gained interest in this population. The case series from Park and colleagues described the above presented 11 patients with lupus nephritis associated with TMA refractory to plasmapheresis, glucocorticoids, and immunomodulatory therapy who were treated with eculizumab. Three patients had concurrent APS and were on anticoagulation. Patients received the CM-HUS approved dosing and 10 patients survived without infection and 5 came off dialysis.⁵⁶ Sciascia and colleagues⁵⁹ reviewed 6 cases of eculizumab use in SLE with renal involvement (5/6 had a TMA), finding improved kidney function, and normalization of serum complement. Kello and colleagues reported on a cohort of 9 patients treated with eculizumab for secondary TMA (7/9 had SLE, 6/9 had APS) and showed improvement in estimated GFR by 25% in half the patients. Two of the 3 patients on dialysis came off dialysis successfully.⁶⁰ Lonze and colleagues⁶¹ reported that 3 patients with APS (2/3 with catastrophic APS) had successful kidney transplantation following pre-emptive treatment with eculizumab.

Anti-CD20 therapy, such as rituximab, in case reports has been beneficial in patients with TMA, especially in the setting of TTP, and APS.^57,62 Sun and colleagues⁶³ reported outcomes in patients with TMA secondary to SLE treated with rituximab. Although the overall survival was higher in patients receiving rituximab, 5 of 7 patients with kidney involvement remained on dialysis.

In summary, TMA in patients with SLE and APS has been treated with plasmapheresis, eculizumab, and rituximab, in addition to hydroxychloroquine and any immunosuppression therapy indicated for concomitant lupus nephritis.³ Mostly case reports and case series exist regarding these therapies and prospective studies are needed.

FUTURE DIRECTIONS

APS and TMA impact the prognosis of patients with SLE and lupus nephritis. However, the understanding of the pathogenesis of these diseases and how SLE interacts with classic presentations of TTP and HUS is evolving. Future research into the molecular mechanisms of disease can help identify targets for treatment and potentially improve mortality. Currently, due to the rarity of these conditions, prospective studies and randomized controlled trials are lacking. Therapies currently undergoing investigation for reducing thrombosis risk in APS include C5 and C5a receptor inhibition, statins, adenosine 2A receptor agonists, and hydroxychloroquine.^3,40 Further research is warranted to determine the efficacy of direct oral anticoagulants in patients with APS. The approval of eculizumab in 2011 was a great leap forward in the management of CM-HUS, including CM-HUS occurring in patients with SLE; however, the role of anticomplement therapy in other TMA syndromes especially APS nephropathy has yet to be established. Small and retrospective studies suggest potential benefit of complement inhibition, MTOR inhibition, B cell depleting therapy, and plasma exchange therapy for patients with SLE and TMA, and prospective investigation of these therapies should be a research priority.

CLINICAL SUMMARY.

• Among patients with SLE and lupus nephritis, the presence of a thrombotic event, unexplained hypertension, thrombocytopenia, or hemolytic anemia should prompt consideration of APS and TMA. APS and TMA negatively impact the renal prognosis of patients with SLE and lupus nephritis.

• TTP and CM-HUS in patients with SLE should be treated in accordance with the standard of care for patients without SLE.

• APS with a documented thrombotic event is treated with anticoagulation. Duration depends upon whether the thrombosis was provoked or unprovoked. Warfarin or low molecular weight heparin is the standard of care, with emerging evidence for the role of direct oral anticoagulants in select patients. The role of hydroxychloroquine as an add-on to anticoagulation is under evaluation.

• The standard of care to manage patients with lupus nephritis and concomitant biopsy-proven TMA, in the absence of a clear TTP or CM-HUS syndrome to guide treatment, is not clear.