The incidence and prevalence of SLE depend on the age, sex, geographic locale, and ethnicity of the population studied, as well as the diagnostic criteria for defining SLE. SLE is newly diagnosed in approximately 400,000 people each year, with a prevalence of approximately 3.4 million. ^, One review found that Poland, the United States, Barbados, and China have the highest incidences of SLE. The incidence of SLE in Poland is the highest, at around 80 per 100,000 person-years. In the United States the SLE incidence is around 5 to 12 per 100,000 person-years. ^, Globally, SLE is more common in women compared with men being reported to be around 8.8 versus 1.5 per 100,000 person-years. Among women the prevalence is around 79 per 100,000 compared with 9.3 per 100,000 among men. These data must be interpreted with caution, however, as data included in the global analysis were only available for 20% of countries. The onset of lupus peaks between 15 and 45 years of age with >85% of patients being younger than 55 years. SLE is more often associated with severe nephritis in children and males. SLE and LN are more common and are associated with more severe kidney involvement, in populations of Asian, Hispanic, and African descent. Nearly half of all patients who develop end-stage kidney failure (ESKF) from SLE in the United States are African-American. The precise roles of biologic-genetic versus socioeconomic factors underlying this heterogeneity have not been clearly defined. The incidence of kidney involvement in SLE varies depending on the populations studied, the diagnostic criteria for kidney disease, and whether involvement is defined by biopsy or clinical findings. Approximately 25% to 50% of unselected lupus patients will have kidney disease at presentation, whereas up to 60% of adults with SLE will develop kidney involvement during their course.

A number of genetic, hormonal, and environmental factors influence the course and severity of SLE. A multiplicity of genes are involved in both SLE and LN. A genetic predisposition is supported by a higher concordance rate in monozygotic twins (25%) than fraternal twins (<5%), the greater risk of relatives of SLE patients developing SLE or other autoimmune disease, the association with distinct major histocompatibility complex class I and II (MHC) or human leukocyte antigen (HLA) genotypes (e.g., HLA-B8, HLA-DR2, and HLA-DR3), inherited deficiencies in complement components (e.g., homozygous C1q, C2, and C4 deficiencies), and Fc receptor polymorphisms. Some HLA alleles in SLE patients appear to be protective against LN (HLA DR4 and DR11), whereas others increase the risk of developing kidney involvement (HLA DR3 and DR13). Genome-wide association studies (GWASs) have identified more than 90 different genetic loci associated with an increased risk of SLE. These candidate susceptibility genes regulate diverse immune functions such as T-cell activation, B-cell signaling, Toll-like receptors (TLRs), signal transduction, neutrophil function, and interferon (IFN) production. ^, A meta-analysis of GWAS studies in SLE patients looking for risk alleles for LN has mapped these alleles to individual genes such as PDGF receptor A gene. Likewise, high-risk groups for LN such as those of African descent have a high frequency of genetic markers that may explain the increased risk, including certain Fc gamma RIIA–R131 alleles and apolipoprotein L1 (APOL1) risk alleles, which can almost triple the risk of ESKD in this population. ^, Inbred spontaneous genetic murine models of SLE and LN are commonly used to model the disease and include the NZB B/W F1 hybrid, BXSB, and MRL/lpr mouse. SLE is inducible in other murine strains through injection of autoantibodies against DNA or by injection of Smith antigen peptides. Evidence for the role of hormonal factors includes the strong predominance of SLE in females of childbearing age and the increased incidence of lupus flares during or shortly after pregnancy. In the F1 NZB/NZW mice, females have more severe disease than males and disease severity is ameliorated by oophorectomy or androgen therapy. Environmental factors other than estrogens also modulate disease expression; these factors include viral or bacterial antigens, exposure to sunlight and ultraviolet radiation, and certain medications. ^{, , ,}

The updated definition of SLE in 2019 by the European League Against Rheumatism (EULAR) and the American College of Rheumatology (ACR) involves a minimum score assigned based on signs and symptoms. For the kidney criterion of the scoring system, histology demonstrating class III or IV LN is assigned a higher score compared with class II LN, class V LN, or proteinuria >0.5 g/day ( Fig. 31.1 ).

Classification criteria for systemic lupus erythematosus.

In patients with SLE, abnormalities of immune regulation lead to a loss of self-tolerance and production of a variety of autoantibodies and immune complexes. ^{, ,} Genetic susceptibility may impact loss of tolerance to SLE antigens. SLE is associated with defective regulation of T cells with decreased numbers of cytotoxic and suppressor T cells; increased helper (CD4 ⁺ ) T cells; dysfunctional T-cell signaling; and abnormal T _H 1, T _H 2, and T _H 17 cytokine production. ^{, , ,} There is also polyclonal activation of B cells and defective B-cell tolerance. Autoreactive B-cell clones differentiate into plasma cells in the bone marrow and memory B cells located in lymphoid organs. B cells also drive the expansion of antigen-specific T-cell clones. The failure of apoptotic mechanisms to delete autoreactive B-cell and T-cell clones can promote their expansion and may trigger immune responses through interactions with TLRs leading to autoantibody production. The role of apoptosis, mediated via the Fas receptor or its ligand (FasL) in the generation of lupus, is unclear. Mutations in the FasL have been observed to lead to lupus-like systemic autoimmune diseases in mice and in some humans, but specific defects in FasL function are not evident in the majority of patients with SLE. In comparison with single antigen glomerular diseases, such as antineutrophil cytoplasmic antibody (ANCA) vasculitis and PLA2R-positive membranous nephropathy, SLE has a greater number of autoreactive T-, B-, and plasma cell clones ( Fig. 31.2 ). The result of these autoreactive clones in SLE is the production of a wide range of autoantibodies including those directed against nucleic acids, nucleosomes (double-stranded DNA [dsDNA] in association with a core of positively charged histones), chromatin antigens, and nuclear and cytoplasmic ribonuclear proteins. ^{, , , ,} Viral or bacterial peptides containing sequences similar to native antigens may lead to “antigen mimicry” and stimulate autoantibody production.

Number of autoreactive T, B, and plasma cell clones in glomerular diseases with single antigen targets, systemic lupus erythematosus, and organ transplant.

Immunological activity of all chronic autoimmune diseases is controlled by the size and antigen affinity of autoreactive clones of memory T cells and memory B cells (T/B) in the lymphoid organs as well as by long-lived plasma cells (PZ) in the bone marrow. In “simple” autoimmune diseases with single antigens such as, e.g. proteinase-3 in granulomatous polyangiitis or phospholipase A2 receptor in primary membranous glomerulonephritis, few immunotherapy is needed to control the few clones. In contrast, SLE involves autoimmunity against many, maybe 10-100 antigens, which goes together with a high number of T-, B-, and plasma cell clones to control, which requires combination therapy. Recipients of an organ transplant are in an even worst condition as each transplanted alloantigen will prime such a set of allo-reactive clones, which can go into the 10,000s of clones, difficult to control even with combination therapy. Differient auto-reactive T, B, and plasma cell clones are depicted with different colours. ANCA, Antineutrophil cytoplasmic antibody; PLA2R, phospholipase A2 receptor; SLE, systemic lupus erythematosus; GC, glucocortcoids; RTX, rituximab.

From Parodis I, Depascale R, Doria A, Anders HJ. When should targeted therapies be used in the treatment of lupus nephritis: early in the disease course or in refractory patients? Autoimmun Rev . 2024;23(1):103418.

In SLE, autoantibodies combine with self-antigens to produce circulating immune complexes that deposit in the glomeruli, activate complement, and incite an inflammatory response. Immune complexes are also detectable in the skin at the dermal-epidermal junction, in the choroid plexus, pericardium, and pleural spaces. Renal involvement in SLE has been considered a human prototype of classic experimental chronic immune complex–induced glomerulonephritis. The chronic deposition of circulating immune complexes plays a major role in the mesangial and endocapillary proliferative patterns of LN. The size, charge, and tissue avidity of the immune complexes, local hemodynamic factors, and ability of the mesangium to clear the complexes all influence the localization of circulating immune complexes within the glomerulus. ^{, ,} In diffuse proliferative LN, the deposited complexes consist of nuclear antigens (e.g., DNA) and high-affinity complement-fixing immunoglobulin G (IgG ₁ and IgG ₃ ) antibodies. In some SLE patients, the initiating event may be the local binding of cationic nuclear antigens such as histones to the subepithelial region of the glomerular capillary wall, followed by in situ immune complex formation. Once glomerular immune deposits form, the complement cascade is activated, leading to complement-mediated damage, activation of procoagulant factors, leukocyte Fc receptor activation with leukocyte infiltration, release of proteolytic enzymes, and production of various cytokines regulating glomerular cellular proliferation and matrix synthesis. Additionally, studies suggest TGFB1 may serve as a potential mediator of complement-related fibrosis. ^, Transcriptome analysis of glomeruli from LN biopsies shows variable expression of B-cell genes, myelomonocytic genes, IFN-inducible genes, and fibrosis genes in both human and murine models. Characterizing the immune profile of the kidney biopsy at the time of LN flare differentiates early treatment responders from nonresponders. Autoantibodies to complement components (C1q and C3b), which are found in some patients, may enhance autoantigen exposure and facilitate immune complex deposition, while autoantibodies to C-reactive protein (CRP) may also lead to further immune activation and worsening of LN. ^, Neutrophils undergoing cell death may release chromatin meshworks (called neutrophil extracellular traps [NETs] composed of chromatin, histones, and neutrophil proteins). These NETs, which are detectable in LN biopsies, are not degraded properly in patients with lupus and are a source of autoantigen presentation and induction of IFN-α by plasmacytoid dendritic cells. ^, There is also evidence for intrarenal autoantibody production in patients with LN. Glomerular damage may be potentiated by mechanisms distinct from immune complex deposition, such as hypertension and coagulation or complement regulatory abnormalities. For example, genetic polymorphism in complement factor H (CFH) influences susceptibility to SLE, and its deficiency accelerates the progression of LN in mice. Some lupus patients with associated ANCAs have documented focal segmental necrotizing glomerular lesions without significant immune complex deposition, resembling a “pauci-immune” glomerulonephritis. ^, The presence of APL antibodies, with their attendant alterations in endothelial and platelet function (including reduced production of prostacyclin and other endothelial anticoagulant factors, activation of plasminogen, inhibition of protein C or S, and enhanced platelet aggregation), can also potentiate glomerular and vascular lesions. Infrequently, SLE patients can develop podocytopathy with no evidence of immune complex deposition. ^,

The histopathology of LN is pleomorphic. ^{, , ,} This diversity is evident when comparing biopsy findings from different patients or even adjacent glomeruli in a single biopsy. Moreover, the lesions have the capacity to transform from one pattern to another spontaneously or following treatment. ^, The World Health Organization (WHO) ^, classified LN by combining glomerular light microscopic (LM), immunofluorescence (IF), and electron microscopic (EM) findings. The 2003 International Society of Nephrology/Renal Pathology Society (ISN/RPS) classification of LN ( Table 31.1 ) has proven more reproducible and provides more standardized definitions for precise clinical pathologic correlations. ^, In 2018, a minor revision of this classification was published. This revision provided a new cutoff for mesangial hypercellularity (>3 mesangial cells per mesangial area) and more precise definitions of cellular and fibrocellular crescents. It replaced the term “endocapillary proliferation” with “endocapillary hypercellularity” to acknowledge the major contribution of infiltrating leukocytes rather than proliferation of resident cells. It also eliminated the subdivisions IV-S and IV-G of class IV LN and applied modified National Institutes of Health (NIH) activity and chronicity indices to all LN classes.

ISN/RPS class I has normal-appearing glomeruli by LM but with mesangial immune deposits detected by IF and EM. Even patients without clinical renal disease often have mesangial immune deposits when studied carefully by the more sensitive techniques of IF and EM.

ISN/RPS class II is defined as mesangial hypercellularity, which is defined as more than three cells in mesangial regions distant from the vascular pole in 3-μm-thick sections, with mesangial immune deposits on IF and EM ( Fig. 31.3 ).

Class II mesangial proliferative lupus nephritis.

Increase in mesangial matrix and cellularity. Peripheral capillary loops are normal. The adjacent tubules in the interstitium are uninvolved (hematoxylin-eosin, ×400).

From Fogo A, Kashgarian M. Diagnostic atlas of renal pathology . 3rd ed. Elsevier, 2017.

ISN/RPS class III, focal LN, is defined as focal segmental and/or global endocapillary and/or extracapillary glomerulonephritis affecting less than 50% of the total glomeruli sampled ( Figs. 31.4 and 31.5 ). Both active and chronic lesions are taken into account when determining the percentage of total glomeruli involved. There is typically focal segmental endocapillary hypercellularity, which variably includes mesangial cells, endothelial cells, and infiltrating mononuclear or polymorphonuclear leukocytes. Class III biopsies can have any combination of active and chronic features. Active lesions may display cellular crescents, fibrinoid necrosis, nuclear pyknosis or karyorrhexis, and rupture of the glomerular basement membrane (GBM). Hematoxylin bodies, swollen basophilic nuclear material resulting from binding to ambient ANAs, are occasionally found within the necrotizing lesions. Subendothelial immune deposits may be visible by LM as “wire-loop” thickenings of the glomerular capillary walls or large intraluminal masses known as “hyaline thrombi.” Chronic glomerular lesions consist of segmental and/or global glomerular sclerosis owing to scarred glomerulonephritis with or without fibrous crescents. In class III biopsies, glomeruli adjacent to those with severe histologic changes may show only mesangial abnormalities by LM. In class III, diffuse mesangial and focal and segmental subendothelial immune deposits are typically identified by IF and EM. The segmental subendothelial deposits are usually present in the distribution of the segmental endocapillary hypercellularity.

Class III lupus nephritis.

Focal and segmental glomerulonephritis. The glomerulus in the upper left pole is normal appearing, whereas the one on the right demonstrates a segmental area of adhesion and necrosis (hematoxylin-eosin, ×400).

From Fogo A, Kashgarian M. Diagnostic atlas of renal pathology . 3rd ed. Elsevier, 2017.

Class III lupus nephritis.

Electron microscopy demonstrating subendothelial and mesangial deposits (transmission electron microscopy, ×4000).

From Fogo A, Kashgarian M. Diagnostic atlas of renal pathology . 3rd ed. Elsevier, 2017.

ISN/RPS class IV, diffuse LN, has qualitatively similar glomerular endocapillary and/or extracapillary lesions as class III but involves more than 50% of the total glomeruli sampled ( Fig. 31.6 ). ^{, , ,} Again, both active (hypercellular) and chronic (sclerosing) lesions are included when determining the percentage of glomeruli affected. All the active features described earlier for class III (including fibrinoid necrosis, leukocyte infiltration, wire-loop deposits, hyaline thrombi, hematoxylin bodies, and crescents) may be encountered in class IV LN. In general, there is more extensive peripheral capillary wall subendothelial immune deposition, and extracapillary hypercellularity in the form of crescents is not uncommon. Class IV lesions may have features similar to those of primary membranoproliferative glomerulonephritis (MPGN; also known as mesangiocapillary glomerulonephritis) with mesangial interposition along the peripheral capillary walls and double contours of the GBMs. Some class III and IV biopsies will have focal necrotizing and crescentic lesions akin to those seen in small vessel vasculitides. Some of these patients have circulating ANCAs, especially of the peripheral IF (P-ANCA) or antimyeloperoxidase ELISA variety. ^,

Class IV LN.

Diffuse proliferative LN involves most or all of the glomeruli. The capillaries here contain “hyaline thrombi,” a lesion caused by massive subendothelial deposits bulging into the capillary lumen. Hyaline thrombi may be thickened to the point to be considered “wire-loop” lesions, due to the extensive subendothelial deposits (hematoxylin-eosin, ×400).

From Fogo A, Kashgarian M. Diagnostic atlas of renal pathology . 3rd ed. Elsevier, 2017.

The >50% or <50% glomerular involvement is somewhat arbitrary, and there is no prognostic significance in this finding alone. ^, For example, if a kidney biopsy contains 11 glomeruli and 6 (>50%) of the glomeruli show proliferation and the findings described earlier, the diagnosis would be class IV LN. If 5/11 glomeruli showed these changes, it would be reported as class III LN. Therefore the distinction between class III and IV may be somewhat random and should not necessarily have implications for therapy.

ISN/RPS class V is defined by regular subepithelial immune deposits producing a membranous pattern ( Fig. 31.7 ). ^, The coexistence of mesangial immune deposits and mesangial hypercellularity in most cases helps to distinguish membranous LN from primary membranous glomerulopathy (and may also be responsible for the microhematuria sometimes noted in membranous LN). Early membranous LN class V may have no identifiable abnormalities by LM, but subepithelial deposits are detectable by IF and EM. In well-developed membranous lesions, there is typically thickening of the glomerular capillary walls and “spike” formation between the subepithelial deposits. Because sparse subepithelial deposits may also be encountered in other classes (III or IV) of LN, a diagnosis of “pure” lupus membranous LN should be reserved only for those cases in which the membranous pattern predominates. When the membranous alterations involve more than 50% of the total glomerular capillaries and are accompanied by focal or diffuse endocapillary hypercellular lesions and subendothelial immune complex deposition, they are classified as class III + V or class IV + V, respectively.

Class V lupus membranous nephritis.

Silver methenamine (Jones) stains reveal a spike and dome pattern to be present along the peripheral capillary loops. Where the wall of the capillaries is cut tangentially, there is a moth-eaten appearance of the capillary wall due to the deposits not staining with silver (Jones, ×400).

From Fogo A, Kashgarian M. Diagnostic atlas of renal pathology . 3rd ed. Elsevier, 2017.

ISN/RPS class VI, advanced sclerosing LN is reserved for biopsies with more than 90% of the glomeruli sclerotic and no residual activity. In such cases, it may be difficult even to establish the diagnosis of LN without the identification of residual glomerular immune deposits by IF and EM or a biopsy history of prior active LN.

Although SLE predominantly affects young females, the clinical manifestations are similar in both sexes and in adults and children. Organ systems commonly affected include the kidneys, joints, serosal surfaces (including pleura and pericardium), central nervous system, and skin. In addition, involvement of the heart, liver, lungs, hematopoietic, and gastrointestinal organ systems is not uncommon.

Renal involvement often develops concurrently or shortly after the onset of SLE and may follow a protracted course with periods of remissions and exacerbations. Clinical renal involvement usually, but not always, correlates with the extent of glomerular involvement. However, some patients may have disproportionately severe vascular or tubulointerstitial lesions that dominate the clinical course. ^{, ,}

Patients with ISN/RPS class I biopsies often have little evidence of clinical renal disease and are often diagnosed following a protocol biopsy. Likewise, most patients with mesangial lesions (ISN/RPS class II) have mild or minimal clinical renal findings. ^{, ,} They may have active lupus serology (a high anti-DNA antibody titer and low serum complement), but the urinary sediment may be inactive, hypertension is infrequent, proteinuria is usually <1 g/day, and the serum creatinine concentration and glomerular filtration rate (GFR) are usually normal. Nephrotic-range proteinuria is extremely rare with class II unless there is a superimposed podocytopathy. ^,

Class III, focal proliferative LN, is often associated with active lupus serology, although the degree of serologic activity does not necessarily correlate with the histologic severity. ^, Hypertension and active urinary sediment are common. Proteinuria is often more than 1 g/day, and one-quarter to one-third of patients with focal LN have nephrotic syndrome at presentation. Many patients, in addition, have an elevated serum creatinine concentration at presentation. Patients with less extensive glomerular proliferation, fewer necrotizing features, and no crescents are more likely to be normotensive and have preserved renal function.

Patients with ISN/RPS class IV, similar to class III, often have high anti-DNA antibody titers, low serum complement levels, and very active urinary sediment, with erythrocytes, and cellular casts on urinalysis. ^{, , , ,} Virtually all have proteinuria and up to 50% of the patients will have nephrotic syndrome. Hypertension and renal dysfunction are typical. Even when the serum creatinine level is in the “normal range,” the GFR is usually depressed. As mentioned earlier, the distinction between “focal” and “diffuse” proliferative LN may depend on tissue biopsy sampling. Class III and IV lupus may present both clinically and histologically with a wide spectrum of involvement, and it is unclear whether current recommendations for the treatment of these classes of LN apply to those who present with milder disease or pathologic findings on biopsy.

Patients with membranous LN, ISN/RPS class V, typically present with proteinuria, edema, and other manifestations of nephrotic syndrome. ^{, , ,} However, up to 40% will have proteinuria of <3 g/day, and 16% to 20% <1 g/day. In contrast to the proliferative variants, only about 60% of membranous LN patients have low serum complement concentration and an elevated anti-DNA antibody titer at presentation. The ANA is usually positive. However, hypertension and renal dysfunction may occur without superimposed proliferative lesions. Patients with membranous LN may present with nephrotic syndrome before developing other clinical and laboratory manifestations of SLE. ^, In addition, they are predisposed to thrombotic complications such as renal vein thrombosis and pulmonary emboli. ^, Patients with mixed membranous and proliferative biopsies have clinical features that reflect both disease components.

Advanced sclerosing LN, ISN/RPS class VI, is usually the result of “burned-out” LN of long duration. Some renal histologic damage may represent nonimmunologic progression of sclerosis mediated by hyperfiltration in remnant nephrons. Although the lesions are sclerosing and inactive, class VI patients may still have microhematuria and proteinuria. Virtually all have hypertension and a decreased GFR. Levels of anti-DNA antibodies and serum complement levels often normalize at this late stage of disease.

“Silent LN” ^{, ,} has been described in patients without clinical evidence of renal involvement despite biopsy evidence of active proliferative LN. Some define silent LN as active biopsy lesions without active urinary sediment, proteinuria, or a depressed GFR, whereas others require negative lupus serologies as well. Although silent LN is well described in some studies, others have been unable to find even isolated examples. ^, It appears to be uncommon, and it is highly likely that even patients with true “silent disease” will manifest clinical renal involvement over time.

Lupus cystitis is an underrecognized manifestation of SLE. Permanent bladder dysfunction has been reported in rare cases. It affects an estimated 0.01% to 2% of SLE patients, mainly women. Most of the reported cases are from East Asia, particularly Japan. The pathophysiology is proposed to be due to immune complex vasculitis in the arterioles of the bladder. This can result in progressive bladder fibrosis and thinning and rarely obstruction. Both a dilated urinary tract and stenosis have been reported. Symptoms may include urgency, dysuria, frequency, nocturia, and lower abdominal pain; reports of copresentation with lupus enteritis have been described. ^, Imaging and cystoscopy are required for diagnosis. Glucocorticoids and cyclophosphamide have been reported as treatment, with other immunosuppression regimens for refractory cases.

It is important to be able to predict systemic and renal relapses and prevent their occurrence through the judicious use of immunosuppressive agents. There may be clinical differences between patients with relapsed disease versus new presentations of LN. Particularly, hypertension may be less frequent in relapsed patients compared with newly diagnosed LN. Serial measurements of many serologic tests (including complement components, autoantibodies, erythrocyte sedimentation rate [ESR], CRP, circulating immune complexes, and, recently, levels of cytokines and interleukins [ILs]) have been used to predict lupus flares. Although there is controversy regarding the value of serum C3 and C4 levels and anti-DNA antibody titers in predicting clinical flares of SLE or LN, these have yet to be replaced by new biomarkers. ^{, ,} Additional factors impacting the use of anti-dsDNA to predict LN activity include the use of immunosuppression, which may affect levels, as well as overlap with class 5 disease, the majority of which are known to be anti-dsDNA negative. In most patients with lupus activity, however, serum levels of anti-dsDNA typically rise and serum complement levels typically fall as the clinical activity of SLE increases, often preceding clinical renal deterioration. A particularly challenging group of patients may be those that remain serologically active but clinically quiescent for long periods of time. However, these patients tend to have more disease activity in concert with a lupus flare.

In patients with active renal involvement, the urinalysis frequently reveals dysmorphic erythrocytes, red blood cell casts, and other formed elements. Given the vagaries of centrifuging urine and preparing a slide, the number of red cells or red cell casts per high-power field should not be used as a measure of the degree of kidney involvement. An increase in proteinuria from levels of <1 g/day to nephrotic levels is a clear indication of either increased activity or a change in renal histologic class. When there is concern about the degree of activity of the SLE and LN, a repeat renal biopsy will often clarify whether to change therapy. ^,

Some data suggest that specific serologic tests, in conjunction with clinical parameters and urine investigations, may help predict the likelihood of histologic activity with LN. Gao and colleagues developed a prediction model composed of mean arterial pressure, hematocrit, serum albumin, eGFR, urinary erythrocytes, serum C3, and anti-dsDNA to predict an activity index >2 on kidney biopsy with good correlation. External validation is still required in diverse patient populations before this prediction tool is recommended for clinical use, but this study suggests that the constellation of the clinical assessment and laboratory parameters may prove useful. The predictability of kidney lesions with alternative biomarkers would be of value where kidney biopsy may be high risk or unavailable. Until this is achieved, in some contexts it may be reasonable to proceed with therapy without subjecting the patient to the biopsy.

A variety of medications may induce a lupus-like syndrome or exacerbate an underlying predisposition to SLE. Those medications metabolized by acetylation, such as procainamide and hydralazine, have been common causes. ^, This occurs more commonly in patients who are slow acetylators due to genetic variants associated with a decrease in hepatic N -acyltransferase activity. Diltiazem, minocycline, penicillamine, isoniazid, methyldopa, chlorpromazine, and practolol, a selective β1-blocker, are other potential causes of drug-induced lupus. Other drugs that have been associated less frequently with this syndrome include phenytoin, quinidine, propylthiouracil, sulfonamides, lithium, β-blockers, nitrofurantoin, para-aminosalicylic acid, captopril, glyburide, hydrochlorothiazide, IFN-α, carbamazepine, sulfasalazine, rifampin, and tumor necrosis factor–α (TNF-α) blockers. ^{, ,}

Clinical manifestations of drug-induced lupus include fever, rash, myalgias, arthralgias and arthritis, and serositis. Central nervous system and renal involvement are relatively uncommon. ^{, ,} While elevated anti-DNA antibodies and depressed serum complement levels are less common in drug-induced lupus, antihistone autoantibodies are present in more than 95% of patients. These are usually formed against a complex of the core histone dimer H2A-H2B and DNA and other histone components. ^, Antibodies are also present in the vast majority of idiopathic, nondrug-related SLE patients, but they are directed primarily against different histone antigens (linker H1 and core H2B). The presence of antihistone antibodies in the absence of anti-DNA antibodies and other serologic markers for SLE is also indicative of drug-induced disease. The diagnosis of drug-induced lupus depends on documenting the offending agent and achieving a remission following withdrawal of the drug. The primary treatment consists of discontinuing the offending drug. For patients with an established diagnosis of SLE and LN, it may be reasonable to allow them to take medications associated with drug-induced lupus as long as they are in remission. However, if they are in a flare of SLE, it may be safest to withhold these drugs given their potential to confound lupus symptomatology.

Because SLE occurs so commonly in women of childbearing age, the issue of pregnancy arises often in the care of this population. Independent but related issues are the health of the mother (in terms of both flares of lupus activity and progression of renal disease) and the fate of the fetus. For more discussion on pregnancy and kidney disease, see Chapter 58 . These maternal and fetal complications in SLE patients include, but are not limited to, higher rates of cesarean section, preeclampsia and hypertension, spontaneous abortion, thromboembolic disease, postpartum infections, premature births and babies small for gestational age, neonatal intensive care unit requirements, congenital defects, and lower live birth rates. Approximately 1 in 5 pregnancies in women with SLE experience an adverse pregnancy outcome. One systematic review and meta-analysis has described the adverse pregnancy risks associated with SLE but recognizes attrition bias and confounding. SLE disease activity at conception or preconception has been associated with increased preterm birth (OR 2.91 [95% CI 1.96–433]) and preeclampsia (OR 2.32 [1.40–3.83]). Other risk factors for adverse pregnancy outcomes include primigravida mothers, use of antihypertensives, and thrombocytopenia. ^, Frequency of adverse pregnancy outcomes also varies by race and ethnicity, with higher rates of complications among patients of African descent.

It is unclear whether flares of SLE occur more commonly during pregnancy or shortly after delivery, but reported flare rates in pregnancy are approximately 30% to 60%. ^, Patients with quiescent lupus for at least 6 months before conception are less likely to experience an exacerbation of SLE during pregnancy. Decreased serum complement levels, particularly C4, during the first trimester have been associated with risk of flare and adverse pregnancy outcomes. ^, Therefore both immunologic and clinical SLE quiescence is optimal before pregnancy to reduce the risk of adverse maternal and fetal outcomes.

Pregnancy in patients with preexisting LN has also been associated with worsening of renal function. ^, This is less likely to occur in patients who have been in remission for at least 6 months. However, even previous LN has also been associated with decreased live birth rate (OR 0.62 [95% CI 0.47–0.81]), increased risk of preterm birth (OR 2.00 [1.55-2.57]), and increased risk of preeclampsia (OR 3.11 [2.35–4.12]). Patients with hypertension are likely to develop higher blood pressure, and those with proteinuria are likely to have increased proteinuria during pregnancy, at least in part as a result of hyperfiltration. Chronic hypertension has been associated with increased risk of disease flare in pregnancy (OR 2.50 [95% CI 1.74–3.58]), preterm birth (OR 2.65 [1.87–3.77]), and preeclampsia (OR 5.86 [3.41–10.06]). Patients with elevated serum creatinine levels are most likely to suffer worsening of renal function and to be at highest risk for fetal loss.

It is generally recommended to continue with hydroxychloroquine during pregnancy, with some studies reporting lower rates of SLE flare and a decrease in congenital heart block. ^, It is worth mentioning one study that demonstrated hydroxychloroquine was associated with a small increased risk of congenital anomalies (OR 1.26) with a daily dose >400 mg in the first trimester. Although it is difficult to dismiss these findings, hydroxychloroquine has long been considered safe in pregnancy and notable confounders were not accounted for in this study. Although high-dose corticosteroids, cyclosporine, tacrolimus, and azathioprine have all been used in pregnant lupus patients, their safety is unclear. Cyclophosphamide is contraindicated due to its teratogenicity. Mycophenolate is also highly teratogenic and associated with first-trimester pregnancy loss and facial deformities such as cleft lip/palate, cartilage abnormalities of the outer ear, absent auditory canals, and limb abnormalities. It should be discontinued at least 6 weeks before planned pregnancies. Newer agents such as rituximab are not recommended due to a lack of data about their safety. Although calcineurin inhibitors have been used in pregnancy, voclosporin is contraindicated in pregnancy due to the high alcohol content. Belimumab, a human monoclonal antibody that inhibits B-cell activating factor (BAFF), is not currently recommended for use in pregnancy. Postmarketing reports in pregnant women with SLE exposed to belimumab are small in number and heterogeneous and provide neither clear risk of harm nor sufficient evidence of safety. ^, The judicial use of rituximab and belimumab in pregnancy should only be considered in those at highest risk of maternal and fetal complications.

The rate of fetal loss in most series is 20% to 40% and may even approach 50% in other selected series. ^{, , ,} While fetal mortality is increased in SLE patients with renal disease, it may be decreasing in the modern treatment era. Patients with anticardiolipin or APL antibodies, hypertension, or heavy proteinuria are at higher risk for fetal loss. Secondary antiphospholipid syndrome (APS) has been associated with decreased probability of livebirth (OR 0.40 [0.27–0.58]), increased risk of pregnancy loss after 20 weeks of gestation (OR 2.77 [1.44–5.31]), and increased risk of preterm birth (OR 1.65 [1.29–2.11]). One review of 10 studies in more than 550 women with SLE found that fetal death occurred in 38% to 59% of all pregnant SLE patients with APL antibodies compared with 16% to 20% of those women with SLE without these antibodies. Prospective studies are needed to better define risk and overcome bias in retrospective studies.

There is significant morbidity and mortality associated with LN, associated with autoimmune disease activity and progression of kidney disease, as well as treatment-related adverse effects from immunosuppressive therapies. Compared with the standard population, those with LN may have up to a sixfold higher rate of mortality. Even in comparison with other primary glomerular diseases, patients with LN experience a higher risk of death and major adverse cardiovascular events. This increased risk is likely driven by progression of atherosclerotic disease from both LN ^, and CKD, but serious infectious complications are an important contributor to mortality in this population. This mortality risk may be partially mitigated by achieving either complete or partial renal remission. ^,

Among patients with LN, about 10% may progress to ESKF over 10 years. ^{, ,} This course is defined by the initial pattern and severity of renal involvement as modified by therapy, exacerbations of the disease, adherence to therapy, and complications of treatment. The prognosis has clearly improved in recent decades with wider and more judicious use of new immunosuppressive medications. ^,

Patients with lesions limited to the renal mesangium generally have an excellent course and prognosis. ^{, ,} If these patients do not transform into other patterns, they are unlikely to develop progressive renal failure and any excess mortality is due to extrarenal manifestations and complications of therapy. Patients with focal proliferative disease have an extremely varied course. Those with mild proliferation involving a small percentage of glomeruli respond well to therapy, and <5% progress to renal failure over 5 years. ^{, , ,} Patients with more proliferation, necrotizing features, and/or crescent formation have a prognosis more akin to patients with class IV diffuse LN. Class III patients may transform into class IV over time and, as mentioned, sometimes the distinction between the two may be the result of random sampling. Some patients with active segmental proliferative and necrotizing lesions resembling ANCA-associated small vessel vasculitis have a worse renal prognosis than other patients with focal proliferative lesions. ^{, ,}

Patients with diffuse proliferative disease have the least favorable prognosis in most older series. ^{, ,} Nevertheless, the prognosis for this group has markedly improved, with renal survival rates now exceeding 90% in some series of patients treated with modern immunosuppressive agents. ^{, ,} In trials from the NIH, the 5-year risk of doubling the serum creatinine concentration, a surrogate marker for progressive renal disease, in patients with diffuse proliferative lupus treated with cyclophosphamide-containing regimens ranged from 35% to <5%. In an Italian study of diffuse proliferative LN, patient survival was 77% at 10 years and more than 90% of extrarenal deaths were excluded. In a U.S. study of 89 patients with diffuse proliferative LN, renal survival was 89% at 1 year and 71% at 5 years. It is unclear whether the improved survival rates in these recent series are largely due to improved immunosuppression or better supportive care and more judicious clinical use of these medications.

In the past, some studies have found age, sex, and race to be as important prognostic variables as clinical features in patient and renal survival in SLE. ^{, , , , ,} A consistent finding is that those of African ancestry have a greater frequency of LN and a worse renal and overall prognosis. This worse prognosis appears to relate to both biologic/genetic and socioeconomic factors. ^{, , , , , ,} In a study from the NIH of 65 patients with severe LN, clinical features at study entry associated with progressive renal failure included age, African ancestry, hematocrit <26%, and serum creatinine concentration >2.4 mg/dL. Patients with a combined activity index (>7) plus chronicity index (>3) on renal biopsy, as well as those with the combination of cellular crescents and interstitial fibrosis also had a worse prognosis. In another U.S. study of 89 patients with diffuse proliferative LN, none of the following features affected renal survival: age, gender, SLE duration, uncontrolled hypertension, or any individual histologic variable. Entry serum creatinine level higher than 3.0 mg/dL, increased activity and chronicity indices on biopsy, and African ancestry predicted a poor outcome. Five-year renal survival rate was 95% for Caucasian patients but only 58% for Black patients. In a study of more than 125 LN patients with WHO class III or IV from New York, both racial and socioeconomic factors were associated with the worst outcomes in African-Americans and Hispanics. An evaluation of 203 patients from the Miami area confirmed worse renal outcomes in African-Americans and Hispanics, argued to be related to both biologic and economic factors.

More rapid and more complete renal remissions are associated with improved long-term prognosis. ^, Repeated renal flares during the course of SLE, including those as a result of nonadherence to therapy, also may predict a poor renal outcome. Relapses of severe LN over 5 to 10 years of follow-up occur in up to 50% of patients and usually respond less well and more slowly to repeated courses of therapy. ^{, ,} A retrospective analysis of 70 Italian patients in whom more than half had diffuse proliferative disease found excellent patient survival (100% at 10 years and 86% at 20 years), as well as preserved renal function with the likelihood of not doubling the serum creatinine concentration to be 85% at 10 years and 72% at 20 years. Most patients in this study were Caucasian, which likely influenced the excellent long-term prognosis. Patients with renal flares of any type had seven times the risk of renal failure, and those with rapid rises in creatinine had 27 times the chance of doubling their serum creatinine concentration. Another Italian study of 91 patients with diffuse proliferative LN showed more than 50% having a renal flare, which correlated with a younger age at biopsy (<30 years old), higher activity index, and karyorrhexis on biopsy. The number of flares, nephritic flares, and flares with increased proteinuria correlated with a doubling of the serum creatinine. The role of relapses in predicting progressive disease has been noted by others as well, although relapse does not invariably predict a bad outcome.

Although an elevated anti-dsDNA antibody titer and low serum complement levels may correlate with active renal involvement, they do not correlate with long-term renal prognosis. ^{, , , ,} In several studies, anemia has been a poor prognostic finding regardless of the underlying cause. ^, Severe hypertension has also been correlated with renal prognosis in some, but not all, studies. Renal dysfunction, as noted by elevated serum creatinine or decreased GFR, heavy proteinuria, and especially nephrotic syndrome, is indicative of a poor renal prognosis in the vast majority of series. ^{, , ,} However, not all studies have found an elevation of the initial serum creatinine to predict a poor long-term outcome, and in some the initial serum creatinine only predicted short-term renal survival. Other renal features, such as duration of LN and rate of decline of GFR, may also predict prognosis. ^,

Finally, histologic features such as the class, the degree of activity and chronicity, and the severity of tubulointerstitial damage have also predicted outcomes. In a number of studies, the pattern of renal involvement, especially when using the ISN/RPS or older WHO classification, has been a useful guide to prognosis. ^{, ,} In NIH trials, patients with severe proliferative LN with a higher activity index or chronicity index were more likely to have progressive renal failure. However, studies with different referral populations could not confirm this finding. ^{, , ,} Regardless, the contribution of chronic renal scarring to a poor long-term outcome has been confirmed by many studies. ^{, , ,} Some studies have found the initial renal biopsy to have little predictive value; rather, certain features on a repeat biopsy at 6 months proved to be a strong predictor of doubling the serum creatinine or progression to renal failure. ^, The lack of predictive value of the initial biopsy also suggests that a high-risk kidney biopsy at presentation may not always be necessary. These include ongoing inflammation with cellular crescents, macrophages in the tubular lumens, persistent immune deposits (especially C3) on IF microscopy, and persistent subendothelial and mesangial deposits or an active NIH activity score. ^, Other studies suggest that reversal of interstitial fibrosis and glomerular segmental scarring along with remission of initial inflammation and immune deposition are important favorable prognostic findings on the 6-month biopsy. Thus this argues that chronic changes on the initial biopsy are not always cumulative or immutable. Surveillance kidney biopsies for prognosis may not be all that useful and should be weighed against the risk of biopsy.

The natural history of membranous LN is less clear. In early studies, the prognosis appeared far better than the prognosis of active proliferative disease. Subsequent studies with longer follow-up suggested a worse outcome for membranous LN with persistent nephrotic syndrome. Even nonnephrotic patients with membranous LN may be at risk for sustained loss of kidney function. Retrospective analyses show that 5-year renal survival rates largely depend on whether patients have pure membranous lesions (class V) or superimposed proliferative lesions in a focal (class III + V) or diffuse (IV + V) distribution. ^, A U.S.-based study found that the 10-year survival rate from death or dialysis was 72% for patients with pure membranous lesions but only 20% to 48% for those with superimposed proliferative lesions. A larger Canadian study described a similar good survival from ESKF of 92% over a median of 8 years for patients with only class 5 lesions; however, nearly 20% of patients developed a sustained 40% decline in eGFR. Black race, younger age, elevated serum creatinine, higher degrees of proteinuria, hypertension, higher NIH chronicity index, and transformation to another WHO pattern all portended a worse outcome. The poor survival in those of African descent with membranous LN may explain the excellent results in retrospective Italian studies, which largely follow Caucasian cohorts. One such Italian study found the 10-year survival rate of membranous LN patients to be 93%. Even in this Italian population, survival for pure membranous LN was far better than in patients with superimposed proliferative lesions. Thus at least in part, the variability of prognosis can be at least partially explained by the differences in racial background and histology.

For patients with class V membranous LN, there have been conflicting data regarding the course, prognosis, and response to treatment. It is reasonable to assume that membranous LN should be treated similarly to other causes of nephrotic syndrome with antiproteinuric and lipid-lowering measures. The long-term renal benefit of acute initiation of RAS inhibitors is unclear, as the priority of initial management is immune therapy. The possible adverse effects of treatments with consideration of the number of different medications being prescribed to a young cohort not used to taking pills should be kept in mind. The use of anticoagulation to prevent thromboembolic events in patients with severe nephrotic syndrome should be individualized.

There is disagreement about which patients with class V membranous LN should be treated with immunosuppressive therapy. The degree of superimposed proliferative lesions greatly influences outcomes in class V patients, and it is unclear if older trials included only pure membranous LN patients. Thus early trials reported low and inconsistent response rates with oral corticosteroids. Excellent long-term results with intensive immunosuppressive regimens from Italian studies and others raise questions of whether the results are related to the therapeutic intervention, the population studied, or better supportive treatments. A retrospective Italian trial found better remission rates with a regimen of chlorambucil and methylprednisolone than corticosteroids alone. In a small nonrandomized trial of cyclosporine in membranous LN, there was an excellent remission rate of nephrotic syndrome with mean proteinuria decreasing from 6 to 1 or 2 g/day by 6 months. At long-term follow-up and rebiopsy, there was no evidence of cyclosporine-induced renal damage, but two patients had developed superimposed proliferative lesions over time. An NIH trial of 42 nephrotic patients with membranous LN compared cyclosporine, prednisone (alternate-day), and intravenous cyclophosphamide and found superior remission rates for the cyclosporine and cyclophosphamide regimens, but more relapses when the cyclosporine was withdrawn. This tendency to relapse suggests that calcineurin inhibitors may suppress proteinuria principally via nonimmunologic mechanisms. Although cyclosporine has been investigated more extensively than tacrolimus in lupus membranous nephropathy, given the similar mechanism of action, and that tacrolimus has been studied in multitarget therapy of proliferative LN, it may be reasonable to use tacrolimus for treatment of those with class V LN and heavy nephrotic syndrome to target reduction in proteinuria. This may be particularly relevant if there is concern about cosmetic or metabolic side effects of cyclosporine.

Both azathioprine and mycophenolate have been investigated in lupus membranous nephropathy. A study of 38 patients with pure membranous LN evaluated long-term treatment with prednisone plus azathioprine. At 12 months 67% of the patients had experienced a complete remission and 22% a partial remission. At 3 years only 12% had relapsed, at 5 years only 16%, and at 90 months only 19% relapsed. At the end of follow-up, no patient had doubled serum creatinine. Clearly in this population, a regimen of steroids plus azathioprine was highly effective. The response of patients with membranous LN to MMF has been varied. There were 84 patients with pure ISN class V membranous LN among the 510 patients enrolled in two similarly designed randomized controlled trials comparing MMF and intravenous cyclophosphamide induction therapy. Rates of remissions, relapse, and course were similar in both treatment groups (and also may suggest that it was the corticosteroid alone that was responsible for the treatment effect). Thus MMF can also be considered a first-line therapy for certain patients with membranous LN.

Given limited data, the treatment of membranous LN should be individualized. Patients with pure membranous LN and a good renal prognosis (subnephrotic levels of proteinuria and preserved GFR) can be managed expectantly. For those at higher risk of progressive disease (African descent, those fully nephrotic), options include cyclosporine, tacrolimus, monthly intravenous pulses of cyclophosphamide, MMF, or azathioprine plus corticosteroids. There are reports of patients responding to rituximab as well. Patients with mixed membranous and proliferative LN are treated in the same way as those with proliferative disease alone, although we have seen patients respond to immunosuppressive therapy for proliferative LN and in parallel develop de novo membranous lupus nephropathy.

Importantly, the time needed for resolution of proteinuria can often be months to years with the current therapies. ^, There may be “residual proteinuria” that can take years to resolve, if at all. If all other indices are favorable, it may put the patient at risk to intensify or continue immunosuppression in attempting to eradicate proteinuria. There may be a greater role in optimizing nonimmune therapies, such as RAS inhibitors, at this stage.

The majority of therapies currently under investigation target B cells in LN ( Fig. 31.10 ). Obinutuzumab, an FDA-approved fully humanized anti-CD20 monoclonal therapy, has shown promising results for their phase 2 clinical trial in proliferative LN. One-hundred and twenty-five patients with proliferative LN receiving MMF and prednisone were randomized to receive obinutuzumab or placebo on day 1 and weeks 2, 24, and 26 and followed through week 104. The primary endpoint was complete renal remission (CRR) at weeks 52 and 104. The CRR was greater for obinutuzumab than placebo at weeks 52 (35% vs. 23%, P = 0.115) and 104 (41% vs. 23%, P = 0.26). Although there were infusion-related reactions (nonsevere), obinutuzumab was not associated with severe adverse events. Compared with rituximab in the LUNAR study, obinutuzumab resulted in a faster, more uniform, and greater degree of B-cell depletion, which may account for the impact seen. The number of autoreactive B-, T-, and plasma cell clones in LN compared with other glomerular diseases may explain why obinutuzumab yielded promising results in LN, whereas rituximab did not.

Current targets of investigational therapies in lupus nephritis. B cells are precursors for plasma cells. B cells present autoantigens to T cells and hence drive the expansion of autoantigen-specific T cell clones. For this reason, drug developers primarily invest in B-cell targeting therapies to minimize autoimmune activity in lupus nephritis, which is mediated, among others, by autoantibodies, cytokines and T cell cytotoxicity.

The results of the phase 3 study of obinutuzumab for patients with LN have been released. Patients were randomized to receive obinutuzumab in one of two dose schedules (1000 mg on day 1 and at weeks 2, 24, 26, and 52, with or without a dose at week 50). The CRR was again greater for the obinutuzumab group than placebo at week 76 (42.7% vs. 30.9%, P = 0.05). There were more serious adverse effects in the obinutuzumab group including COVID-19 pneumonia and neutropenia. The authors highlight that most of the SARS-CoV-2 infections were early in the pandemic before vaccines were available, and no serious infections were observed in the latter half of the study once vaccination programs were commenced. Nonetheless, this increased risk of COVID-19 pneumonia when obinutuzumab was added to mycophenolate mofetil must be considered in the treatment pathway and overall immunosuppression burden.

Ocrelizumab, another fully humanized anti-CD20 monoclonal antibody, FDA approved for severe multiple sclerosis, has the advantage of avoiding first-dose infusion reactions and the development of human antichimeric antibodies that are potential problems with rituximab therapy. However, a controlled randomized trial using this agent in patients with LN was terminated early due to adverse events.

Atacicept, a soluble fully humanized recombinant fusion protein that inhibits B-cell stimulating factor (BLISS) and a proliferation-inducing ligand (APRIL), failed in initial trials with patients with LN.

Epratuzumab, a humanized monoclonal antibody against CD22, a marker of mature B cells but not plasma cells, is currently being studied. IFN-α is an antiviral cytokine induced in SLE patients that may lead to inflammation, autoimmunity, and renal damage.

Anifrolumab, a monoclonal antibody against the IFN-α type I receptor, has been shown in a trial of more than 300 SLE patients to add to the efficacy of standard therapy.

Chimeric antigen receptor (CAR)-T cell therapy is a promising therapy for SLE. Mackensen and colleagues conducted a phase I study of 5 patients with refractory SLE who received CAR-T cell therapy. Autologous T cells were transduced with a lentiviral anti-CD19 CAR vector and reinfused after lymphodepletion with fludarabine and cyclophosphamide. This resulted in a depletion of B-cells, improvement in clinical symptoms, and normalization of laboratory values. All five patients achieved remission by 3 months. This resulted in a decrease in the median SLEDAI (Systemic Lupus Erythematosus Disease Activity Index) score from 16 to 0. Treatment was well tolerated with only a mild cytokine release syndrome. Follow-up did not show any relapses of SLE while still being off any SLE-associated medications. A 24-month update of these 5 patients, with an additional 3 patients, was released. This update demonstrated a sustained reduction in proteinuria. Five of the 8 patients developed grade 1 cytokine release syndrome, for which 3 patients received tocilizumab. Given these promising results, other CAR-T cell therapies are in clinical trials for patients with SLE. Given that cytokine release syndrome can be fatal, larger-scale studies are needed to confirm the safety in addition to efficacy. These results also indicate, as mentioned earlier, that targeting B cells appears to be a promising therapy in LN. Numerous other therapies targeting B cells are under investigation for LN. ^,

T-lymphocyte activation requires two signals. ^, The first occurs when the antigen is presented to the T-cell receptor via major histocompatibility complex (MHC) class II molecules on antigen-presenting cells. The second mechanism of activation is by the interaction of costimulatory molecules on T lymphocytes and antigen-presenting cells. Disruption of costimulatory signals interrupts the (auto)immune response. Two clinical trials using different humanized anti-CD40L monoclonal antibodies in LN patients to block B and T-cell costimulation have not been successful. ^, Another costimulatory pathway is mediated through the interaction of CD28 with CD80/86. CTLA4 Ig, abatacept, a fusion molecule that combines the extracellular domain of human CTLA4 with the constant region (Fc) of the human IgG1 heavy chain, interrupts the interaction betweenCD28 and CD80/86. CTLA-4 is similar to the T-cell costimulatory protein CD28, and both bind to CD80 and CD86, also known as B7-1 and B7-2, on antigen-presenting cells. It is FDA approved for the treatment of RA. Two major randomized controlled trials with abatacept in patients with severe LN treated concurrently with intravenous cyclophosphamide and steroids were negative. ^{, ,}

Definitions of remission differ among clinical studies and published guidelines. KDIGO defines a “complete remission” as reduction in urine PCR (from 24-hour urine collection) to <442 mg/g (50 mg/mmol) and stabilization of kidney function within 10% to 15% of baseline within 6 to 12 months of initiating treatment. A novel treatment response definition added to the 2024 KDIGO update is the criterion of “primary efficacy renal response,” which is defined by a UPCR <0.7 g/g (70 mg/mmol), eGFR that is no worse than 20% below the preflare value or ≥60 mL/min/1.73 m ² , and no use of rescue therapy for treatment failure. This proteinuria remission target of <0.7 g/g has also been described as a good predictor of long-term kidney outcomes in other glomerular diseases. Joint statements from the EULAR, European Renal Agency and European Dialysis and Transplant Association (ERA/EDTA) define a “complete clinical response” as proteinuria reduction to <0.5 to 0.7 g/day by 12 months. In addition, they suggest that improvement in proteinuria should be noted by 3 months, with at least a 50% reduction by 6 months. Given that proteinuria may take months to years to improve, assessments of other biomarkers and clinical symptoms factor into the treatment response during this time. Resolution of microscopic hematuria has only variably been included in definitions of treatment response, while remission of microhematuria likely correlates with reduction in inflammation, its association with improved long-term kidney outcomes has not been demonstrated in LN.

Persistent proteinuria after treatment induction and maintenance may be due to ongoing glomerular inflammation or glomerular scarring. Although serology and symptoms of SLE may aid in this assessment, there may be uncertainty when proteinuria is discordant to serology and symptoms. The role of repeat kidney biopsies remains controversial, particularly given the described discrepancy with clinical response for some patients. A histologic activity index may take months to years to resolve, while chronic damage accrues with recurrent episodes of LN. Data continue to emerge regarding the optimal timing of repeat kidney biopsies. Therefore there may be a greater utility of repeat biopsies when there hasn’t been a clinical remission, there is concern about ongoing immunologic activity, and sufficient time has passed to allow resolution of histologic activity parameters after treatment was initiated. Ongoing studies will hopefully shed light on the sequence of immunologic, clinical, and histologic remission as has been characterized in other glomerular diseases, to allow either intensification or deescalation of immunosuppressive therapies.

In patients not able to achieve remission, verifying adherence remains pivotal, as outlined by the KDIGO guidelines. Nonadherence is typically defined by <80% of the prescribed doses taken by patients. Measurement of nonadherence may be challenging with self-administered questionnaires or even clinician assessments, and therefore more objective measures such as drug levels or drug dispensing records may be needed. Patients with SLE and LN have high rates of nonadherence, which in turn leads to worse renal outcomes. Nonadherence to hydroxychloroquine has been associated with SLE flares and mortality. If available, monitoring of hydroxychloroquine levels can serve as a measure of nonadherence. Many factors contribute to barriers to adherence, comprising those that are patient related, health care related, medication related, and disease related. ^, Patient priorities and self-reported symptoms being perceived as less important by physicians have been identified as a recurring barrier. Numerous interventions for improving adherence have been proposed including improved communication among health care professionals and patients, patient education, regularly addressing adherence, involvement of caregivers, and well-coordinated transition from pediatric to adult care. Specifically pertaining to communication and education, patients would like more weight and discussion given to self-reported symptoms and quality of life in medication decisions. This counseling can be done in the context of the disease with appropriate and clear explanation of risks versus benefits of treatment. When performed regularly and with involvement of caregivers, this approach may improve rates of nonadherence. Many patients with lupus may not be aware that it is a potentially fatal disease and stop taking their medications at the first sign of a side effect, such as acne or hair growth. Education of patients about the seriousness of their illness to encourage adherence to their immunosuppression is important. Another way to try to improve adherence is to be mindful of the number of medications prescribed, particularly when patients have identified this to be an important issue. While it is reasonable to prescribe lipid-lowering medicine, RAS inhibition to lower proteinuria, and now the consideration of SGLT2 inhibitors, these maneuvers, unproven in lupus nephritis, must be weighed against the total burden of medications prescribed to these typically young patients and may be better reserved outside the induction immunosuppression window.

The incidence of APS is estimated to be around 5 cases per 100,000 people per year, and its prevalence is approximately 40 to 50 cases per 100,000 people. Among patients with APL antibodies, 30% to 55% have primary APS in which there is no associated autoimmune disease. ^{, , , ,} APL antibodies are found in 25% to 75% of SLE patients, although most patients never experienced clinical features of the APS. A variable percentage (0%–23%) of patients in different series initially felt to have idiopathic APS have evolved over time into SLE-associated APS. In an analysis of 29 published series with more than 1000 SLE patients, 34% were positive for the lupus anticoagulant and 44% for anticardiolipin antibodies. Most studies have found a higher incidence of thrombotic events in SLE patients positive for APL antibodies. Lupus anticoagulants may carry a greater risk of thrombosis than anticardiolipin antibody, with the highest risk for those with triple positivity. ^, A European study of almost 575 SLE patients found the prevalence of IgG anticardiolipin antibodies to be 23% and of IgM 14%. Patients with IgG antibodies had a clear association with thrombocytopenia and thromboses. A multicenter European analysis of 1000 SLE patients found thromboses in 7% of patients over 5 years. Patients with IgG anticardiolipin antibodies again had a higher incidence of thromboses, as did those with a lupus anticoagulant. APL antibodies are also found in up to 2% of normal individuals and in those with a variety of infections (commonly in patients with HIV or hepatitis C virus) and drug reactions, but these are not usually associated with the clinical features of the APS.

Susceptible individuals may develop APL antibodies after exposure to infectious or other noxious agents. Among SLE patients there may be a genetic predisposition associated with HLA-DRB1 loci. However, despite the presence of APL antibodies, a “second hit” (such as pregnancy, contraceptive use, nephrotic syndrome, SLE, or hyperlipidemia) may be necessary for thrombotic events. RNA-sequencing from kidney biopsy samples for patients with APS reveal increased expression of interferon, NET-related genes, and complement. These results suggest innate immunity may be playing a role in the pathogenesis. The mechanism(s) of the procoagulant effect are likely multifactorial. APL antibodies exert procoagulant effects at multiple sites in the clotting cascade including prothrombin, protein C, annexin V, factors VII and XII, platelets, serum proteases, and tissue factor procoagulant. They may also impair fibrinolysis through inhibition of factors such as tissue-type plasminogen activator. APL antibodies may also be procoagulant due to inhibition of the mechanistic target of rapamycin (mTOR) complex intracellular pathway. mTOR, also previously known as the mammalian target of rapamycin or the FK506-binding protein 12-rapamycin-associated protein 1 (FRAP1), is a kinase encoded by the MTOR gene in humans. mTOR belongs to the phosphatidylinositol 3-kinase-related kinase family of protein kinases. As noted earlier, neutrophil extracellular traps (NETs) are scaffolds of decondensed chromatin and are thought to be involved in triggering thrombosis through activation of coagulation factors, endothelium, platelets, and the complement system. Other studies support involvement of the classical, lectin, and alternative complement systems. Many patients have hypocomplementemia suggestive of complement protein consumption. The result of all these abnormalities is endothelial damage and intravascular coagulation. In cases of pregnancy-associated APS, the complications may be mediated predominantly by trophoblast dysfunction and complement activation rather than by thrombosis.

The clinical features of APS relate to thrombotic events and consequent ischemia. Among 1000 APS patients the most common features were deep vein thrombosis (32%), thrombocytopenia (22%), livedo reticularis (20%), stroke (13%), pulmonary embolism (9%), and fetal loss (9%). Patients may also experience pulmonary hypertension, cardiac involvement, memory impairment and other neurologic manifestations, fever, malaise, and constitutional symptoms. ^{, ,} As mentioned earlier, patients who test positive for all three diagnostic tests (lupus anticoagulant, anticardiolipin antibodies, and β ₂ -glycoprotein antibodies) are at higher risk for thromboembolic events. Catastrophic APS (CAPS), a rare event (occurring in <1% of APS patients), is associated with rapid development of thromboses in multiple organ systems and has a high fatality rate. ^{, ,}

Renal involvement is generally an uncommon finding in patients with APS. ^, Reports of renal involvement vary from as low as 3% to as high as 25% of patients with primary APS. ^{, ,} Kidney involvement ranges from large vessels, such as renal artery stenosis or renal vein thrombosis, to small vessel involvement including glomerular thrombotic microangiopathy ( Fig. 31.12 ). ^{, , ,} Lesions involving the arteries and arterioles often have both an acute thrombotic component and a reactive or proliferative chronic one with intimal mucoid thickening, subendothelial fibrosis, and medial hyperplasia. ^, Interstitial fibrosis and cortical atrophy may occur due to tissue ischemia. Glomerular lesions include glomerular capillary thrombosis with associated mesangiolysis, mesangial interposition and duplication of GBMs, and subendothelial accumulation of electron-lucent, flocculent material (“endotheliosis”), resembling the changes in other forms of glomerular thrombotic microangiopathy such as HUS and TTP.

Antiphospholipid nephropathy.

(A) Severe fibrous intimal hyperplasia lesion in a native kidney for a patient with APS. (B) Thrombotic microangiopathy lesions in patients with APS (arrows).

Among APL-positive SLE patients, APL-nephropathy was found in two-thirds of those with APS and in one-third of those without APS. Although patients with APL-nephropathy had a higher frequency of hypertension and elevated serum creatinine levels at biopsy in this series, they did not have a higher frequency of progressive renal insufficiency, ESKD, or death at follow-up. This is in contrast to another series of more than 100 SLE patients, which found the presence of APL antibodies to be associated with both thrombotic events and a greater progression to renal failure. In patients with APL-nephropathy, renal biopsies with thrombotic microangiopathy may be misclassified as FSGS, membranous nephropathy, and membranoproliferative GN. However, some patients with APS may develop a number of other glomerular histologic patterns on LM examination including membranous nephropathy, minimal change/focal sclerosis, mesangial proliferative glomerulonephritis, and pauci-immune rapidly progressive glomerulonephritis (RPGN).

Kidney biopsies can be high risk in those presenting with APS because of thrombocytopenia or the need for continuous anticoagulation. Holding anticoagulation before and after a kidney biopsy, even if only for a few days, may be enough to activate systemic thrombosis in susceptible patients, and there is a risk of late bleeding with resumption of anticoagulation. Therefore a risk-benefit analysis should be undertaken before considering kidney biopsy.

The most frequent clinical renal findings are proteinuria, at times in the nephrotic range, active urinary sediment, hypertension, and progressive renal dysfunction. ^{, , , ,} Some patients present with an acute deterioration in renal function. With major renal arterial involvement, there may be renal infarction and renal vein thrombosis may be silent or present with sudden flank pain and a decrease in renal function. Renal artery stenosis has been reported with and without malignant hypertension.

About 10% of biopsied patients with LN have glomerular microthromboses as the major histopathologic finding. Therapy of this glomerular lesion clearly differs from that of immune complex–mediated glomerulonephritis. One study of 114 biopsied SLE patients found vasoocclusive lesions in one-third of biopsies, which correlated with both hypertension and increased serum creatinine. In SLE, features that correlate well with high titers of IgG APL antibodies are thrombocytopenia, the presence of a false-positive VDRL for syphilis, and a prolonged activated partial thromboplastin time. ^{, ,} Neither the titer of anti-dsDNA antibodies nor the serum complement levels correlate well with the APL antibody levels. In SLE, high titers of IgG anticardiolipin antibody usually correlate well with the risk of thrombosis. However, in one study of 114 biopsied SLE patients, renal thrombi were related to lupus anticoagulant but not anticardiolipin antibodies. The clinical features of APS in SLE patients are identical to those of primary APS. An important study documents the prevalence of APL antibodies in 26% of 111 patients with LN followed for a mean of 173 months. Of the APL antibody–positive patients, 79% developed a thrombotic event or fetal loss, and the presence of antibodies was strongly correlated with the development of progressive CKD.

Optimal treatment of patients with APL antibodies and the APS remains to be defined. ^{, ,} Many patients with APL antibodies do not experience thrombotic events. On the basis of limited data in asymptomatic patients with APL antibodies but no evidence of thrombotic events or the APS, low-dose aspirin as primary prophylaxis may be beneficial. EULAR 2023 recommendation updates for the management of APS suggest consideration of low-dose aspirin for asymptomatic patients with SLE or a high-risk profile (lupus anticoagulant, double/triple positivity, or persistently high aPL titers). ^,

Because patients with higher titers of IgG APL antibody have a greater incidence of thrombotic events, they may benefit from anticoagulation. ^, In patients with full APS, anticoagulation with heparin followed by warfarin has proven more effective in preventing recurrent thrombosis than no therapy, aspirin, or low-dose anticoagulation. Studies in patients with predominantly venous events suggest no additional benefit with a target INR of 3 to 4 versus 2 to 3. ^, It is unclear if this higher INR target would be of benefit for patients with arterial, as opposed to venous, thrombotic events. Of note, many patients did not routinely achieve the higher target INR and very few patients with arterial thrombosis were included. Preliminary studies suggest that the newer direct oral anticoagulants (DOACs) may be inferior to warfarin for the prevention of thrombotic events in patients with APS. One meta-analysis suggests that the use of DOAC versus vitamin K antagonists is associated with increased odds of subsequent arterial thrombotic events (OR 5.43), especially stroke. The risk of major bleeding between the two was similar in this study. Therefore the 2023 EULAR guideline updates recommend that therapy with DOACs only be considered as second-line treatment if target INR 2 to 3 cannot be achieved despite good adherence to warfarin. In the event of recurrent thrombosis despite warfarin and achievement of target INR, treatment considerations may include adding aspirin, increasing target INR to 3-4, or switching to low-molecular-weight heparin. The role of immunosuppression is uncertain in APS. ^{, ,}

In patients with SLE, the anti-dsDNA antibody titer and serum complement may normalize with immunosuppression without a significant change in a high titer of IgG APL antibody. In rare patients who cannot tolerate anticoagulation due to bleeding, who have thromboembolic events despite adequate anticoagulation, or who have catastrophic APS, plasmapheresis with corticosteroids and other immunosuppressives have been used with some success. ^, It is uncertain whether hydroxychloroquine, used mostly in SLE patients, can prevent thromboembolic events in APS. ^{, ,} There is insufficient and conflicting data on whether newer agents such as rituximab lower the levels of APLs or decrease the risk of thromboembolism. ^, Eculizumab and the longer-acting ravulizumab have been reported in refractory cases of catastrophic APS. Outcomes may be improved with earlier administration of these anticomplement therapies.

Other therapies have also been described in APS. In recent years belimumab has been reported to reduce aPL levels in patients with SLE. Currently there is a pilot trial under way investigating belimumab for refractory or noncriteria manifestations of APS. HMG-CoA reductase inhibitors (statins) have demonstrated reductions in thrombus size for mice with APS, and mechanistic studies in humans demonstrate a favorable proinflammatory and prothrombotic response. However, efficacy of statins for reduction of thrombotic events for patients with APS has not been proven. The use of other treatments, such as intravenous γ-globulin and stem cell transplant, are reported only in isolated patients. ^,