Diagnosis and treatment of lupus nephritis: a summary of the Consensus Document of the Spanish Group for the Study of Glomerular Diseases (GLOSEN)

ABSTRACT

Lupus nephritis (LN) is the most frequent serious manifestation of patients with systemic lupus erythematosus (SLE). Up to 60% of SLE patients develop LN, which has a significant impact on their quality of life and prognosis. Recent advances have improved the diagnostic approach to LN, and new drugs that block specific pathways and kidney damage progression have been developed. Several randomized and well-powered clinical trials have confirmed the efficacy of these agents in terms of proteinuria remission and preservation of kidney function in the medium and long term, with an acceptable safety profile and good tolerance. The combination of different therapies allows for reduction of the dose and duration of corticosteroids and other potentially toxic therapies and leads to an increase in the number of patients achieving complete remission of the disease. This consensus document carried out by the Spanish Group for the Study of Glomerular Diseases (GLOSEN) provides practical and updated recommendations, based on the best available evidence and clinical expertise of participating nephrologists.

INTRODUCTION

Systemic lupus erythematosus (SLE) is the paradigm of systemic autoimmune disease. A significant number of SLE patients develop lupus nephritis (LN), with serious consequences for their prognosis and quality of life. In recent years several important studies have demonstrated the efficacy of new drugs and therapeutic combinations, providing a greater precision in the definition of treatment objectives and prognostic markers.

We think that a consensus document that collects, analyzes and summarizes all this new information in an easy-to-read and practical document will be useful for a better management of LN patients. This was the objective of the present work, the complete form of which has been published previously [1].

In this summarized version we present the most important parts of the consensus document, focused on the diagnosis, definitions, objectives and treatment of the different classes of LN and some special situations. Each section includes some brief recommendations, followed by a rationale. The purpose of the tables and figures is to summarize and illustrate the content of the recommendations.

Excellent and comprehensive documents and guidelines on LN management have been published in recent years, such as the 2019 Update of the Joint European League Against Rheumatism and European Renal Association–European Dialysis and Transplant Association (EULAR/ERA–EDTA) recommendations for the management of lupus nephritis [2] and the KDIGO 2021 Clinical Practice guideline for the management of Glomerular Diseases [3]. The central aims of our consensus document were to adapt the different combinations of drugs, the efficacy of which has been demonstrated in recent trials, to the different clinical profiles and special situations of LN, as well as to summarize these treatment regimens in algorithms and tables that are easy to read and translate to clinical practice. The document was prepared by Spanish Group for the Study of Glomerular Diseases (GLOSEN) nephrologists with extensive experience in the diagnosis and treatment of LN patients. Several virtual and face-to-face meetings were held for coordination, section assignments and content discussion. An exhaustive and systematic search of the literature was carried out, which included, among others, the following databases: PubMed, EMBASE, Cochrane Library, Google Scholar and ClinicalTrials.gov, as well as the abstract books of national and international congresses. Details about search strategy and levels of evidence of the selected studies are provided in the supplementary material of the complete version [1].

1. DIAGNOSIS

Recommendations

• 1.1. The diagnosis of SLE should be based on the characteristic signs and symptoms of the disease, according to the criteria established by the 2019 EULAR/American College of Rheumatology classification [4].

• 1.2. In patients with SLE, a determination of serum creatinine, estimated glomerular filtration rate (eGFR) and a urinalysis [urinary albumin/creatinine ratio (uACR) and/or urinary protein/creatinine ratio (uPCR), urinary sediment] should be performed periodically. These determinations should be more frequent (at least once or twice a year) in those patients with a higher propensity to develop LN (persistent extrarenal clinical manifestations, positivity of serological markers, early onset of SLE and non-Caucasian races).

• 1.3. In a significant proportion of patients, LN is asymptomatic and is detected only by analytical data. Special attention should be paid to the appearance of edema and/or de novo hypertension in SLE patients.

• 1.4. Kidney biopsy is needed for the diagnosis and classification of LN and should be evaluated by expert nephropathologists. It is recommended in patients with SLE and proteinuria >0.5 g/24-h (or uPCR >0.5 g/g). The presence of hematuria, leukocyturia, or urinary casts and/or unexplained impairment of renal function further reinforces the biopsy indication. Patients without proteinuria but with an active urinary sediment and/or impairment of renal function require a careful study to rule out etiologies other than SLE before performing a kidney biopsy.

Rationale

LN represents one of the most frequent and serious complications of SLE [5–7]. Early diagnosis is crucial and, since LN is asymptomatic in many patients [8], periodic blood and urine analysis are recommended. A higher prevalence of atherosclerotic complications in patients with SLE has been reported [9, 10], so regular determination of kidney function and albuminuria/proteinuria, factors that are clearly linked to an increased cardiovascular risk, are essential even in inactive lupus patients. The periodicity of these determinations must be guided by patient characteristics. The incidence of LN is higher in patients with extrarenal lupus manifestations, early onset of SLE and persistent serological markers of immunological activity (hypocomplementemia, anti-dsDNA positivity), and in non-Caucasians, so analytical determinations should be performed more frequently in such patients [5–7, 10, 11]. A significant proportion of patients are asymptomatic at the time of LN diagnosis. In such cases, the suspicion of LN is raised by abnormalities detected in urine and blood laboratory tests [12, 13].

Kidney biopsy is mandatory for LN diagnosis. It is recommended in patients with SLE with persistent urinary abnormalities not attributable to other causes or with unexplained impairment of renal function [14–17]. The 2018 International Society of Nephrology/Renal Pathology Society (ISN/RPS) LN classification includes a semiquantitative analysis of active and chronic lesions having prognostic value [18]. Description of vascular and interstitial lesions should be also included [19–21]. Other lesions of considerable prognostic and therapeutic importance may also be found, such as thrombotic microangiopathy (TMA), vasculitis, podocytopathy and acute tubulointerstitial nephritis [15–17].

The histological study requires light microscopy and immunofluorescence techniques, and electron microscopy is recommended. The biopsy should be interpreted by expert nephropathologists. Figure 1 summarizes the histological classes proposed by the latest revision of the ISN/RPS [14, 18] classification.

Figure 1:

Indications of kidney biopsy in patients with SLE and major histological classes according to the 2018 ISN/RPS classification.

2. MONITORING OF LUPUS NEPHRITIS AFTER DIAGNOSIS

Recommendations

• 2.1. The identification, assessment and follow-up of extrarenal clinical manifestations is essential in patients with LN. The Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) scoring system is recommended to monitor these manifestations.

• 2.2. In LN patients, the amount of proteinuria (expressed as uPCR or as 24-h proteinuria), the urinary sediment, and the determination of serum creatinine and eGFR are the most important parameters to assess disease evolution and treatment efficacy.

• 2.3. Some factors independent of SLE activity can influence proteinuria amount, urinary sediment and kidney function, and should be carefully evaluated (Table 1). The amount of proteinuria does not always have a direct relationship with LN activity, especially in relapsing patients and in those with significant chronic lesions.

Table 1:

Main clinical factors that can influence proteinuria levels, urine sediment and kidney function independently of lupus activity.

Renal parameter	Modifying factor or condition
Proteinuria	Weight changes (increased proteinuria with weight gain, decreased with weight loss)
	BP changes (increased proteinuria with high blood pressure)
	Kidney function changes (increased proteinuria with a higher eGFR, decreased with worsening kidney function)
	Changes in the dose of drugs modifying glomerular hemodynamics:
	‒ RAS blockers (ACEi, ARB)
	‒ Inhibitors of the glucose-sodium tubular cotransporter 2 (SGLT2i)
	‒ Aldosterone antagonists (spironolactone, eplerenone)
	‒ Diuretics (thiazides, amiloride)
	‒ Fibrates
	‒ Endothelin antagonists
Urinary sediment	Urinary infections
	Renal lithiasis, crystalluria, tumors or malformations of the urinary tract
	Familial persistent microhematuria/genetic abnormalities of collagen IV
Kidney function	eGFR reduction at the beginning of some renoprotective treatments (ACEi, ARB, SGLT2i)
	eGFR reduction due to diuretics
	eGFR reduction due to excessive blood pressure control
	AKI due to causes different of active LN

• 2.4. There is no general agreement on the indications for repeating kidney biopsy in LN. It could be considered in refractory patients, in patients with sustained proteinuria to evaluate the predominance of active or chronic lesions, in some LN relapses with diagnostic or therapeutic doubts or when a condition unrelated to LN is suspected. Repeated biopsy could be also useful before the withdrawal of immunosuppression.

Rationale

Monitoring of extrarenal systemic manifestations is essential in any patient with LN. The SLEDAI scoring system allows evaluation of such activity with a systematic and reproducible approach [22].

The amount of proteinuria is the most important marker for prognosis and for treatment response assessment in LN [23–26]. A precise measurement, that should be expressed as 24-h urinary protein excretion or as uPCR in an isolated urine sample, is essential. To analyze urinary sediment, automatized techniques have replaced the traditional manual techniques and have been validated for the routine monitoring of patients [27, 28]. For eGFR calculation, the Chronic Kidney Disease Epidemiology Collaboration formula is recommended. However, not all changes in analytical parameters are related to immunological lupus activity (Table 1). Changes in body weight have an important influence on proteinuria. An impairment in renal function not accompanied by increases in proteinuria or changes in urinary sediment should force the physician to consider factors independent of LN activity, as well as changes in the urinary sediment without increases/reappearance of proteinuria and without changes in kidney function.

3. DEFINITIONS AND TREATMENT OBJECTIVES

3.1. Definition of complete remission, partial remission and no response

Recommendations

1. 3.1.1. The efficacy of any therapeutic strategy for LN must be evaluated by obtaining a complete remission (CR) or partial remission (PR), as defined in Table 2.

Table 2:

Definitions of renal outcomes: CR, PR, non-response and relapse.

Outcome	Definition
CR	Proteinuria ≤0.5 g/24 h or uPCR ≤0.5 g/g
	Inactive urinary sediment (≤5 RBC/hpf)
	Serum albumin ≥3.5 g/dL
	Normal eGFR or ≤10% inferior to baseline values
PR	Proteinuria reduction ≥50% with values between 0.6 and 3.5 g/24-h or uPCR 0.6–3.5 g/g
	Hematuria reduction (≤10 RBC/hpf)
	Serum albumin ≥3 g/dL
	Normal eGFR or ≤25% inferior to baseline values
Non-response	Absence of CR or PR
Relapse	Recurrence or significant increase in hematuria (>15 RBC/hpf) with dysmorphic RBC and/or RBC casts
	Sustained proteinuria increase:
	‒ ≥1 g/24-h or ≥1 g/g in patients with CR
	‒ ≥50% in patients with PR
	eGFR reduction ≥25% not attributable to other causes

RBC/hpf: red blood cells per high-power field.

• 3.1.2. Time elapsed until obtaining a CR or PR is an important prognostic marker. The timeline to achieving objectives must be established according to individual patient's characteristics (Fig. 2). The optimal goal is to achieve CR, since long-term kidney survival is significantly better than that of patients who achieve PR alone.

Figure 2:

Objectives in the treatment of LN. RBC/hpf: red blood cells per high-power field.

Rationale

There is no universally agreed definitions of CR or PR, although all the guidelines coincide in proteinuria reduction and the recovery of kidney function as fundamental objectives [3, 25, 29, 30]. Although normalization/increase in serum albumin is not usually included in the definitions of CR or PR, we prefer to include this parameter considering that hypoalbuminemia can be found in some patients with proteinuria levels below the nephrotic range (3.5 g/24 h) (Expert opinion). Improvement or normalization of the urinary sediment is not included within CR or PR criteria in some guidelines [29]. We prefer to maintain this criterion, since reappearance of hematuria can be a sensitive marker of clinical activity, especially in patients with residual proteinuria associated with chronic lesions. A >50% reduction in proteinuria at 6 months is associated with a better kidney survival and a proteinuria level <0.7 g/24-h at 12 months is an excellent predictor of a good long-term kidney outcome [23, 24].

In addition to obtaining CR or PR, the time elapsed until reaching these objectives is important. The evolution of clinical and analytical parameters must be carefully evaluated at every visit, verifying their progressive improvement (Fig. 2). Such evolution should be adapted to the individual patient's characteristics, especially in the most aggressive forms of presentation [severe nephrotic syndrome, acute kidney injury (AKI)] [25, 31, 32]. In this type of patient, a slower achievement of the objectives shown in Fig. 2 can be acceptable, as long as a clear and progressive clinical and analytical improvement is evident.

3.2. Definition of relapse (flare)

Recommendation

• 3.2.1. A substantial number of patients with LN present relapses after reaching CR or PR. Definition of relapse is shown in Table 2.

Rationale

The incidence of LN relapses oscillates between 10% and 50%, according to different studies. The risk is higher among patients who reach only PR and during the first years after the first episode (especially when immunosuppression is reduced or discontinued) [33–35]. However, relapses can occur at any time, even after decades of inactivity. For this reason, long-life periodic revisions are recommended in any LN patient. Most relapses are detected by the reappearance or increase in hematuria and proteinuria (Table 2). In the most severe cases, these changes in hematuria and proteinuria can be accompanied by a worsening of kidney function. However, it is exceptional to observe kidney function impairment not accompanied by proteinuria or urinary sediment abnormalities (hematuria, leukocyturia, red blood cell casts), so other causes of kidney function worsening not related to lupus activity should be ruled out (Table 1). The diagnosis of relapse is more difficult in patients with PR and in those with repeated flares of LN that have caused extensive chronic lesions and residual proteinuria. In these cases, a new renal biopsy can provide useful information about the presence of histological activity [36–39].

4. TREATMENT OF LUPUS NEPHRITIS: INDUCTION AND MAINTENANCE

4.1. Classes I and II (Fig. 3)

Figure 3:

Treatment of classes I and II. CV: cardiovascular.

Recommendations

• 4.1.1. In patients with LN class I, we recommend treatment with hydroxychloroquine (HCQ), general nephroprotection and general treatment of SLE according with extrarenal manifestations.

• 4.1.2 In patients with LN class II with proteinuria <1 g/24-h and normal urinary sediment, we recommend same treatment as in LN class I.

• 4.1.3 In patients with LN class II with proteinuria >1 g/24-h despite optimal renin–angiotensin system (RAS) blockade and/or urinary sediment showing glomerular hematuria, we recommend a second assessment of kidney biopsy or to perform a new biopsy to rule out changes in histological class or a concurrent lupus podocytopathy. If class II is confirmed, we suggest adding corticosteroids and mycophenolic acid analogues (MPAA) for 6–12 months to the general treatment, with subsequent gradual tapering according to evolution. If lupus podocytopathy is found, a therapeutic approach similar to minimal change disease is suggested (see Section 4.5).

Rationale

Classes I and II LN are considered mild forms of renal involvement with a good prognosis [40]. There are no controlled studies supporting a specific therapeutic regimen, although the use of HCQ is recommended. Patients with mild proteinuria (<1 g/24-h) can be managed conservatively with optimal RAS blockade [angiotensin-converting enzyme inhibitors (ACEi) or angiotensin receptors blockers (ARB) regardless of blood pressure (BP)], due to their renoprotective and anti-inflammatory effects [40, 41].

Additionally, a strict control of BP, dyslipidemia and other cardiovascular risk factors is mandatory. It is important to keep in mind that LN classes can change over time. The presence of proteinuria >1 g/24-h despite optimized RAS blockade, with or without hematuria or kidney function impairment, should prompt the revision of kidney biopsy or the performance of a new one to rule out changes to more severe histological classes [36].

Scarce information exists about the management of patients with confirmed LN class II who maintain proteinuria >1 g/day despite optimized RAS blockade and/or glomerular hematuria. Some studies have shown favorable results of immunosuppressive treatment (corticosteroids and MPAA for 6–12 months) [42, 43], although no controlled studies have been published.

4.2. Classes III/IV ± V: initial treatment (Fig. 4)

Figure 4:

Initial treatment for classes III/IV ± V. HCQ: initial dose 4–5 mg/kg/day (maximum 400 mg daily). If eGFR <30 mL/min/1.73 m², do not exceed 200 mg daily. MPAA: dose equivalent to 2 g of MMF. Cyclophosphamide: Eurolupus regimen preferred (6 biweekly i.v. pulses of 500 mg each, cumulative dose = 3 g). CNI: cyclosporine (initial dose 100–200 mg daily, blood target level 60–100 ng/mL), tacrolimus (initial dose 0.05–0.07 mg/kg/day, blood target level 4–7 ng/mL), voclosporine (23.7 mg twice daily for 1–2 years, no blood target level required). Belimumab: 10 mg/kg i.v. on days 1, 15 and 29, then every 28 days. i.v.: intravenous.

Recommendations

• 4.2.1. All patients with LN classes III, IV or III/IV + V should receive corticosteroids as initial treatment (unless contraindicated) along with other immunosuppressants. As a preferred corticosteroid regimen, we suggest intravenous pulses of methylprednisolone (250–500 mg/day) for three consecutive days followed by a reduced-dose regimen (oral prednisone or equivalent 0.5–0.6 mg/kg/day).

• 4.2.2. In patients with proteinuria <3 g/24-h, good therapeutic compliance, infertility concerns, or contraindications or intolerance to cyclophosphamide, we suggest double initial immunosuppressive treatment with corticosteroids and MPAA [in doses equivalent to 2 g/day of mycophenolate mofetil (MMF)].

• 4.2.3. In cases with proteinuria <3 g/24-h, risk of therapeutic non-compliance, or contraindications or intolerance to MPAA, we suggest double initial immunosuppressive treatment with corticosteroids and intravenous cyclophosphamide (Eurolupus regimen). After the administration of the sixth pulse of cyclophosphamide, treatment with MPAA will be started (or azathioprine in case of intolerance to MPAA).

• 4.2.4. In those patients who have not achieved a reduction in proteinuria of at least 25% after 2–3 months of treatment with corticosteroids + MPAA or at the end of the cyclophosphamide boluses, we suggest adding belimumab to the treatment (especially if immunological activity persists), or a calcineurin inhibitors (CNI) (especially if significant proteinuria persists).

• 4.2.5. We suggest initial triple immunosuppressive therapy with corticosteroids + MPAA + belimumab in those patients fulfilling the profile described in recommendation 4.2.2, but presenting in addition extrarenal SLE manifestations, intense serological activity, need to rapidly reduce corticosteroids or previous LN relapses.

• 4.2.6. In those patients with proteinuria >3 g/24-h or complete nephrotic syndrome and fulfilling otherwise the other characteristics described in recommendation 4.2.2, we suggest initial triple immunosuppressive therapy with corticosteroids + MPAA + CNI (cyclosporine, tacrolimus, voclosporine), provided that eGFR is ≥45 mL/min/1.73 m².

• 4.2.7. In cases with acute deterioration of kidney function, we suggest applying therapeutic schemes adapted to the patient's clinical profile (recommendations 4.2.1 to 4.2.6) but avoiding CNI in patients with eGFR <45 mL/min/1.73 m².

Rationale

Intravenous pulses of corticosteroids significantly improve acute inflammation and immune activity in LN [44, 45]. There is a current trend to use lower doses of oral corticosteroids following intravenous pulses, due to the lower risk of adverse effects [3, 25, 46]. The Aurinia Renal Response in Active Lupus with Voclosporin (AURORA)-1 study confirmed the efficacy and safety of low-dose corticosteroid regimens, reaching 2.5 mg/day at week 12 after starting treatment [47] (Table 3). However, other regimens with higher doses of corticosteroids can be considered in patients with more aggressive presentations or worse responses to treatment (Table 3).

Table 3:

Recommended dose of corticosteroids for the treatment of LN.^a

Regimen	Standard dose	Moderate dose	Low dose
Methylprednisolone (i.v., up to 3 days)	0.25–0.5 g/day	0.25–0.5 g/day	0.25–0.5 g/day
Prednisone (oral, mg/day)
Week 0–2	0.8–1.0 mg/kg (maximum 80 mg)	0.6–0.7 mg/kg (maximum 50 mg)	0.5–0.6 mg/kg (maximum 40 mg)
Week 3–4	0.6–0.7 mg/kg	0.5–0.6 mg/kg	0.3–0.4 mg/kg
Week 5–6	30 mg	20 mg	15 mg
Week 7–8	25 mg	15 mg	10 mg
Week 9–10	20 mg	12.5 mg	7.5 mg
Week 11–12	15 mg	10 mg	5 mg
Week 13–14	12.5 mg	7.5 mg	2.5 mg
Week 15–16	10 mg	7.5 mg	2.5 mg
Week 17–18	7.5 mg	5 mg	2.5 mg
Week 19–20	7.5 mg	5 mg	2.5 mg
Week 21–24	5 mg	<5 mg	2.5 mg
Week >25	<5 mg	<5 mg	<2.5 mg

^aBased on references [3] and [47].

i.v.: intravenous administration

MPAA represent the first line of treatment in compliant patients with proliferative or mixed LN (Fig. 4). As induction therapy, they show similar [2, 48, 49] or superior efficacy to cyclophosphamide [50], with a better safety profile. This benefit has been observed in different ethnic groups [51].

Cyclophosphamide is still a first-line therapeutic option, especially in noncompliant patients or when contraindications to MPAA exist [52, 53]. Intravenous administration is preferred over oral route due to its lower risk of adverse effects and lower cumulative dose [54, 55].

The Euro-Lupus Nephritis Trial, conducted in European and predominantly Caucasian patients with LN classes III, IV or V, demonstrated that low doses of cyclophosphamide (six fortnightly pulses of 500 mg each, total dose 3 g) followed by oral azathioprine, have the same clinical and immunological efficacy in the short and long term as monthly intravenous pulses with higher doses, and with a lower risk of infections [56, 57]. Although this Eurolupus regimen was evaluated in patients with relatively preserved kidney function and without representation of other ethnic groups, it is an effective, safe and frequently used treatment in patients with proliferative LN [3, 48, 57, 58].

Recent studies have shown superiority of the so-called triple immunosuppressive therapy over double immunosuppressive regimens (corticosteroids plus another immunosuppressive agent). The BLISS-LN study demonstrated that the combination of corticosteroids, belimumab, and MMF or cyclophosphamide was significantly superior to treatment with corticosteroids, MMF or cyclophosphamide and placebo in inducing CR [59]. This beneficial effect of belimumab was not observed in patients with significant proteinuria (>3 g/24-h), Black patients or those who had received cyclophosphamide. Subsequent analysis of the BLISS-LN study showed that belimumab also prevented LN relapses and slowed the loss of kidney function [60]. On the other hand, several studies have previously shown the efficacy of belimumab to control extrarenal manifestations, decrease lupus immune activity and reduce corticosteroid doses [61–64].

CNI have shown an important antiproteinuric and podocyte cytoprotective effect in different glomerular diseases, including LN [65]. Studies conducted in China showed that the combination of corticosteroids, MMF and tacrolimus was superior to corticosteroids plus cyclophosphamide in obtaining CR [66, 67]. Prospective studies with voclosporine have shown that this drug added to corticosteroids and MMF induces CR and PR in a significantly greater number of patients than therapy with corticosteroids plus MMF and placebo, without increasing adverse effects [46, 68]. Due to the acute and chronic nephrotoxicity that CNI (cyclosporine, tacrolimus) can induce, these agents are usually avoided (or administered with a close monitoring) in LN patients with reduced kidney function.

Although there are no studies designed in this regard, we suggest adding belimumab (especially if serological activity persists) or a CNI (especially if significant proteinuria persists) in those patients treated with dual immunosuppressive therapy without >25% reduction in proteinuria after 2–3 months of treatment.

Scarce information exists about the most effective treatments in LN patients with AKI at presentation. A post hoc analysis of the ALMS (Aspreva Lupus Management study) clinical trial compared monthly intravenous pulses of cyclophosphamide with MMF in 32 patients with eGFR <30 mL/min/1.73 m² at presentation [69]. The number of responses was similar (17% and 20%, respectively) and serious adverse events were more frequent in the group treated with cyclophosphamide.

4.3. Class V (membranous): initial treatment (Fig. 5)

Figure 5:

Initial treatment for class V (membranous). HCQ: initial dose 4–5 mg/kg/day (maximun 400 mg daily). If eGFR <30 mL/min/1.73 m², do not exceed 200 mg daily. MPAA: dose equivalent to 2 g of MMF. CNI: cyclosporine (initial dose 100–200 mg daily, blood target level 60–100 ng/mL), tacrolimus (initial dose 0.05–0.07 mg/kg/day, blood target level 4–7 ng/mL), voclosporine (23.7 mg twice daily for 1–2 years, no blood target level required). CV: cardiovascular.

Recommendations

• 4.3.1. In patients with proteinuria <1 g/24-h, we suggest treatment with HCQ, general nephroprotective measures and treatment of SLE according with extrarenal manifestations.

• 4.3.2. In patients with proteinuria between 1.0–3.5 g/24-h, we suggest double immunosuppressive treatment with corticosteroids and CNI (tacrolimus, voclosporine, cyclosporine). If after 3–4 months of treatment a reduction in proteinuria of at least 25% has not been obtained, and especially when persistent immunological activity coexists, we suggest adding MPAA.

• 4.3.3. In patients with proteinuria >3.5 g/24-h or complete nephrotic syndrome, we suggest triple immunosuppressive treatment from the onset, with corticosteroids, CNI (tacrolimus, voclosporine, cyclosporine) and MPAA.

Rationale

LN class V (membranous) frequently presents with nephrotic syndrome. There is a consensus regarding the indication of immunosuppressive treatment in cases with complete nephrotic syndrome. However, the management of patients with non-nephrotic proteinuria is more controversial. It seems reasonable to recommend immunosuppressive treatment in those cases who maintain proteinuria >1 g/24-h despite optimized RAS blockade [70].

Different immunosuppressive treatments have shown efficacy in class V LN, but there are no conclusive comparative studies. Corticosteroids alone are ineffective. MMPAs can induce remission but require more time to achieve therapeutic goals than CNI [71]. In patients with severe nephrotic syndrome, a more rapid and sustained reduction in proteinuria can be obtained with triple immunosuppressive therapy (corticosteroids, CNI and MPAA [47, 72].

4.4. Classes III/IV ± V and class V maintenance therapy (Fig. 6)

Figure 6:

Maintenance treatment for classes III/IV ± V and V. CV: cardiovascular; MRA: mineralocorticoid receptors antagonists.

Recommendations

• 4.4.1. We suggest a low-dose scheme of corticosteroids for maintenance therapy (for example, prednisone 2.5–5 mg/day from the sixth month of treatment), although such a scheme must be adapted to the patient's characteristics. Withdrawal of corticosteroids at month 18–24 is recommended in patients with clinical and serological remission.

• 4.4.2. In patients achieving clinical and serological remission, we recommend a progressive and slow reduction of MPAA dose after 18–24 months of treatment. Total duration of MPAA treatment should be at least 3–5 years.

• 4.4.3. We recommend azathioprine (1.5–2 mg/kg/day) for those patients who do not tolerate MPAA. Dose reduction and total duration of treatment is similar to MPAA.

• 4.4.4. Belimumab is suggested as maintenance therapy in patients who do not tolerate HCQ and/or corticosteroids, in relapsing patients, or in those with persistent extrarenal manifestations and/or persistent serological activity.

• 4.4.5. In patients with LN class V initially treated with corticosteroids and CNI, reduction of corticosteroids should follow the recommendation 4.4.1. For CNI, we suggest 12–18 months of treatment followed by a gradual tapering for 6–12 months (or 12–18 months in patients with PR and significant proteinuria).

• 4.4.6. We recommend treatment with RAS blockers, titrating doses to achieve and/or maintain BP and proteinuria targets. If proteinuria goals are not achieved, we suggest considering the addition of sodium-glucose cotransporter-2 inhibitors (SGLT2i) (Expert opinion).

Rationale

Regimens based on a reduced dose of corticosteroids have demonstrated similar efficacy to those based on higher doses, with a lower rate of side effects. When clinical and serological remission is achieved, corticosteroid withdrawal is recommended after 18–24 months of treatment, especially in patients on HCQ or another immunosuppressive agent. Intravenous cyclophosphamide in not recommended as maintenance therapy owing to the risk of severe adverse events related to excessive cumulative dose. MPAA or azathioprine have demonstrated a greater efficacy with a lower rate of side effects [73]. The ALMS study extension demonstrated that MMF was superior to azathioprine in terms of relapse prevention in patients who had achieved remission [74]. In the MAINTAIN trial, no differences in LN flares were observed between azathioprine-treated and MMF-treated patients, and the rate of adverse was similar except for hematological cytopenias, which were more frequent in the azathioprine group [58]. MPAA are preferred for maintenance therapy, although azathioprine is an effective alternative in those patients who cannot tolerate MPAA. A minimum duration of 3 years is recommended for both drugs, with a slowly progressive reduction before its complete withdrawal.

A post hoc analysis of BLISS-LN study showed that belimumab decreased the risk of relapses [59, 60]. Therefore, belimumab is a good alternative for maintenance therapy in patients with frequent relapses, intolerance or contraindications to HCQ or persistent extrarenal manifestations.

In LN class V the optimal maintenance regimen is not well established, given the scarcity of specific studies. A randomized study in patients with classes IV and V that compared cyclosporine with azathioprine showed a similar incidence of relapses [75]. In another randomized trial, cyclosporine was associated with a higher rate of remission at 12 months than cyclophosphamide or corticosteroids, but the incidence of relapses was significantly higher among cyclosporine-treated patients [76]. MPAA and tacrolimus showed a similar efficacy as maintenance treatment in another study [77]. We suggest a slow and progressive decrease in CNI doses in patients with class V LN who have achieved clinical remission after initial therapy with CNI and corticosteroids.

Non-immunosuppressive treatment (Fig. 6) is extremely important during the maintenance phase. A healthy lifestyle (regular physical exercise, avoidance of overweight and smoking) as well as the control and treatment of all cardiovascular risk factors is especially important in LN patients [78, 79]. Special attention should be paid to HCQ compliance to avoid relapses [80, 81], as HCQ prolongs remission during immunosuppressive therapy and improves renal prognosis [82, 83]. Treatment with RAS blockers is essential both for BP control and for achieving proteinuria goals [84]. A low-salt diet optimizes the effects of RAS blockers. There are other treatments that have demonstrated an antiproteinuric and renoprotective effect in different proteinuric kidney diseases, such as, SGLT2i, aldosterone receptor blockers, and anti-endothelin agents [85–88]. No conclusive studies have been published about their efficacy in LN, but preliminary data suggest an important antiproteinuric effect in LN patients, particularly in those with PR [89, 90].

4.5. Lupus podocytopathy

Recommendations

• 4.5.1. We recommend a similar management to minimal change disease, including corticosteroids as the first-line therapy and CNI as the second-line in corticosteroids-resistant patients. Rituximab, cyclophosphamide, MPAA or CNI are therapeutic alternatives in corticosteroid-dependent or frequently relapsing patients.

• 4.5.2. In corticosteroid-resistant lupus podocytopathy with renal function impairment not otherwise explained and/or presence of glomerular hematuria, a new kidney biopsy is recommended to rule out change to a different histological class.

Rationale

The pathogenesis of lupus podocytopathy is largely unknown and there are no controlled therapeutic studies. However, most of the reported cases presented with a complete nephrotic syndrome and showed a response to corticosteroids similar to that of minimal change disease [91–100]. Therefore, we recommend a therapeutic approach similar to this entity.

5. SPECIAL SITUATIONS

5.1. Refractory or resistant lupus nephritis (Fig. 7)

Figure 7:

Recommendations and therapeutic options in refractory LN.

Recommendations

• 5.1.1. We suggest defining refractory or resistant LN as the lack of response (or worsening of proteinuria and/or kidney function) after at least 3 months with any regimen of triple immunosuppressive therapy shown in Fig. 4.

• 5.1.2. In patients whose initial treatment was based on MPAA, we suggest adding rituximab or changing MPAA to cyclophosphamide (Eurolupus regimen).

• 5.1.3. In refractory LN after cyclophosphamide-based initial treatment, we suggest adding rituximab or extending cyclophosphamide pulses to complete 6 months of treatment.

• 5.1.4. In patients with no response to 5.1.2 and 5.1.3 recommendations, we suggest treatment with new anti-CD20 drugs (obinutuzumab) or anti-plasma cell drugs (daratumumab, bortezomib), or inclusion in clinical trials with new biological drugs under evaluation.

Rationale

Up to 30% of LN patients are refractory to initial treatment according to some studies [101]. It is important to consider, however, that there is no general agreement about the definition of refractory LN [102, 103]. Before establishing this diagnosis, it is essential to rule out non-compliance, sub-optimal doses of treatment, the existence of genetic factors influencing refractoriness and, in those cases with previous LN flares, chronic histological lesions by means of a new kidney biopsy [36, 37, 104, 105].

Although rituximab has not demonstrated superiority as add-on therapy to MPAA + corticosteroids [106], several observational studies including ethnically diverse patients have shown its effectiveness as rescue therapy in refractory LN [107–112].

Patients who do not respond to rituximab or extended cyclophosphamide therapy could benefit from newer anti-CD20 (obinutuzumab) or anti-plasma cell (bortezomib, daratumumab) therapies [113–115]. Cellular therapy with CAR-T cells could be an alternative for patients with LN refractory to several approaches [116, 117]. Ideally, LN patient refractory to several lines of treatment should be included in ongoing clinical trials evaluating new therapies.

5.2. Prevention and treatment of relapses (Fig. 8)

Figure 8:

Prevention and treatment of relapses.

Recommendations

• 5.2.1. We suggest treating LN relapses according to the clinical profile of the patient, according to the algorithms shown in Figs 4 and 5.

• 5.2.2. We recommend investigating and ensuring adequate therapeutic compliance. In cases with repeated relapses, we suggest considering a new kidney biopsy to evaluate histological activity, chronic and irreversible lesions, or changes in LN class.

• 5.2.3. After obtaining relapse remission, we suggest extending the duration of treatment with MPAA. In patients with repeated relapses, we suggest maintaining MPAA indefinitely at reduced doses.

• 5.2.4. We suggest adding belimumab to maintenance treatment in relapsing patients.

Rationale

There are no studies specifically designed to compare different treatments in LN relapses. We suggest treating relapses as in the first episode, according to the patient's clinical and analytical profile (Figs 4 and 5). When the clinical profile advises a new cyclophosphamide course, we suggest not exceeding the total cumulative dose of 10 g. It is crucial in relapsing patients to ensure strict compliance with HCQ and with other prescribed drugs. HCQ has been associated with a lower incidence of relapses [80–82], so this drug should be prescribed indefinitely, except in cases of contraindications or intolerance. It can be difficult to differentiate residual proteinuria due to healing of previous flares from that produced by active lesions in patients with repeated relapses. In these cases, a new biopsy can provide useful information [36, 39, 118]. MPAA treatment could be extended beyond the 3–5 years recommended for the initial flare in relapsing patients, and it may even be recommendable to maintain low doses of MPAA indefinitely in some cases with repeated relapses. In a sub-analysis of the BLISS-LN study, belimumab reduced the risk of LN relapse by 55% [60]. Treatment with this drug may therefore be useful in relapsing patients.

5.3. Pregnancy

Recommendations

• 5.3.1. Pregnancy in LN is associated with an increased risk of maternal and fetal complications and can be planned after a minimum of 6 months of LN remission. Pregnancy in LN patients should be controlled by multidisciplinary teams within high-risk obstetric units.

• 5.3.2. Teratogenic or potentially deleterious drugs should be avoided. Corticosteroids, azathioprine and CNI can be prescribed when immunosuppression is required during pregnancy.

• 5.3.3. HCQ should be maintained to reduce the risk of LN relapse. Low aspirin dose (100 mg/day) should be prescribed before the 12th week of pregnancy to reduce the risk of pregnancy complications.

Rationale

The risk of hypertension, preeclampsia, prematurity and preterm delivery is increased in LN patients. Pregnancy in any LN patient must be carefully planned and requires at least 6 months of remission to decrease the risk of relapse [119–121]. The first episode of LN can occur in pregnant SLE patients; a renal biopsy can be performed after the 20th week of pregnancy, provided there are no contraindications [122, 123]. The differential diagnosis between preeclampsia and LN flare is challenging; however, the evaluation of several clinical, laboratory, and imaging data is useful to establish the right diagnosis [124–126] (Table 4).

Table 4:

Differential diagnosis of LN flare and preeclampsia/HELLP in pregnant women.

Clinical characteristic	LN	Preeclampsia/HELLP syndrome
Impaired kidney function	Occasionally	Occasionally (HELLP)
Hypertension	Common	Yes
Proteinuria	Yes	Yes
Hematuria	Yes	No
Low serum Complement (C3/C4)	Yes	No
Positive anti-dsDNA	Yes	No
Extrarenal manifestations	Yes	No
Fever	Occasionally	No
Appearance	At any time during pregnancy	From week 20
sFlt-1	Normal	Increased
PIGF	Normal	Decreased
sFlt-1/PIGF ratio	Normal	Increased
Endoglin	Normal	Increased
Uterine arteries (Doppler US)	Normal	Abnormal

HELLP: Hemolysis, Elevated Liver enzymes, Low Platelets; PIGF: placental growth factor; SFlt-1 (or sVEGFR-1): soluble fms-like tyrosine kinase-1; US: ultrasonography.

The prescription of low doses of aspirin before the 12th week of pregnancy reduces the risk of preeclampsia and intrauterine growth retardation. HCQ should be maintained during pregnancy, unless contraindicated, in order to decrease the risk of SLE flares and LN relapses [16, 82, 127–129].

Cyclophosphamide and MPAA are contraindicated in pregnancy due to their teratogenic effects. Rituximab can cause B cell depletion and neutropenia in the fetus and newborn, increasing the risk of infection. When immunosuppression is required during pregnancy (to treat either a first LN outbreak or an LN relapse, or as maintenance therapy in patients at high risk for relapse), corticosteroids, azathioprine and CNI can be used (Table 5).

Table 5:

Use of immunosuppressive drugs during pregnancy in patients with LN.

Drug	Safety	Side effects
Azathioprine	Yes	Hematologic toxicity
CNI	Yes	Hypertension, gestational diabetes
HCQ	Yes	Its withdrawal can cause relapse
Corticosteroids	Yes	Gestational diabetes, premature rupture of membranes
Cyclophosphamide	No	Teratogenicity, myelosuppression
Leflunomide	No	Teratogenicity
Methotrexate	No	Teratogenicity
MPAA	No	Teratogenicity
Rituximab	No	Neonatal B-cell depletion

5.4. Thrombotic microangiopathy

Recommendation

• 5.4.1. In patients with LN and concomitant TMA, determination of antiphospholipid antibodies and ADAMST-13 activity should be performed. Although immunosuppressive treatment of LN alone can resolve LN-related TMA in some cases, the addition of anticoagulant therapy, plasma exchange or complement blockers may be indicated depending on the underlying abnormalities detected.

Rationale

LN patients, especially those with classes III/IV ± V, can present TMA lesions on renal biopsy. In some cases, these lesions are accompanied by the typical hematological manifestations of TMA (microangiopathic hemolytic anemia with schystocytes and low serum haptoglobin, thrombocytopenia). The etiology of LN-associated TMA is variable: deposition of immunocomplexes in the microvascular walls can cause endothelial damage severe enough to trigger TMA. In these cases, treatment will consist of the immunosuppressive approaches discussed for LN classes III/IV ± V. Some SLE patients develop antibodies against ADAMTS-13, a protein that prevents the aggregation of von Willebrand factor multimers. When ADAMTS-13 activity is less than 5%–10% of normal, the diagnosis of thrombotic thrombocitopenic purpura (TTP) can be established [130]. The PLASMIC score has demonstrated its usefulness in situations in which the determination of ADAMTS-13 is not possible [131]. In LN-related TTP, the addition of plasma exchange to immunosuppressive treatment can facilitate the recovery of renal function and improvement of renal lesions.

Up to 30%–40% of SLE patients present antiphospholipid antibodies and can develop antiphospholipid syndrome (APS), an important risk factor for vascular and renal damage in SLE [21, 132–137]. In addition to a high risk for arterial and/or venous thrombosis, APS can also cause the so-called APS-associated nephropathy, with renal lesions very similar to those of TMA, in addition to renal artery stenosis [137, 138]. While immunosuppression by itself does not induce favorable effects in APS, long-term anticoagulation is the treatment of choice in APS [139, 140]. Finally, some patients with LN-related severe TMA have been treated with complement blockers such as eculizumab, with variable results [141–149]. Although some cases may present mutations in complement regulatory genes [149–151], several studies have shown a transient complement activation in TMA of different etiologies, without identifiable genetic abnormalities [152]. These findings could support the temporary use of complement blockers in LN-related TMA unresponsive to immunosuppressive therapy [153, 154], although controlled prospective studies are needed.

6. HOLISTIC APPROACH TO LUPUS NEPHRITIS PATIENTS: PREVENTIVE ACTIONS, AND CONTROL OF COMORBIDITIES AND THERAPY TOXICITY

The objectives in the treatment of patients with LN go beyond renal remission. SLE is a chronic inflammatory and autoimmune disease that affects different organs and systems. Immune dysfunction and accelerated atherosclerosis are key conditions associated with SLE and LN [2, 4, 9, 10]. The development of chronic kidney disease, with progressive deterioration of renal function and persistent proteinuria, added to the short- and long-term adverse effects of anti-inflammatory and immunosuppressive drugs, has a significant impact on the development of acute and chronic complications. Among other things, LN patients have a greater susceptibility to infections and neoplasms, cardiovascular events, metabolic disturbances, involvement of the musculoskeletal system and bone, cutaneous and ocular damage, and infertility, with important deleterious consequences on quality of life and life expectancy [155]. Therefore, the prevention and appropriate treatment of all these complications within a holistic and multidisciplinary approach, is essential. Supplementary data, Tables S1 and S2 summarize the most common general recommendations to avoid these risks.

AUTHORS’ CONTRIBUTIONS

All authors participated in the regular consensus meetings, developed at least one specific section of the main document, including search and literature review, and approved and signed the final document, including tables and figures and the supplementary material before submission. M.P. carried out the design and general coordination of the study. M.P. and J.E.R.-R. designed and performed the main and supplementary figures and the correction, editing and layout of the main and supplementary tables. J.E.R.-R. and C.G.-C., in addition to their specific sections in the main document, performed the final general editing of the manuscript.

CONFLICT OF INTEREST STATEMENT

J.E.R.-R.: paid lectures and consultancy for GSK, Otsuka and Alexion. C.G.-C.: paid lectures and consultancy for AstraZeneca, Esteve, Novonortis, Boehringer Ingelheim, Astellas, Otsuka, Novartis, Mundifarma, Baxter and Vifor. A.I.Á.: nothing to disclose. M.Espino: paid lectures for Alexion. M.Espinosa: paid lectures and consultancy for Alexion. G.F.-J.: paid lectures for GSK and Otsuka. X.F.: paid consultancies for Otsuka and Novartis. M.G.: paid lectures and consultancy for GSK. M.M.: nothing to disclose. E.M.: paid lectures for GSK. L.F.Q.: paid lectures for GSK and Otsuka. M.P.: paid lectures and consultancy for GSK, Osuka, Novartis, Apellis, Alexion, Sanofi, Vifor and Travere.

DATA AVAILABILITY STATEMENT

The data underlying this article are available in the article and in its online supplementary material.