Introduction

Renal involvement represents a severe complication of Systemic Lupus Erythematosus (SLE) and biopsy-proven lupus nephritis (LN) occurs in 20–40% of patients [1,2,3]. The management of LN has witnessed advances over the past decades thus resulting in improved outcomes [4, 5]. Still, a considerable proportion (10–30%) of patients will develop chronic renal insufficiency and/or end-stage renal disease (ESRD) [3, 6,7,8,9]. In addition, LN and ESRD are associated with reduced health-related quality of life [3, 10], increased healthcare utilization and medical costs [11], comorbidities [12, 13], and most importantly, 6- to 26-fold increased mortality compared with the general population [8, 14].

In agreement with earlier studies, recent data have highlighted a number of factors that are associated with worse prognosis in LN, such as male gender, non-white race, class IV (with or without concomitant class V) disease (especially IV-global subtype [15]), arterial hypertension, low glomerular filtration rate (GFR), increased histological activity, and chronicity lesions (Table 1) [4, 16, 17]. Notably, tubulointerstitial lesions such as tubulitis, interstitial fibrosis, and tubular atrophy [17,18,19], as well as thrombotic microangiopathy [20], are increasingly recognized as risk factors for progression into ESRD, whereas the significance of serology is less clear [21]. Although appreciation of the prognostic impact of the aforementioned factors offers the possibility for patient-tailored medical care, still, the management of LN is rather generic. In this review, we summarize recent evidence pertaining to the diagnosis, therapy, and monitoring of LN, including special issues such as the treatment of pediatric LN, during pregnancy, and ESRD. Notwithstanding the lack of comparative data, we provide some guidance with regard to appropriate treatment regimen based on the clinical scenario. (Table 2)

Table 1 Major risk factors associated with adverse renal outcomes in patients with lupus nephritis
Full size table
Table 2 Therapeutic regimens in lupus nephritis
Full size table

Early Diagnosis of Lupus Nephritis

Delay in diagnosis of LN and initiation of immunosuppressive treatment have been linked to lower renal response and increased ESRD rates [16, 22, 23]. Accordingly, vigilance is required for the prompt identification of signs and symptoms suggestive of kidney disease. To facilitate early classification/diagnosis, the 2012 Systemic Lupus Collaborating Clinics (SLICC) [24, 25] and the 2019 European League Against Rheumatism/American College of Rheumatology (EULAR/ACR) [26•] classification criteria enable the classification of SLE based merely on histological evidence of LN coupled with positive ANA or lupus autoantibodies. Notably, a study evaluating cases diagnosed with LN from 1970 to 2016 found decreasing trends of renal insufficiency and histological chronicity, and increasing rates of isolated urinary abnormalities at the time of LN presentation [4], all suggestive of tendency for early than delayed diagnosis. In the same study, an increase in patient age and longer time elapsing between SLE onset and LN occurrence was observed towards more recent times, which might be due to improved disease management [27].

To this end, risk factors for incident LN include younger age, male gender, non-white race, high anti-dsDNA titer, and presence of anti-Sm antibody [28]. Anti-C1q antibodies have also been shown to predict active proliferative LN with high specificity; however, they lack standardization and are not universally assayed [29, 30]. Importantly, patients with LN are characterized by higher genetic burden [31, 32] and neutrophil gene signature in the peripheral blood [33, 34], which could help to define high-risk groups among SLE patients [35] and possibly implement preventative strategies.

Treatment of Proliferative Lupus Nephritis: an Expanding Range of Therapeutic Regimens

New Data on the Efficacy and Safety of Conventional Immunosuppressive Regimens

At present, first-line treatment of active proliferative (class III or IV, with or without concomitant class V) nephritis includes the combination of high-dose glucocorticoids (typically, pulses of intravenous methylprednisolone followed by 0.5 mg/kg/day oral prednisone) with either low-dose cyclophosphamide (500 mg intravenous bolus administered biweekly for 3 months) or oral mycophenolate (2 to 3 g/day, for 6 months) [36,37,38]. Following this initial period (“induction phase”), less intensive therapy with gradually tapered dose of glucocorticoids and either mycophenolate (1 to 2 g/day) or azathioprine (2 mg/kg/day) is administered for several years to consolidate and maintain the response (“maintenance phase”). The aforementioned recommendation is supported by randomized controlled trials (RCTs) demonstrating comparable efficacy and more favorable toxicity profile of low-dose cyclophosphamide and mycophenolate compared with the high-dose cyclophosphamide regimen [39, 40].

According to a recent Cochrane Collaboration Group systematic review and meta-analysis of RCTs, mycophenolate is as effective as intravenous cyclophosphamide in inducing remission in LN (8 studies, 828 participants; relative risk [RR] 1.17, 95% confidence interval [CI] 0.97 to 1.42) [41•]. Additional observational studies have suggested efficacy of mycophenolate across different ethnic groups [42, 43], and in terms of certain long-term outcomes such as chronic kidney disease and renal flares [44, 45]. Notwithstanding, a post hoc analysis of a controlled trial showed that induction with intravenous cyclophosphamide versus mycophenolate was associated with a lower likelihood of treatment failure after 3 years (odds ratio 0.50; 95% CI 0.2 to 1.0) [21]. To this end, more evidence is still needed with regard to the long-term efficacy of mycophenolate especially against “hard” renal outcomes.

Low-dose intravenous cyclophosphamide, followed by maintenance with azathioprine or mycophenolate, has yielded long-term (10 years) effectiveness against ESRD, yet the majority of patients will require chronic treatment with immunosuppressive agents and low-dose glucocorticoids [46, 47]. More recently, the regimen has been successfully used in non-Caucasians including Asians [42], Africans [48], and African-Americans [49], although this has not always been confirmed [50,51,52]. According to the present treatment paradigm, high-dose cyclophosphamide should be reserved for selected severe cases such as with nephritic urine sediment and impaired renal function, or histological crescents or necrosis affecting > 25% of glomeruli [36].

Calcineurin Inhibitors: Comeback in Lupus Nephritis

Several studies have recently explored the efficacy of calcineurin inhibitors (CNI), namely tacrolimus, cyclosporin, and voclosporin, either as monotherapy or in combination with mycophenolate. A meta-analysis of five RCTs showed that tacrolimus was more effective than intravenous cyclophosphamide at inducing complete renal remission (RR 1.59; 95% CI 1.16 to 2.19), although the difference was non-significant when mycophenolate was used as comparator (RR 0.97; 95% CI 0.64 to 1.46) [53]. Similarly, tacrolimus was non-inferior to mycophenolate in a trial involving 150 Asian patients with extended follow up. At 5 years, the cumulative incidence of a composite outcome (decline of creatinine clearance by at least 30%, development of chronic kidney disease or death) was comparable between the two groups [54•].

The combination of mycophenolate with CNI and glucocorticoids (multitarget regimen) has recently drawn attention in LN. Notably, a large RCT comparing mycophenolate (1 g/day) plus tacrolimus (4 mg/day) against intravenous cyclophosphamide (0.75 mg/m2, 6 monthly boluses)—both given on a background of glucocorticoids—in 368 Chinese patients with LN reported nearly 2-fold higher remission rates with the former regimen at 24 weeks [55]. Notwithstanding the aforementioned data are limited by the inclusion of almost exclusively Asian patients, results from the multi-ethnic phase 2 [56•] and phase 3, https://ir.auriniapharma.com/press-releases/detail/164/aurinia-announces-positive-aurora-phase-3-trial-results. https://ir.auriniapharma.com/press-releases/detail/164/aurinia-announces-positive-aurora-phase-3-trial-results. controlled trials of voclosporin/mycophenolate versus mycophenolate alone have confirmed the superior efficacy (OR 2.65; 95% CI 1.64 to 4.27) with acceptable safety of the multitarget regimen, at least at the short-term. To this end, it remains unclear as to whether the multitarget regimen should be viewed as a universal first-line treatment of LN, or whether it should be considered for selected patients such as those with severe nephrotic syndrome or podocytopathy at presentation, or with inadequate reduction in proteinuria after initial treatment with mycophenolate [38••]. Skepticism against their use pertains to the fact that CNIs may exert predominantly antiproteinuric—rather than immunomodulatory—effects, as well as their possible nephrotoxicity with long-term use [57].

Biologics in Lupus Nephritis

Despite its failed trial, B cell depletion with rituximab (anti-CD20 monoclonal antibody) has shown promising results in numerous non-randomized LN studies, especially when used as salvage therapy [50]. Renal response rates approximate 70% (40% complete, 30% partial response) and correlate positively with complete (OR 5.8) and long-lasting (> 71 days; OR 4.1) B cell depletion, and negatively with longer time to achieving depletion (OR 0.89) [58], thus suggesting that monitoring peripheral B cell counts might be a useful biomarker [59]. Blissfully, obinutuzumab, a humanized anti-CD20 monoclonal antibody, is currently evaluated in a phase 3 trial (NCT04221477) based on encouraging results when used in combination with mycophenolate, https://www.roche.com/media/releases/med-cor-2019-11-11b.htm. https://www.roche.com/media/releases/med-cor-2019-11-11b.htm. Moreover, ofatumumab, another human monoclonal antibody against CD20, has been successfully used instead of rituximab in cases of intolerance/immunogenicity to the latter [60].

Belimumab, a monoclonal antibody targeting BAFF (B cell activating factor) currently approved for the treatment of active SLE, has also yielded evidence for efficacy in renal disease [61]. Interestingly, BAFF has been implicated in LN by inducing renal tertiary lymphoid structures and regulating the position of intra-glomerular T lymphocytes [62]. Indeed, the Efficacy and Safety of Belimumab in Patients With Active Lupus Nephritis (BLISS-LN) trial was recently announced to meet its primary endpoint demonstrating increased renal response rates over 2 years in patients who received belimumab compared with placebo (both combined with standard therapy) (43% versus 32%, p = 0.031), https://www.gsk.com/en-gb/media/press-releases/gsk-announces-positive-headline-results-in-phase-3-study-of-benlysta-in-patients-with-lupus-nephritis/. https://www.gsk.com/en-gb/media/press-releases/gsk-announces-positive-headline-results-in-phase-3-study-of-benlysta-in-patients-with-lupus-nephritis/. A detailed look at the results of this trial, once published, will be needed in order to determine the position of belimumab in the therapeutic armamentarium of LN. To this end, belimumab could be considered add-on therapy to mycophenolate in cases of inadequate renal response, inability to taper off prednisone to less than 7.5 mg/day, and/or presence of extra-renal lupus activity [63]. Finally, anifrolumab, a human monoclonal antibody to type I interferon receptor subunit 1, has demonstrated efficacy in SLE [64•] and its efficacy in LN is currently being assessed (NCT02547922).

Maintenance Treatment in Lupus Nephritis: For how Long?

The two landmark RCTs of maintenance immunosuppressive treatment in proliferative LN, the MAINTAIN trial [46] and the Aspreva Lupus Management Study (ALMS) [65], despite differences in their design and the induction regimens, both provided evidence for the long-term efficacy of mycophenolate and azathioprine. Notably, in a meta-analysis of 4 RCTs including 452 LN patients, treatment with azathioprine was associated with increased risk of relapse as compared with mycophenolate (RR 1.75; 95% CI 1.20 to 2.55) although the two regimens did not differ in terms of ESRD or mortality [41•]. Based on these results, mycophenolate may be preferred over azathioprine in severe LN cases or when the former drug was also used for induction [66]. Limited data support the use of CNIs as maintenance treatment in proliferative LN [67], still they can be a useful option in cases of pregnancy or intolerance to other agents. Finally, a controlled study in Asian patients has demonstrated efficacy of a multitarget-based maintenance regimen (mycophenolate 0.5–0.75 g/day, tacrolimus 2–3 mg/day) however, in combination with unacceptably high dose of prednisone (10 mg/day) over a period of 18 months [68].

To date, there is no study designed to evaluate the feasibility of discontinuing immunosuppressive therapy in LN. In a retrospective analysis of 73 Caucasian LN patients who achieved remission, successful (i.e., free of renal relapse) withdrawal of immunosuppressives was accomplished in 71.2% of cases, particularly those with longer treatment duration (average 98.1 versus 31.0 months in cases who relapses) and longer period on remission (52.8 versus 12.0 months) [69]. Accordingly, treatment should be maintained for at least 3 to 5 years after remission has been achieved, although it may be extended in patients with adverse prognostic factors such as African ancestry, renal flares, and chronic kidney disease. Interestingly, among patients with clinically quiescent LN who underwent per-protocol repeat kidney biopsy, the risk for relapse during the ensuing 24 months was found to be independently associated with the histological activity score [70•], suggesting that kidney biopsy might be used to guide duration of maintenance treatment in LN [71].

Treatment of Membranous Lupus Nephritis

Patients with pure class V (membranous) LN and nephrotic-range proteinuria may benefit from immunosuppressive treatment such as mycophenolate or high-dose intravenous cyclophosphamide, in combination with glucocorticoids [36]. Other options have also been used albeit supported by less evidence. Specifically, low-dose cyclophosphamide was evaluated in a small RCT yielding comparable response rates with mycophenolate (71.4% versus 66.7%, respectively) [42•]. A network meta-analysis of RCTs and cohort studies has also suggested efficacy of CNIs in membranous LN [72], which however, are considered second-line agents for induction or maintenance therapy due to their safety profile and the need for prolonged administration to prevent relapses [73]. Similar to proliferative nephritis, the multitarget regimen consisting of mycophenolate, tacrolimus, and glucocorticoids demonstrated superior short-term efficacy than intravenous cyclophosphamide and glucocorticoids in a subgroup of 69 Chinese patients with class V LN (response rates 33.1% versus 7.8%, respectively) [55]. Finally, a small retrospective study reported high renal response rates (13 out of 15 patients) of rituximab used in combination with glucocorticoids [74], although this option is typically reserved for refractory/relapsing cases.

Monitoring Lupus Nephritis

Treat-to-Target

The present paradigm in the management of rheumatic diseases has introduced the concept of “treating-to-target” towards improvement of long-term disease and patient prognosis [75]. In clinical practice, assessment of the renal response to treatment can be challenging as it encompasses a variety of relevant parameters such as GFR, proteinuria, urinalysis, serological markers, blood pressure, body weight, hematocrit, and serum albumin [36]. Among these, changes in proteinuria have been identified as the best individual predictor of long-term renal outcomes [76]. Accordingly, a complete renal response is typically defined as a decrease in proteinuria to very low levels (less than 500 mg/24 h) coupled with normalization or stabilization (within 10% of normal GFR if previously abnormal) of renal function, with or without clearance of abnormal urinary sediment. This state has been consistently associated with favorable long-term renal prognosis in the context of RCTs and long-term observational studies [77,78,79,80,81], and therefore, is considered the ultimate therapeutic goal in LN [36, 38, 66]. Attainment of more stringent proteinuria cut-offs (< 150 mg/24 h) is associated with even lower risk of 5-year renal flare rate and development of chronic kidney disease, but it may be less frequently encountered [82].

In addition, post hoc analysis of the low-dose intravenous cyclophosphamide RCTs has indicated that proteinuria below 700–800 mg/24 h at 12 months following treatment initiation predicts favorable long-term renal outcome with sensitivity 71–81% and specificity 75–78% [76, 83]. Importantly, the prognostic value of proteinuria < 800 mg/24 h at 12 months was confirmed (sensitivity 90%, specificity 78%, negative predictive value 94%) in a long-term observational study of 94 patients with biopsy-proven LN [84]. Collectively, these data suggest that treatment in LN should aim at a proteinuria level below 700–800 mg/24 h at 12 months and below 500 mg/24 h at 24 months [38, 77, 78]. Interim goals at 3 to 6 months should include a consistent decrease in proteinuria by at least 25–50% [85]. Although normalization of anti-dsDNA soon after induction treatment has also been linked to favorable renal response [21], therapeutic adjustments based on serological markers alone are currently not recommended.

Defining Refractory and Relapsing Disease and the Role of Repeat Kidney Biopsy

Although early resolution of proteinuria is associated with excellent prognosis, data from RCTs suggest that only 25–40% of LN patients will achieve complete renal response within the first 6 months since treatment initiation [86, 87]. Indeed, it has long been appreciated that therapy-induced clearance of kidney immune deposits and inflammation is a chronic process and that heavier baseline proteinuria takes longer to resolve [88]. In accordance, the proportion of treated LN patients who meet the renal response criteria may increase over time [89]: therefore, monitoring the kinetics of proteinuria may be more appropriate indicator of the effectiveness of immunosuppressive treatment. Still, a considerable proportion (20–30%) of patients will response poorly or inadequately to first-line treatment, which is associated with increased risk for ESRD [78, 90].

Exacerbation of LN in a patient who previously responded to treatment represents yet another adverse outcome, and a contemporary observational study found that renal relapse-free survival rate was 69% and 57% at 5 and 10 years, respectively, post induction treatment [45]. Renal flares can be classified as proteinuric or nephritic (i.e., accompanied by decrease in GFR by ≥ 10% and re-activation of urine sediment) [91], the latter carrying the highest risk for progression into ESRD [92, 93]. Serologic reactivation may predate renal flares [94]; however, pre-emptive treatment is currently not recommended.

In the context of refractory or relapsing LN, (repeat) kidney biopsy may be particularly useful to evaluate the histological class, activity, and chronicity lesions, exclude other pathologies, and provide therapeutic guidance [95]. Notably, several studies have highlighted a clinical–histological discordance in patients with clinically quiescent LN [96, 97], and that persistence of histologically active disease in spite of reduced proteinuria is associated with increased likelihood for renal relapse upon treatment withdrawal [70, 98]. Together, these findings suggest the possibility of using per-protocol kidney biopsy to determine the optimal intensity and duration of maintenance immunosuppressive therapy in LN [71], although this will require further validation in prospective studies.

Treatment of Refractory and Relapsing Lupus Nephritis

For patients with refractory LN and following assessment of medication adherence, the usual approach includes switching to another first-line treatment, for instance, from mycophenolate to cyclophosphamide (including the high-dose regimen) and vice versa, although there is limited supporting evidence [99, 100]. CNIs, particularly as part of multitarget regimens, represent another option [101, 102], especially in the absence of significant renal fibrosis. B cell depleting agents have been extensively used off-label in LN refractory to one or more conventional immunosuppressive agents. In a meta-analysis of three studies including 57 LN patients with refractory disease, treatment with rituximab induced renal response in 70% (95% CI 55% to 81%), and this was accompanied by a steroid-sparing effect (mean reduction: 12.5 mg/day; 95% CI 6.4 to 18.6) [103]. Finally, addition of belimumab to background mycophenolate could also be considered although evidence regarding the effectiveness of this combination in refectory cases is still limited [61].

The management of a LN relapse will depend on its severity (increase in proteinuria, reduction in GFR, histological findings in case of repeat biopsy) and the possible co-existence of extra-renal lupus activity. There is paucity of evidence but in general, therapeutic measures include initiation or increase in glucocorticoids (including pulses of intravenous methylprednisolone), re-induction with any of the available agents, or using the same regimens as described above for refractory LN [100, 104,105,106]. To this end, considering the lack of randomized evidence and head-to-head comparisons, treatment of relapsing or refractory nephritis cannot follow the “one size fits all” rule. For patients who are on mycophenolate and manifest inadequate renal response, a CNI or belimumab can be added, the latter being preferred if there is history of flares or generalized SLE activity. In the scenario of a severe relapse, induction with cyclophosphamide or rituximab may be considered.

Lupus Nephritis in Special Patient Groups

Pediatric Lupus Nephritis

Childhood-onset LN tends to have more aggressive presentation, often leading to earlier and higher damage accrual. Evidence on its management is mainly extrapolated from studies in adults. A kidney biopsy offers a definitive diagnosis and guides therapeutic decisions. Similar to adult LN, the therapeutic goal is complete or partial renal response attained no more than a year since treatment initiation [107]. Both cyclophosphamide and mycophenolate, combined with glucocorticoids, are indicated as induction therapy. In the largest published cohort including 51 children with proliferative LN followed over a 3-year period, there was no difference between the two treatments in terms of disease activity, urine albumin/creatinine ratio, and serum creatinine [108]. Concerns over cyclophosphamide-related gonadal toxicity and the chronic implications from long-term use of corticosteroids have increased the efforts for alternative regimens with lower cumulative doses; however, there are no long-term efficacy data in children with LN. To this end, limited observational data support the effectiveness of CNIs, multitarget treatment, and biologic agents both in refractory or new-onset disease [109, 110].

Lupus Nephritis During Pregnancy

Lupus affects primarily women of child-bearing age, and therefore, pregnancy is not uncommon throughout the course of the disease. In a large prospective study of 383 pregnant SLE patients, low C4 was associated with renal flares or new-onset LN (OR 5.59; 95% CI 1.64 to 19.13) but not low C3 or positive anti-dsDNA alone [111•]. Patients with active LN are at increased risk for adverse fetal and maternal outcomes [112]. In a recent meta-analysis of four studies including 285 patients, the pooled RR for preterm birth in active LN was 1.78 (95% CI 1.17 to 2.70) compared with counterparts with quiescent disease [113]. Similarly, prospective studies suggest that maternal outcomes such as pre-eclampsia or disease flares are more frequent among women with LN [114, 115]. In line with these findings, the EULAR recommends that the pre-pregnancy counseling and risk assessment should consider both SLE and LN activity [116•]. Inactive/stable disease may permit pre-conception treatment modifications to avoid teratogenic agents such as cyclophosphamide, mycophenolate, methotrexate, and angiotensin-converting enzyme inhibitors. Other agents such as azathioprine, tacrolimus, and cyclosporin are considered safe/acceptable. Finally, there is a consensus regarding the use of hydroxychloroquine in all pregnant women with SLE based on its safety profile and beneficial effects especially on maternal outcomes [117, 118].

Lupus Nephritis During End-Stage Renal Disease

Despite advances, the long-term renal survival in LN has essentially remained unchanged since the 1990s. Interestingly, as LN progresses towards ESRD, disease activity tends to diminish; thus, extra-renal flares are not common post-dialysis [119]. All renal replacement therapies (hemodialysis, peritoneal dialysis, kidney transplantation) can be considered for SLE patients with transplantation demonstrating a significant survival benefit, and it can be safely performed in quiescent LN patients. Recently, a nationwide cohort study with data from the United States Renal Data System including 20,974 LN-ESRD patients found that transplantation was associated with a 70% reduction in all-cause mortality (hazard ratio 0.30; 95% CI 0.27 to 0.33) [120]. Finally, regarding renal graft survival, most studies have found comparable rates between LN and control recipients [121, 122].

The Cardiovascular Burden of Lupus Nephritis

Patients with SLE have an increased risk of cardiovascular events owing both to traditional and disease-specific risk factors [123]. LN is particularly associated with an additional (2.8- to 8.5-fold) cardiovascular risk compared with non-renal lupus [124,125,126]. Thus, in a Danish population-based cohort study with 1644 SLE patients, LN patients had significantly increased cardiovascular mortality compared with other SLE patients (HR 8.5, 95% CI 2.2 to 33) [126]. In another retrospective study, Sun et al. [127] performed a class-specific analysis and found a trend for excessive cardiovascular risk associated with proliferative LN. The heightened cardiovascular burden in LN may in part be attributed to the presence of chronic kidney disease and increased prevalence of traditional atherosclerotic risk factors (hypertension, dyslipidemia) and exposure to glucocorticoids. Accordingly, a multifaceted approach is mandated for primary prevention [123], although specific recommendations are still missing.

Conclusions

Despite its lower frequency compared with other manifestations, LN is a serious complication of SLE associated with the need for prolonged intake of glucocorticoids and immunosuppressive or cytotoxic drugs, and increased risk for several comorbidities. A low threshold is required for the early detection of kidney involvement in patients with established SLE, as much as for the diagnosis of SLE in patients who present with abnormal renal function tests; to facilitate the latter, the recent classification criteria for SLE have included high-sensitivity algorithms for classifying LN. In both scenarios, a diagnostic kidney biopsy should be performed promptly, since delay in initiation of immunosuppressive treatment may incur renal damage. First-line treatment with either low-dose intravenous cyclophosphamide or mycophenolate, followed by long-term maintenance immunosuppression, is effective in the majority of cases; however, a considerable proportion of patients will demonstrate delayed or inadequate improvement, accrual of kidney fibrosis, progression into chronic kidney disease, or even ESRD. To expedite or augment the renal response, novel multitarget approaches based on the combination of mycophenolate with either CNI or belimumab have yielded promising results, although it remains unclear whether their added benefit justifies their universal use as induction regimens in LN. Nonetheless, these combinatory regimens together with B cell depleting agents certainly represent useful alternatives in refractory or relapsing disease. Notably, the identification of specific thresholds of proteinuria associated with favorable long-term renal prognosis paves the way for implementing a treat-to-target strategy in LN. Depending on the depth and duration of renal response, gradual tapering of treatment (glucocorticoids first) can be attempted in some cases, and recent evidence suggests that patient selection may be aided by the findings of repeat kidney biopsy.

Notwithstanding the aforementioned advances, several unmet needs exist such as the need for individualized therapeutic decisions based on the clinical presentation, histological findings, and overall SLE status. The evidence base for the management of recalcitrant or flaring disease is limited, and thus, controlled studies are eagerly awaited. In this context, the indications for performing repeat kidney biopsy or whether per-protocol biopsies should be performed in patients with LN remain to be determined. For the future, we remain optimistic that ongoing efforts to decipher the cellular and molecular heterogeneity of lupus kidney inflammation and its progression [128,129,130] will eventually lead to the discovery of accurate biomarkers and targetable pathways pertaining to the underlying pathophysiology.