Novel Treatments Paradigms: Membranous Nephropathy

Abstract

Primary membranous nephropathy (MN) is a kidney-specific autoimmune glomerular disease and the leading cause of nephrotic syndrome (NS) in White adults, usually caused by antiphospholipase A2 receptor (PLA2R) antibodies, although several new target antigens have been recently identified. It is characterized by the diffuse thickening of the glomerular basement membrane secondary to immune complex deposition. In patients with persistent NS without response to maximizing conservative therapy including the use of renin-angiotensin system (RAS) blockers, the use of immunosuppressive agents is warranted. However, the optimal immunosuppressive treatment has not yet been established. Classical immunosuppressants, such as cyclophosphamide plus steroids, are effective but may cause clinically relevant adverse effects, limiting their use. Rituximab offers efficacy with a better safety profile whereas calcineurin inhibitors (CNIs) are marred by high relapse rates and nephrotoxicity. Nevertheless, up to 30% of patients fail to respond to standard therapy. Novel and specific therapies targeting B cells and plasma cells have shown encouraging preliminary results, in terms of clinical efficacy and safety profile, especially in patients with poor tolerance or refractory to conventional treatments. In this brief review, we discuss the benefits and limitations of the current therapeutic approach to MN and describe emerging novel therapies that target its pathogenesis.

MN is the main cause of NS in White adults, the second cause in African American and Hispanic individuals, and almost twice more frequent in men.¹ Primary MN is a kidney-specific autoimmune glomerular disease caused by circulating podocyte-targeted autoantibodies, mainly anti-PLA2R (70%−75%).¹ Recently, novel autoantibodies and podocyte antigens have been described using laser-capture microdissection/mass spectrometry (Table 1).² The formation and deposits of immune complexes containing immunoglobulins and complement induce podocyte damage and alter the glomerular basement membrane, resulting in the development of proteinuria, that frequently progress to full blown NS, and if persistent, progression to kidney failure^1,2 (Figure 1).

Table 1.

Antigens and autoantibodies associated with membranous nephropathy

Antigen	Biological Function	Expression site	Serum antibodies	IgG deposition site	IgG subclass	Clinical phenotype	Frequency	Clinical utility
PLA2R	Receptor for PLA2 type M	Podocyte	Anti-PLA2R	Subepithelial	IgG4	“primary” MN	70%–85%	Diagnosis, prognosis, monitoring
THSD7A	Angiogenesis	Podocyte	Anti-THSD7A	Subepithelial	IgG4	“primary” MN, cancer	2%–5%	Diagnosis, prognosis, monitoring
HTRA1	Serine protease	Podocyte	Anti-HTRA1	Subepithelial	IgG4	“primary” MN	1%	Diagnosis
NTNG1	NS development	Podocyte	Anti-NTNG1	Subepithelial	IgG4	“primary” MN	Uncommon	Diagnosis
CNTN1	NS development	GBM	Anti-CNTN1	Subepithelial	IgG4	“primary” MN	Uncommon	Diagnosis
SEMA3B	NS development	GBM	Anti-SEMA3B	Subepithelial	IgG1, IgG3	children, females	<1%	Diagnosis, prognosis?
TGFBR3	TGFβ coreceptor	GBM, glomerular capillary	No	Subepithelial Subendothelial Mesangial	Variable	Lupus MN, females	5%–10% SLE MN	Diagnosis
FAT1	Gap junction formation	GBM, TBM	NA	Subepithelial	IgG4, IgG2	HSCT-associated MN	Uncommon	Diagnosis
EXT1/EXT2	HS biosynthesis	GBM	No	Subepithelial Subendothelial Mesangial	IgG1, IgG2	Lupus MN	30% SLE	Diagnosis, prognosis
NELL1	Cell growth, differentiation	Osteoblasts, tubular cells	Anti-NELL1	Subepithelial	IgG1, IgG3	“primary” MN, cancer	3%–4%	Diagnosis
PCDH7	Cellular adhesion	GBM	Anti-PCDH7	Subepithelial	IgG1, IgG4	“primary” MN	1%–3%	Diagnosis
NCAM1	NS development	Glomerular capillary, TBM (rare)	Anti-NCAM1	Subepithelial	IgG4	“primary” and lupus MN	2% (MN), 6.6% (SLE)	Diagnosis

CNTN1, contactin 1; EXT1/EXT2, exostosin 1/2; FAT1, Protocadherin FAT1; GBMg glomerular basement membrane; HS, heparan sulfate; HSCT, hematopoietic stem cell transplant; HTRA1, High Temperature Requirement A serine peptidase 1; Ig, immunoglobulin; MN, membranous nephropathy; NA, not available; NCAM1, neural cell adhesion molecule 1; NELL1, NEL-like protein 1; NS, nervous system; NTNG1, netrin G1; PCDH7, protcadherin-7; PLA2R, phospholipase 2 receptor; SEMA3B, semaphorin 3B; SLE, systemic lupus erythematosus; TBM, tubular basement membrane; TGFBR3, TGFβ receptor type 3; THSD7A, thrombospondin type-1 domain-containing protein 7A.

Table adapted and partially modified from reference 2 with permission.

Figure 1.

Classical immunosuppressors and targeted novel therapies for the treatment of membranous nephropathy. MN is characterized by the presence of subepithelial immune deposits in the glomerular capillary wall. Although these deposits may contain exogeneous planted antigens (e.g., cationic bovine serum albumin), more often, they are composed of autoantibodies directed against podocyte antigens (Table 1) and complement. Complement targeting prevents preclinical podocyte injury and MN and there is systemic evidence for complement activation in human MN. T cells cooperate in the immune response with CD20⁺ B cells. Affinity-maturated, class-switched B cells may circulate as CD20⁺ memory B cells or terminally differentiate into CD20^-/CD38⁺ plasma cells, the main source of antibodies and autoantibodies. Cyclophosphamide principally affects the activation and proliferation of T cells and early B cells. CNIs inhibit T-cell differentiation and proliferation, but they can also reduce the glomerular expression of PLA2R (tacrolimus) and decrease proteinuria through hemodynamic effects and stabilization of podocyte cytoarchitecture. Rituximab and other anti-CD20 drugs induce death of CD20+ B cells at early and intermediate stages, reducing the progression to late B cells and plasma cells and reducing the production of autoantibodies. Belimumab targets BAFF, leading to B-cell apoptosis, blockade of progression to plasmacytes, and reduced autoantibody production. Anti-CD38 drugs and proteasome inhibitors target plasma cells and reduce the production of autoantibodies, especially in relapsing and refractory cases. Eculizumab and avacopan inhibit the final stage of the complement system, but there is no experience with these drugs in primary MN, except for cases with concurrent thrombotic microangiopathy for eculizumab. Iptacopan (LNP023) inhibits the effect of factor B on the C3 convertase, limiting the continuous complement activation. An ongoing RCT phase 2 is comparing iptacopan and rituximab for MN. BCX9930 inhibits Factor D and an ongoing RCT is enrolling MN patients, as for narsoplimab (OMS721) which inhibits MASP-2, i.e., the lectin pathway for complement activation. PLA2R

In patients with persistent proteinuria >4 g/24 h despite RAS blockers, several studies have consistently demonstrated that rituximab, cyclophosphamide, and CNIs ± glucocorticoids, can induce proteinuria remission. However, up to 30% of patients have a poor response, which is associated with a high risk of progression to kidney failure.³ Some patients develop drug intolerance or serious adverse effects (e.g., major infections, myelosuppression, neoplasms, nephrotoxicity, metabolic disorders). In addition, relapses are common. This has led to the search for more specific and better tolerated immunomodulatory drugs, aiming at improving long-term clinical outcomes. In this brief review, we discuss the current therapeutic approach and novel immunomodulatory agents and their potential role in the treatment of MN. We also highlight the unmet needs that should be addressed in future research.

Current Therapies Available for MN

Patients with MN should be treated with supportive nephroprotective and cardiovascular disease preventive strategies, and immune suppression should be considered for patients at risk for progressive kidney injury (Figure 2a).³ Patients are at low risk of progression if they have normal or stable estimated glomerular filtration rate (eGFR) over the preceding 6 months with proteinuria <4 g/24h with either negative PLA2R or low PLA2R antibody levels (<50 RU/ml) that are decreasing by ≥25%. These patients do not require immunosuppressive therapy and should be treated with supportive therapy (see next section). Renal function, serum albumin, and 24-hour proteinuria should be monitored. Patients at moderate risk of progression are those who have normal or stable eGFR over the preceding 6 months but have proteinuria between 4 g/24h and 8 g/24h (in the absence of life-threatening NS, e.g., complications such as a thromboembolic event), with moderately elevated PLA2R levels (50−150 RU/ml) that are stable over a span of 6 months. These patients may need to be treated depending on how they progress over time. For example, if the proteinuria increases, eGFR declines, or PLA2R titer rises, then the patients should be treated with immunosuppression (see next section). Patients who are at high risk of progression should be treated with immunosuppression without any additional observation period. These patients include those who have a decline in their eGFR ≥25% related to the MN, have >8 g/24h of proteinuria, have life-threatening NS, or have PLA2R antibodies that are elevated (≥150 RU/ml) and are not declining >25% within a 1-month to 2-month interval (Figure 2a).

Figure 2.

Current approach to membranous nephropathy therapy and role of anti-PLA2R in monitoring disease (2a). Therapeutic algorithm designed, adapted, and modified from figures 30 and 31 of reference 3. Control of anti-PLA2R serum levels should be performed at intervals of 3 to 6 months from baseline determination. However, shorter, or longer intervals could be considered depending on the patient's clinical evolution. ^†Selectivity index is calculated as clearance of IgG/clearance of albumin. ^††Cardionephroprotection involves strict control of blood pressure and proteinuria using the maximum tolerable dose of RAS inhibitors. Weight control, dietary salt restriction and, in general, healthy lifestyle habits are also mandatory.

Response to any type of treatment necessarily implies a decrease in anti-PLA2R antibodies, initially accompanied or not by a proteinuria reduction, which is usually posterior to antibodies drop.

∗The second cycle of immunosuppression depends on the first used treatment. For patients with rituximab as initial treatment, a one-second cycle of rituximab can be used with the addition of (if kidney function is stable) or without the addition of CNIs. In addition, a valid approach is directly changing to CYC+GC if there is a decrease in kidney function (especially in patients with eGFR <45 ml/min). For patients with CYC+GC as initial treatment, rituximab is a valid option in the second cycle of treatment, or if the patient had a good initial tolerance to alkylating agents, give a second course of CYC+GC, trying to use the lowest effective dose possible.

∗∗A refractory MN implies nonresponse to 2 cycles of different treatments, including at least, a course of CYC-GC.

Before insisting on further immunosuppression, it would be important, if the patient's clinical condition allows, to perform a kidney biopsy to rule out the presence of chronic irreversible changes in the renal tissue that would justify discontinuing immunosuppression. If immunosuppression is still feasible, treatments targeting B cells (obinutuzumab or belimumab) or plasma cells (bortezomib or daratumumab) could be given. Other alternatives include the combination of different drugs (rituximab-belimumab, rituximab-cyclophosphamide, etc.), patient derivation to expert centers in the treatment of glomerular diseases, or inclusion in RCTs focused on new therapeutic targets. (2b). The therapeutic approach to the patients depends to a great extent on the baseline values (risk of progression) and the monitoring of the anti-PLA2R antibodies. In case of values <50 RU/ml, the probability of spontaneous remission is very high and only periodic monitoring is required. On the contrary, very high baseline values (>150 RU/ml) justify the start of immunosuppressive therapy. For intermediate values, more frequent monitoring can be performed, and immunosuppression can be started if an increase in anti-PLA2R or proteinuria is evident. The response to treatment is also assessed by the decrease in anti-PLA2R antibodies that usually precedes the reduction in proteinuria by weeks or months. The black curves imply a poor outcome in response to immunosuppressive treatment, and the blue curves show an adequate response of anti-PLA2R antibodies after immunosuppressive treatment or supportive treatment.

ACEi, angiotensin converting-enzyme inhibitors; aPLA2R, anti-PLA2R autoantibodies; ARB, angiotensin II receptor blocker; BP, blood pressure; CNI, calcineurin inhibitors; CYC, cyclophosphamide; GC, glucocorticoids; GFR, glomerular filtration rate; IS, Immunosuppression or immunosuppressive; MN, membranous nephropathy; MRA, mineralocorticoid receptor antagonist; PLA2R, type M phospholipase A2 receptor; RCTs, randomized clinical trials, RTX, Rituximab; ST, supportive therapy; uPCR, urinary protein-creatinine ratio (or its equivalent in g/24 h).

Supportive Therapy: Optimized Kidney and Cardiovascular Protection

Nonspecific, nonimmunosuppressive treatment involves restricting dietary sodium to less than 2 g/d; restricting protein intake (0.8−1 g/kg/d); and controlling blood pressure (systolic <120 mmHg), hyperlipidemia, and edema to decrease kidney and cardiovascular risks.³ RAS blockers should be up-titrated to maximally tolerated or allowed dose to reduce proteinuria. More recently SGLT2 inhibitors improved kidney and cardiovascular outcomes in both diabetic and nondiabetic chronic kidney disease (CKD).⁴ Specifically, the Dapagliflozin in Patients with Chronic Kidney Disease (DAPA-CKD) and Empagliflozin in Patients with Chronic Kidney Disease (EMPA-KIDNEY) trials enrolled a combined total of 136 patients with MN, representing 14% of the “other glomerulonephritis” category, in which dapagliflozin or empagliflozin decreased the risk of CKD progression by 32% (hazard ratio 0.68, 95% confidence interval 0.46–1.02).⁴ However, results for MN were not specifically reported. Beyond the kidney, the 2021 ESC cardiovascular prevention guidelines suggest considering adding an SGLT2 inhibitor to the standard of therapy for cardiovascular disease prevention purposes in nondiabetic patients with CKD and recommend their use if the CKD patient is diabetic.⁵ Experience with steroidal mineralocorticoid receptor antagonist in MN has been generally negative from the point of view of tolerance and effectiveness on proteinuria.⁶ However, finerenone, a nonsteroidal mineralocorticoid receptor antagonist, improved kidney and cardiovascular outcomes in patients with diabetes and CKD, the best results being observed in patients on both SGLT2 inhibitor and finerenone.⁷ However, no data are available for patients with MN. In addition, these agents have relatively weak antiproteinuric effect and unlikely to significantly improve proteinuria in patients with severe NS (e.g., >10 g/24 h). In addition, the majority (80%) of patients with MN are normotensive and unable to tolerate maximum RAS blockade let alone the combined use of a mineralocorticoid receptor antagonist. Nevertheless, a holistic approach to cardiorenal protection combining RAS blockade and SGLT2 inhibitor (adding a nonsteroidal mineralocorticoid receptor antagonist will depend on blood pressure levels) could be explored as supportive therapy in patients who reach immunologic remission but remain with subnephrotic proteinuria following immunosuppression therapy in MN.

Classical Immunosuppression

Current recommendations for immunosuppressive treatment of MN are based on the stratification of the risk of progression.³ Immunosuppressive treatment should be started for persistent NS or proteinuria despite RAS blockers, impaired renal function and/or high anti-PLA2R autoantibody titers (Figure 2). The evidence on classical immunosuppressants is summarized in Table 2, and their advantages, and limitations are summarized in Table 3.

Table 2.

Summarized evidence of major recent clinical trials in membranous nephropathy

Study	Year	Target	Design	Analysis	N	Primary end point	Interventions	CR + PR	CR	IR	Relapses	SAEs	ESRD
GEMRITUX8	2017	B cells	Open-RCT Phase 3 (Superiority)	ITT and PP	78	6 moa	RTX vs. NIAT	35% vs. 21%	19% vs. 3%	50% vs. 12%	Not analyzed	16% vs. 13%	0% vs. 0%
MENTOR9	2019	B cells	Open-RCT phase 3 (Noninferiority)	ITT and PP	130	24 mo	RTX vs. CsA	60% vs. 20%b	35% vs. 0%	66% vs. 13%	5% vs. 53%	17% vs. 31%	0% vs. 2%
STARMEN10	2020	T and B cells	Open-RCT, phase 3 (Noninferiority)	ITT and PP	86	24 mo	CYC+GC vs. TAC+RTX	84% vs. 58%b	60% vs. 26%	88% vs. 83%	2.7% vs. 12%	19% vs. 14%	2% vs. 0%
RI-CYCLO11	2021	T and B cells	Open-RCT (Pilot)	ITT and PP	74	12 moc	CYC+GC vs. RTX	73% vs. 62%	32% vs. 16%	56% vs. 62%	22% vs. 13%	14% vs. 19%	5% vs. 0%
Ramachandran12	2021	T cells	Open-RCT phase 3 (Noninferiority)	ITT and PP	70	72 mo	CYC+GC vs. TAC+GC	88% vs. 53%b,d	59% vs. 22%	Not specified	33% vs. 64%	11% vs. 17%e	Not specified

CR, complete remission; CsA, cyclosporine A; CYC, cyclophosphamide; eGFR, estimated glomerular filtration rate (expressed in ml/min per 1.73 m²); ESRD, end-stage renal disease; GC, glucocorticoids, IR, immunologic remission (defined as negative anti-PLA2R); ITT, intention to treat analysis; N, sample size; NIAT, nonimmunosuppressive antiproteinuric therapy; PP, per protocol analysis; PR, partial remission; RASi, renin-angiotensin system inhibitors; RCT, randomized clinical trial; Ref, reference number; RTX, Rituximab; SAESs, serious adverse events; TAC, tacrolimus.

Note that none of the ESRD cases was reported in patients randomized to rituximab. All patients were adults with biopsy-proven primary nephropathy, with nephrotic proteinuria, eGFR ≥ 30 ml/min per 1.73 m², and on top of RASi therapy.

^aIn the extended follow-up (median 17 mo), post hoc analysis showed 65% vs. 34% in favor of rituximab.

^bP-value < 0.05 (including relapses);

^cSecondary end point at 24 mo showed no differences in remission rates between the 2 regimens.

^dResults were 62% vs. 28% for relapsed-free remission (P-value < 0.05).

^eConsidering only deaths and major infections.

Table 3.

Advantages and limitations of immunosuppressive drugs currently used in membranous nephropathy

	Therapeutic intervention
Characteristic	CYC+GC	CNIs	Rituximab
Evidence-supporting studies	RCT, cohort studies	RCT, cohort studies	RCT, cohort studies
More recommended profile patient	Very-high risk	Moderate risk	Moderate and high risk
Short-term efficacya (3–6 mo)	74%–79%	44%–74%	35%–60%
Medium-term efficacya (18–24 mo)	84%–86%	20%–75%	60%–80%
Long-term efficacya (>24 mo)	80%–88%	53%	60%–65%
Nonresponse (24 mo)	15%–20%	25%–80%	25%–30%
Relapses (24 mo)	3%–33%	53%–64%	5%–13%
Main adverse effects	Cytopenia Severe infection Cushing syndrome, Infertility, cancerb	Nephrotoxicity HT, Hyperkalemia Metabolic disorders Distal tremor	Infusion reaction Leukopenia (rare) Mild infection (rare) Ig depletionc
NNT-Bd (95% CI) for CR+PR	4 (2–14)	4 (2–13)	3 (2–4)
Patient tolerance	Low-moderate	Moderate	High
Patient adherence	Variable	Variable	High
Costse	Low	Low-medium	Medium-high

95% CI, 95% confidence interval; CR, complete remission; CNIs, calcineurin inhibitors (includes cyclosporine and tacrolimus); CYC, cyclophosphamide; GC, glucocorticoids; HT, hypertension; Ig, immunoglobulins; NNT-B, number needed to treat (Benefit); PR, partial remission; RCT, randomized controlled trial; SAEs, serious adverse event.

^aEfficacy was defined as complete and partial remission (CR + PR). Data were obtained from major RCTs, and cohort studies included in this review.

^bEspecially with the use of high doses and by long duration.

^cUncommon effect. This is more frequent in antineutrophil cytoplasmic antibodies vasculitis, principally because of repeat dose of rituximab.

^dThe NNT-B indicates the number of patients who need to be treated with the drug to obtain a clinical benefit (in our case, to achieve CR+PR) and is calculated as the inverse of the risk difference of outcome between the experimental drug and the control drug. The optimal value is 1, which means that for each patient treated, the desired outcome is obtained. In this review, NNT-B was calculated for CR+PR at 24 months, as following: from the STARMEN study for CYC+GC, from Ramachandran et al.¹² study for CNIs, and from the MENTOR study for rituximab.

^eThey may vary depending on the final dose administered of rituximab, the type of CNI, the geographic region and the type of health care system.

Rituximab: the First Option for Most Patients

According to 2021 KDIGO guidelines, for moderate-high risk MN, the first option is rituximab (±CNIs).³ Rituximab is an anti-CD20 chimeric IgG1 monoclonal antibody that depletes CD20⁺ pre-B/mature B cells (Figure 1) for at least 6 to 12 months through complement-dependent cytotoxicity, antibody-dependent cellular cytotoxicity, and apoptotic cell death. Rituximab may also protect podocytes by stabilizing sphingomyelin-phosphodiesterase-acid-like-3b expression and preventing downregulation of acid-sphingomyelinase activity, thereby decreasing actin cytoskeleton disruption and apoptosis.¹³

In observational studies, rituximab was effective in reducing anti-PLA2R antibody titers and nephrotic proteinuria, improving kidney function, with an acceptable safety profile.¹⁴ In an Italian cohort of 100 nephrotic MN patients, rituximab induced complete remission or partial remission (CR+PR) in 65% after 29 months, without differences according to prior immunosuppressors, with good tolerance, and without serious adverse events (SAEs).¹⁵ Several observational studies showed that serum anti-PLA2R antibody reduction preceded the proteinuria remission, becoming a potential early marker of response to treatment.16, 17, 18

GEMRITUX was the first randomized controlled trial (RCT) with rituximab,⁸ After 6 months, rituximab showed no difference when compared to supportive therapy for CR+PR (35% vs. 21%) in 75 nephrotic MN patients, although serum albumin increased and anti-PLA2R antibodies decreased. Interestingly, after 17 months, rituximab was significantly superior to RAS blockade (CR: 19% vs. 3%, CR+PR: 65% vs. 35%, respectively). SAEs were uncommon, but contrary to previous results, kidney function was lower with rituximab.

MENTOR, a multicenter and noninferiority RCT in 130 patients with persistent nephrotic MN, showed that rituximab (1 g, 2 doses, repeated at 6 months in patients who showed >25% reduction in proteinuria at 6 months) was superior to cyclosporine A (CsA, 3−5 mg/kg/d for 6−12 months) for achieving CR+PR from month 18 (62% vs. 23%) up to month 24 (60% vs. 20%).⁹ Treatment failure was defined as reduction of proteinuria <25% at 6 months. Rituximab induced a faster (52% vs. 28% at 6 months) and persistent immunologic response, and a better follow-up kidney function, compared to patients treated with CNI who exhibited persistent nephrotoxicity at 24 months despite CNI discontinuation. Adverse events (AEs) were similar in both groups, but more SAEs were seen with CsA (31% vs. 17%).

There is also evidence on rituximab efficacy in refractory or advanced MN. In a prospective study of 36 Chinese MN patients without response to prior immunosuppression, rituximab produced CR+PR in 42% of patients.¹⁹ Similar efficacy has been shown in Italian patients who failed to respond to previous immunosuppressive treatment.²⁰ Even more, when anti-PLA2R titers remain persistently high, a second course of rituximab may be effective.²¹ In a retrospective analysis of 13 nephrotic MN patients with anti-PLA2R and eGFR <30 ml/min per 1.73 m², rituximab improved kidney function with remission of NS after 41 months in 69% (9/13) of patients, and immunologic remission was associated with clinical response.²² A recent Indian 2-year prospective observational study with 64 incident and resistant nephrotic MN patients with eGFR 30 to 60 ml/min per 1.73 m² showed similar proteinuria remissions between rituximab and cyclophosphamide-glucocorticoids (38.5% vs. 49%), with a better safety profile for rituximab.²³

CNIs: Valid but Secondary Partners

CNIs (CsA, tacrolimus) are effective in the treatment of MN because of immunosuppressive and antiproteinuric effects. CsA forms a complex with cyclophilin, whereas tacrolimus binds to FK-binding protein FKBP12. Both inhibit calcineurin, inactivating the nuclear factor of activated T cells and decreasing the synthesis of proinflammatory cytokines (Figure 1). In podocytes, CNIs block calcineurin-dependent synaptopodin dephosphorylation, preventing its cathepsin-mediated degradation, and preserving the actin cytoskeleton.²⁴

CNIs induce remission of NS in 70% to 75% of MN patients in comparison with nonimmunosuppressive therapy;²⁵ and in Asian patients, the efficacy could be similar and even superior to cyclophosphamide, with fewer AEs.^12,26 However, the frequent relapse after withdrawal (∼50%) and nephrotoxicity are major limitations. Alternatively, the combination of CNI-rituximab can be an option in patients with high risk of relapse postwithdrawal²⁷ and was explored in the STARMEN trial discussed below.

Cyclophosphamide: Confirmed Efficacy, but Safety Concerns

Cyclophosphamide is an alkylating agent, whose active metabolite, phosphoramide mustard, decreases T-cell (CD4⁺ > CD8⁺) and B-cell function, including pathogenic autoantibody production (Figure 1).

Alternating cyclophosphamide-glucocorticoids is the main therapeutic option for patients with very high-risk of progression, life-threatening NS, or rapid or progressive deterioration of kidney function.^1,3 The initial evidence over 30 years ago, showed that alkylating agents (cyclophosphamide, chlorambucil) were superior to supportive treatment to reach proteinuria remission and prevent CKD progression.³ Long-term follow-up of cyclophosphamide-treated patients confirmed these initial findings.²⁸ However, cyclophosphamide-glucocorticoids induced more SAEs, including increased risk of cancer, and these small studies lacked appropriate control groups (suboptimal blood pressure target and diet, poor uptake of RAS blockers). Importantly, a European comparative cohort study, showed that over a median follow-up of 40 months, rituximab induced significantly fewer AEs than cyclophosphamide for serious (including cancer and death) and nonserious AEs (24% vs. 61%), but achieving similar efficacy for complete remission (40% vs. 42%).²⁹ Recently, better quality evidence emerged from trials that compared cyclophosphamide-glucocorticoids with regimens that included rituximab, with a more appropriate supportive therapy.

STARMEN was an open-label RCT of 86 nephrotic MN patients.¹⁰ At 24 months, cyclophosphamide-glucocorticoids were superior to sequential treatment with tacrolimus (9 months) plus rituximab (1 g at 6 months) for achieving CR+PR and CR (84 vs. 58%, and 60 vs. 26%, respectively). This superiority was observed starting at month 3, but there were no differences at 24 months in the proportion of patients with preserved kidney function (eGFR ≥45 ml/min per 1.73 m²) (93% vs. 86%). The immunologic response occurred earlier with cyclophosphamide-glucocorticoids but was similar after 24 months (88% vs. 83%). Cyclophosphamide-glucocorticoids induced more AEs per patient, without relevant differences in SAEs (19% vs. 14%). Within the sequential treatment regimen, most of AEs were related to tacrolimus. The addition of rituximab after 6 months reduced the relapse rate post-tacrolimus withdrawal (∼12%) as compared with historic studies but resulted in a lag time in anti-PLA2R antibodies reduction of 6 months from the first cyclophosphamide-glucocorticoids dose, which together with the low rituximab dose used could have limited its efficacy.

In the RI-CYCLO pilot trial of 75 nephrotic MN patients with relatively preserved kidney function (eGFR 84 ± 24 ml/min per 1.73 m²) and milder proteinuria (6 g/24h), CR and CR+PR at 12 months for cyclophosphamide-glucocorticoids and rituximab (1g, days 1−15) were 32% versus 16%, and 73% versus 62%, respectively.¹¹ At 24 months, in which 77% of the initial population was assessed, CR and relapses were 35% versus 42%, and 22% versus 13% for cyclophosphamide-glucocorticoids and rituximab, the difference no longer significant, respectively. The immunologic responses and the percentage of patients with AEs and SAEs were similar at 12 months. In their subgroup analysis, both trials showed greater efficacy of cyclophosphamide-glucocorticoids in males and in patients with more severe disease.

Other cyclophosphamide-based regimens have been tried, including the use of lower doses, and intravenous routes with encouraging results.³⁰ In a retrospective cohort study of 60 nephrotic MN patients, the combination of rituximab, low-dose cyclophosphamide, and a rapid tapering glucocorticoids regimen was effective and safe, achieving 100% of immunologic response at 6 months, 83% of CR at 24 months, and a low relapse rate (10%).³¹

In patients with advanced CKD, the evidence is scarce. In a British multicenter RCT of 108 nephrotic MN patients with creatinine clearance ≤50 ml/min and progressive loss of kidney function, cyclic chlorambucil-based regimen diminished the probability of kidney function reduction by ≥20% with respect to CsA or supportive therapy, after 3 years of follow-up, but was associated with more AEs and SAEs, especially cytopenias.³²

We conclude that alternating cyclophosphamide-glucocorticoids is an effective option to treat MN. Tolerance and safety concerns could be reduced by using lower doses, especially less intravenous methylprednisolone, shorter treatment courses, or combining with rituximab, although these therapeutic schemes require validation through appropriate RCTs.

Monitoring of MN: Role of Anti-PLA2R Antibodies

The presence or absence of anti-PLA2R antibodies adds important information to clinical and immunopathologic data in patients with MN. Levels of anti-PLA2R antibodies tightly correlate with disease activity. Low baseline (<50 RU/ml) and decreasing anti-PLA2R antibody levels strongly predict spontaneous remission, thus favoring conservative therapy. Conversely, high baseline (>150 RU/ml) or increasing anti-PLA2R antibody levels associate with NS and progressive loss of kidney function, thereby guiding initiation of immunosuppressive therapy (Figure 2b). Serum anti-PLA2R antibody profiles reliably predict response to therapy, and levels at completion of therapy can forecast long-term outcome. Re-emergence of or increase in antibody titers precedes a clinical relapse. Therefore, an individualized serology-based approach to the diagnosis and treatment of MN was proposed.³³

Recent trials GEMRITUX, MENTOR, STARMEN, and RI-CYCLO have confirmed that approach showing that an early decline of serum anti-PLA2R titers predicts response to treatment.³⁴ Undetectable anti-PLA2R allows discontinuation of immunosuppression, or progressive dose reduction for CNI. Persistent anti-PLA2R antibody levels (after initial fall), in the presence of recovering CD20+ B cells (>5 cells/μl) may require a second dose of rituximab. Increasing anti-PLA2R levels despite immunosuppressive therapy requires modification of therapy.^1,3. Anti-PLA2R titers at 6 months are key to decide the continuation of treatment. Absent or low titers allow immunosuppressant withdrawal and/or tapering for CNI. Patients with persistent or increasing levels anti-PLA2R levels without proteinuria remission despite adequate immunosuppression should be considered as resistant to therapy (Figure 2b).

Novel Approaches: Targeted Therapies

AEs associated with classic immunosuppressants and the lack of response in 20% to 30% of patients has led to the search of alternative therapies.^1,3 These include new therapeutic options aiming at more effectively B-cell depletion or targeting plasma cells, which could potentially produce pathogenic autoantibodies (Figure 1).

There is no standardized definition of resistant or refractory MN. Before the clinical use of anti-PLA2R antibodies, resistant or refractory NM was based on persistent proteinuria after immunosuppressive therapy. Currently, for patients with PLA2R-associated MN, we define resistant as persistence of anti-PLA2R antibodies despite adequate immunosuppression. For patients treated with rituximab that means unchanged or increasing anti-PLA2R levels despite persistent anti-CD20+ B cells depletion (e.g., 0 cells/μl) at 6 months. The same is true for patients treated with cyclical cyclophosphamide-based regimen. Patients treated with a CNI at adequate target levels that show no changes or increasing anti-PLA2R levels at 6 months should also be considered resistant to therapy (Figure 2a and 2b).

Failure to respond to one therapy does not implies resistant to alternative therapies. Therefore, if there is lack of response to initial treatment with rituximab, patients can be considered for treatment with a CNI if kidney function is preserved, or switching to a cyclical CYC-GC regimen. Regardless of treatment use, we recommend monitoring anti-PLA2R and proteinuria levels every 2 to 3 months, initially (Figure 2b). Patients who are resistant to 2 or more lines of treatment with currently available immunosuppressants are candidates for new therapeutic approaches,^1,3 which we discuss in the next section.

Alternative or Additional Anti-B-cell Therapies: Obinutuzumab and Belimumab

Obinutuzumab is a humanized anti-CD20 type-II monoclonal antibody that produces greater cytotoxicity and B-cell apoptosis than rituximab, by targeting a different epitope on CD20 and has a higher affinity for Fcɣ-RIII, thus enhancing antibody-dependent cellular cytotoxicity and phagocytosis by NK cells and macrophages (Figure 1). It is effective in B-cell neoplasms, and recently, as add-on therapy in patients with lupus nephritis, including membranous lupus glomerulonephritis. In 2 case series, 1 of them including nephrotic patients with primary and post-transplant MN, refractory to several immunosuppressors, obinutuzumab depleted anti-PLA2R antibodies and induced proteinuria remission in >60% of patients.^35,36 A pilot, open-label, and noncontrolled Italian phase-2 trial will enroll 20 nephrotic MN patients with eGFR ≥30 ml/min per 1.73 m² who are resistant, dependent, or intolerant to rituximab, to evaluate CR+PR at 12 months and the safety profile (NCT05050214). Another randomized open-label US phase-3 trial will enroll 140 nephrotic MN patients with eGFR ≥40 ml/min per 1.73 m² to compare CR at 2 years between obinutuzumab and tacrolimus (NCT04629248). Results are expected in early 2025.

Belimumab is a human IgG1λ monoclonal antibody that binds to soluble B-lymphocyte-stimulating protein (BlyS/BAFF/TNFSF13B), blocking binding to receptors on B cells. Thus, belimumab decreases B-cell survival and differentiation to immunoglobulin-producing plasma cells (Figure 1). Belimumab has been authorized to treat active lupus nephritis following the BLISS-LN 2-year phase-3 trial, which included patients with membranous class. In a recent open-label prospective British study with 14 persistent nephrotic MN patients (3 with prior immunosuppression), belimumab (10 mg/kg/month for 104 weeks) induced CR+PR in 9 patients (64%). Anti-PLA2R reduction was observed from week 12 onwards and proteinuria reduction from week 36.³⁷ There were few AEs and no patient deaths. Exploring potential add-on effect on CD20+ cell depletion and efficacy, an ongoing multicenter phase-2, double-blind RCT will compare the impact of belimumab-rituximab versus rituximab on CR after 2 years in 124 nephrotic adult MN patients with eGFR ≥30 ml/min per 1.73 m² (NCT03949855). Results are expected in 2027.

Targeting Plasma Cells: Proteasome Inhibitors and Anti-CD38 Agents

Antiplasma cell therapy is a logical approach for patients with relapsing MN or resistance to conventional immunosuppressants. Bortezomib is a selective and reversible proteasome inhibitor in high-turnover plasma cells, producing caspase-mediated apoptosis and plasma cell depletion (Figure 1). It is successfully used to treat multiple myeloma and non-Hodgkin lymphoma. It can induce response in refractory lupus nephritis, and similarly, in refractory nephrotic MN patients, producing anti-PLA2R depletion and proteinuria remission, including CR at 12 months.38, 39, 40

Daratumumab and felzartamab (MOR202) are human IgG1κ monoclonal antibodies targeting CD38 protein, which is highly expressed on plasmablasts and short or long lived plasma cells, that depletes plasma cells through complement-dependent cytotoxicity, antibody-mediated cytotoxicity, and antibody-dependent cellular phagocytosis. Daratumumab is approved for the treatment of multiple myeloma and light-chain amyloidosis but has not yet been formally studied in MN. One recent case report showed that daratumumab induced a rapid anti-PLA2R and proteinuria reduction in a woman with MN refractory to cyclophosphamide and rituximab, but the effect was short lasting, probably because of induced B-cell hyperreactivity.⁴¹ The M-PLACE proof-of-concept study (NCT04733040) has so far shown that felzartamab rapidly and substantially reduces anti-PLA2R antibodies titers in patients with anti-PLA2R+ MN (interim results from the M-PLACE Study. ASN Kidney Week. Published 2021. Accessed May 30, 2022. https://www.asn-online.org/education/ kidney week/2021/program-abstract.aspx? control). A small open-label phase-2 trial will assess the efficacy, safety, and pharmacokinetics or pharmacodynamics of felzartamab in nephrotic anti-PLA2R⁺ MN patients (NCT04145440), and another trial is evaluating felzartamab as rescue therapy in MN patients with prior failure of anti-CD20 therapies (NCT04893096). These studies would be completed in 2024 (Table 4).

Table 4.

Ongoing trials with novel immunomodulators agents

Study	Region	Drug	Target (MoA)	Design	N	Selection criteria	Follow-up (wks)	Control	Primary outcome	Secondary Outcomes	Expected end-date
NCT04629248	USA	Obinutuzumab	B cell (anti-CD20)	Open-RCT (Phase III)	140	Adults 18 to 75 yr UPCR >5 g/g eGFR ≥ 40 ml/min/1.73 m2	104	Tacrolimus	CR	CR+PR Δ eGFR Δ anti-PLA2R Adverse events	June 2028
NCT05050214 (ORION Study)	Italy	Obinutuzumab	B cell (anti-CD20)	Noncontrolled clinical trial (Pilot)	20	Adults ≥ 18 yr Proteinuria >3.5 g/24 h eGFR ≥ 30 ml/min/1.73 m2 Intolerance, resistance, or Dependence to RTX	52	None	CR+PR Adverse events	CR+PR Δ eGFR Δ anti-PLA2R	April 2025
NCT03949855 (REBOOT Study)	USA	Belimumab + RTX	B cell (anti-BLyS and anti-CD20)	Open-RCT (Phase III)	124	Adults 18–75 yr Proteinuria ≥ 4 g/24 h eGFR ≥ 30 ml/min/1.73 m2 Anti-PLA2R+	104	RTX + Placebo	CR	CR at 52–156 wks PR at 52104-156 wks Relapses - Adverse events	October 2026
NCT04893096 (Rescue therapy for resistant patients to CD20 therapy)	Italy	Felzartamab (MOR202)	Plasma cell (anti-CD38)	Open single noncontrolled trial	10	Adults ≥ 18 yr Prot 24 h >3.5 g/24 h eGFR > 30 ml/min/1.73 m2 Dependence/resistance to RTX	24	None	CR+PR Δ Proteinuria 24h	- Not specified	October 2024
NCT04145440 (M-PLACE Study)	USA (cohort 1) & France (cohort 2)	Felzartamab (MOR202)	Plasma cell (anti-CD38)	Open single clinical trial (Phase Ib-IIA)	30	Adults 18–80 yr Proteinuria ≥ 3.5 g/24 h or UPCR ≥3 g/g eGFR ≥ 50 ml/min/1.73 m2 Anti-PLA2R+	24–52	None	Safety and tolerability (at 24 weeks)	Δ anti-PLA2R (1 year) Immunogenicity (1 yr) Adverse events (1 yr)	November 2022
NCT04733040 (NewPLACE Study)	Europe	Felzartamab (MOR202)	Plasma cell (anti-CD38)	Open single clinical trial (Phase IIa)	24	Adults 18–80 yr Proteinuria ≥ 3.5 g/24 h or UPCR ≥3 g/g GFR ≥ 50 ml/min/1.73 m2 Anti-PLA2R > 50 RU/ml	12–96	None	Δ anti-PLA2R	Immunologic complete response Overall proteinuria response Adverse events, PK	January 2024
NCT05162066 (RENEW study)	Europe	BCX9930	Factor D inhibitor	Open single clinical trial (Proof of concept)	42 (14 MN)	Adults ≥18 yr GFR ≥ 50 ml/min/1.73 m2 Body weight ≥ 40 kg IgAN, MN, C3G	24	None	Δ Proteinuria 24 h, or Δ UPCR	Partial remission Δ eGFR	July 2023
NCT04154787	USA Argentina Europe China	Iptacopan (LNP023)	Factor B inhibitor	Open-RCT (Proof of concept)	52	Adults ≥18 years GFR≥30 ml/min/1.73m2 Proteinuria ≥3.5/24 h Anti-PLA2R≥60 RU/ml	24	Rituximab	Δ UPCR	CR, PR Δ eGFR Plasma levels of Bb & sC5b-9 PK	March 2024
NCT02682407	USA	Narsoplimab (OMS721)	MASP-2 inhibitor (Lectin pathway)	Open single clinical trial (Phase II)	52	Adults ≥18 yr eGFR≥30 ml/min/1.73 m2 IgAN, MN, LN, C3G	104	None	Adverse events	PK of Narsoplimab Δ UPCR and Δ UACR	Not actualized date

C3G, C3 glomerulopathy; CR, complete remission; eGFR, estimated glomerular filtration rate; IgAN, IgA nephropathy; LN, lupus nephritis; MN, membranous nephropathy; MoA, mechanism of action; PK, pharmacokinetics, PR, partial remission; RCT, randomized clinical trial; UACR, urinary albumin/creatinine ratio; UPCR, urinary protein/creatinine ratio; Δ, change.

Targeting Complement System: Beyond Thrombotic Microangiopathy

Despite evidence of a key role of the complement system in the pathogenesis of MN,² currently there is no published experience with complement inhibitors, beyond some case reports of patients with primary or post-transplant MN patients who had concurrent thrombotic microangiopathy and were treated with eculizumab.⁴² Ongoing clinical studies are exploring the inhibition of several complement pathways (NCT05162066/ NCT04154787/NCT02682407) (Figure 1, Table 4).

Conclusions

MN is considered the prototype of organ-specific autoimmune disease. The current therapeutic approach is based on the risk of progression. All patients should receive maximum antiproteinuric therapy to achieve adequate renal and cardiovascular protection. In case of no response, or in high-risk patients (e.g., anti-PLA2R levels >150 RU/ml), treatment with immunosuppressants should be started promptly, mainly with rituximab because of its efficacy and safety, with or without CNIs. Very high-risk cases or with rapid progression can be treated with cyclophosphamide-glucocorticoids based regimens. New drugs targeting B cells and plasma cells are currently options for refractory cases, although their clinical efficacy in a broader patient population needs to be confirmed in ongoing RCTs (Table 4). The efficacy of complement inhibitors is under assessment. Future studies should standardize the endpoints, optimize the doses of the immunosuppressants, and minimize the associated risks (Table 5).

Table 5.

Unmet needs in Membranous Nephropathy to be addressed in future research

Issue	Potential solutions in future research
Safety, tolerance, and doses	CYC-based regimens for shorter and/or lower doses Minimize the dose and duration of glucocorticoids Optimize dose of rituximab, based on CD20-B cell and anti-PLA2R response Pharmacogenomics-guided prescription of immunosuppressants
Multitarget therapy	Design RCTs with combined or sequential therapies, directed toward different immunologic targets, including old and new immunomodulators (e.g., RTX-CYC, Belimumab-RTX, RTX-CNIs, etc.)
Efficacy, effectiveness and cointerventions	To include immunologic response as part of the complete remission definition (considering time to response) Randomization of patients according to key baseline clinical factors (age, baseline proteinuria, kidney function, and anti-PLA2R levels) Studies with a minimum of 3–5 years of follow-up Extended follow-up in recently published RCTs to analyze potential long-term risks (kidney failure, latent infections, neoplasms, death) Lower dose and slower tapering in case of use of CNIs Carry out cost-effectiveness and cost-utility studies, considering hard endpoints and the differences between the various health systems
Selection criteria	Inclusion of patients with advanced CKD (eGFR up to 30 ml/min/1.73 m2) Balanced representation of all ethnic groups
Refractory and relapsing MN	Standardization of criteria for refractory or resistant disease
Biobank	Generate biobank to study new neoantigens, autoantibodies, and early markers of response to therapy and relapse

CKD, chronic kidney disease; CNIs, calcineurin inhibitors; CYC, cyclophosphamide; eGFR, estimated glomerular filtration rate; MRAs, mineralocorticoids receptor antagonists; MN, membranous nephropathy; PLA2R, phospholipase A2 receptor; RCTs, randomized clinical trials; RTX, Rituximab; SGLT2i, sodium-glucose cotransporter type 2 inhibitor; THSD7A, thombospondin type-1 domain-containing protein 7A.

Disclosure

JER-R has received consultancy or speaker fees from GSK, Otsuka, and Alexion. AO has received grants from Sanofi and consultancy or speaker fees or travel support from Advicciene, Astellas, Astrazeneca, Amicus, Amgen, Fresenius Medical Care, GSK, Bayer, Sanofi-Genzyme, Menarini, Mundipharma, Kyowa Kirin, Alexion, Freeline, Idorsia, Chiesi, Otsuka, Novo-Nordisk, Sysmex, and Vifor Fresenius Medical Care Renal Pharma and is Director of the Catedra Mundipharma-UAM of diabetic kidney disease and the Catedra Astrazeneca-UAM of chronic kidney disease and electrolytes. FCF reports having consultancy agreements with Alexion Pharmaceuticals, Alnylam, and ByoCrystal; receiving research funding from Chemocentryx, Genentech Inc., Janssen Pharmaceutical, and Questcor/Mallinckrodt; being a scientific advisor for, or member of, Kidney International, Nephrology, Nephrology Dialysis Transplantation, and UpToDate; and receiving honoraria from UpToDate.