Membranous nephropathy: Treatment and prognosis

INTRODUCTION — Membranous nephropathy (MN) is among the most common causes of the nephrotic syndrome in nondiabetic adults, accounting for up to one-third of biopsy diagnoses in some regions. The term MN reflects the pattern of histologic change noted on light microscopy: glomerular basement membrane (GBM) thickening with little or no cellular proliferation or infiltration.

MN in adults is most often primary (approximately 75 to 80 percent of cases) and caused by circulating autoantibodies against podocyte antigens. In approximately 20 to 25 percent of cases in adults, the MN lesion is associated with various disorders, including infections (such as hepatitis B and syphilis); autoimmune diseases; malignancies; allogeneic hematopoietic stem cell transplantation; and the use of certain drugs such as nonsteroidal antiinflammatory drugs (NSAIDs), alpha-lipoic acid, and certain traditional medicines.

The treatment and prognosis of MN in the native kidney will be reviewed here. Other aspects of MN are presented separately:

●(See "Membranous nephropathy: Pathogenesis and etiology".)

●(See "Membranous nephropathy: Clinical manifestations and diagnosis".)

●(See "Membranous nephropathy and kidney transplantation".)

TERMINOLOGY — The phospholipase A2 receptor (PLA2R) is the most important target antigen in patients with primary MN (see "Membranous nephropathy: Pathogenesis and etiology", section on 'Phospholipase A2 receptor'). The following terminology will be used throughout this topic:

●PLA2R-associated MN – MN with either positive immunofluorescence staining for PLA2R antigen on kidney biopsy or with detectable serum anti-PLA2R antibodies. The most sensitive technique to diagnose PLA2R-associated MN is immunostaining of the kidney biopsy for PLA2R antigen. In the majority of patients with PLA2R-associated MN, circulating autoantibodies directed against PLA2R can be detected in the serum.

●Non-PLA2R-associated MN – MN with neither immunofluorescence staining for PLA2R antigen on kidney biopsy nor detectable serum anti-PLA2R antibodies. Patients with non-PLA2R-associated MN may be positive for other target antigens associated with MN, such as thrombospondin type-1 domain-containing 7A (THSD7A), neural epidermal growth factor-like 1 (NELL-1), semaphorin 3B (Sema3B), protocadherin 7 (PCDH7), and other novel antigens [1]. (See "Membranous nephropathy: Pathogenesis and etiology", section on 'Other antigens'.)

●Anti-PLA2R antibody positive – In a patient with PLA2R-associated MN, circulating autoantibodies directed against PLA2R are detectable.

●Anti-PLA2R antibody negative – In a patient with PLA2R-associated MN, circulating autoantibodies directed against PLA2R are not detectable. In these patients it is likely that the kidneys act as a "sink" by binding all available anti-PLA2R autoantibodies. If disease activity persists, anti-PLA2R antibodies may become detectable with longer follow-up [2]. Alternatively, some patients with undetectable antibodies may have already developed an immunologic remission; in such patients, a gradual decrease in proteinuria will become apparent.

PRETREATMENT CONSIDERATIONS

Likelihood of spontaneous remission — In view of the potential toxicity of the drugs used to treat primary MN, the decision to initiate therapy is based, in part, upon an understanding of the natural history of untreated patients, with and without features of the nephrotic syndrome at presentation [3-6]:

●Spontaneous complete remission of proteinuria occurs in 5 to 30 percent at five years [4-7].

●Spontaneous partial remission (proteinuria ≤2 g/day) occurs in 25 to 40 percent at five years [4-6].

●The occurrence of end-stage kidney disease (ESKD) in untreated patients with nephrotic syndrome is approximately 14 percent at five years, 35 percent at 10 years, and 41 percent at 15 years [3,6]. However, among untreated patients who remain non-nephrotic for the duration of their disease, the rate of ESKD at 10 years may be as low as 2 percent [8].

Risk factors for disease progression — In view of the often indolent clinical course and toxicity of available therapies, immunosuppressive agents should be considered only in patients with primary MN who are most at risk for progressive disease, have severe symptomatic nephrotic syndrome, or are at risk for or have complications of nephrotic syndrome. (See 'Which patients need treatment' below.)

●Clinical features – Clinical features associated with a higher risk of developing ESKD include nephrotic-range proteinuria (particularly if protein excretion exceeds 8 to 10 g/day) and an increased serum creatinine at presentation [4,9-12]. Although male sex and older age at onset have been associated with a higher risk of developing ESKD, this higher risk is mostly explained by the severity of disease (ie, higher proteinuria, lower glomerular filtration rate) [13,14].

Conversely, non-nephrotic-range proteinuria, a progressive decline in protein excretion, and presentation with normal kidney function have been associated with a relatively benign course [3,5,8,15-17]. In addition, patients of Japanese ancestry seem to have a better long-term prognosis than other ancestries [9].

●Biomarkers – Certain biomarkers have been associated with an increased risk of disease progression among patients with primary MN:

•Serum anti-phospholipase A2 receptor (PLA2R) antibody – Circulating autoantibodies to the M-type PLA2R are present in approximately 70 to 80 percent of patients with primary MN. Among patients with PLA2R-positive primary MN, a high level of anti-PLA2R antibody at the time of diagnosis has been associated in cross-sectional analysis with an increased risk for progressive loss of kidney function [18-21], a lower risk of spontaneous remission [18,22], and, among patients with non-nephrotic proteinuria, an increased risk for developing nephrotic proteinuria [23]. However, because the thresholds used to define a high titer varied in these studies, it remains unclear what specific antibody level should be used to predict the risk of disease progression. (See "Membranous nephropathy: Pathogenesis and etiology", section on 'Phospholipase A2 receptor'.)

Conversely, a low titer or absence of anti-PLA2R antibody is associated with a higher likelihood of spontaneous clinical remission [17,18,24,25]. Similarly, a reduction in anti-PLA2R antibody levels within the first six months after diagnosis, with or without treatment, is also commonly associated with clinical remission [17,26].

•Serum anti-thrombospondin type-1 domain-containing 7A (THSD7A) antibody – Circulating autoantibodies to THSD7A are present in approximately 3 percent of patients with primary MN [27,28]. Data on the prognostic role of anti-THSD7A antibody levels are limited [27,29]. The available data suggest that, similarly to anti-PLA2R antibodies, anti-THSD7A antibodies correlate with disease activity [28].

•Urine low-molecular-weight proteins – Urinary excretion of the low-molecular-weight proteins beta-2 microglobulin and alpha-1 microglobulin has been shown to correlate with disease progression in patients with primary MN [30,31]. The prognostic utility of these markers has not been directly compared with that of serum anti-PLA2R antibody levels. These markers have not been shown to predict the response to immunosuppressive therapy.

●Histologic findings – Histologic findings are frequently regarded as important predictors of outcome, as the risk of progression is increased in patients with glomerular scarring (segmental and/or global glomerulosclerosis) and correlates more closely with the severity of the tubulointerstitial disease than with the degree of glomerular injury [9,32,33]. This observation is typical of most glomerular disease. (See "Secondary factors and progression of chronic kidney disease", section on 'Tubulointerstitial fibrosis'.)

GENERAL MEASURES IN ALL PATIENTS — General supportive measures in all patients with MN include dietary sodium and protein restriction, blood pressure control, minimization of proteinuria with renin-angiotensin system inhibition, treatment of dyslipidemia, and, in select patients, anticoagulation. Sodium-glucose cotransporter 2 (SGLT2) inhibitors may also be of benefit, although formal studies in nephrotic patients are lacking. Other aspects of therapy include diuretics to control edema and maintenance of adequate nutrition [34]. This approach is consistent with the 2021 Kidney Disease: Improving Global Outcomes (KDIGO) Clinical Practice Guideline for the Management of Glomerular Diseases [35]. These issues are discussed in greater detail elsewhere:

●Dietary sodium and protein restriction (see "Dietary recommendations for patients with nondialysis chronic kidney disease", section on 'Salt intake' and "Dietary recommendations for patients with nondialysis chronic kidney disease", section on 'Protein intake')

●Antihypertensive therapy (see "Antihypertensive therapy and progression of nondiabetic chronic kidney disease in adults")

●Renin-angiotensin system inhibition (see "Antihypertensive therapy and progression of nondiabetic chronic kidney disease in adults", section on 'Renin-angiotensin system inhibitors')

●SGLT inhibitors (see "Overview of the management of chronic kidney disease in adults", section on 'Patients with proteinuria')

●Lipid lowering (See "Overview of the management of chronic kidney disease in adults", section on 'Dyslipidemia'.)

●Anticoagulation (see "Hypercoagulability in nephrotic syndrome", section on 'Patients with membranous nephropathy')

●Treatment of edema (see "Overview of the management of chronic kidney disease in adults", section on 'Volume overload')

INITIAL THERAPY FOR PRIMARY MN

Which patients need treatment — In patients with primary MN, our approach to initial therapy is based upon an assessment of the patient's risk of progressive disease, which we determine using both clinical and immunologic parameters (table 1) [36]:

●In patients with primary MN, regardless of phospholipase A2 receptor (PLA2R) status, we consider patients who present with two or more of the following features to be at very high risk for progressive disease:

•Serum creatinine ≥1.5 mg/dL [≥133 micromol/L] considered to be due to active MN

•Progressive decline in kidney function (eg, decrease in estimated glomerular filtration rate [eGFR] ≥25 percent from baseline over the prior two years) considered to be due to active MN

•Severe, disabling, or life-threatening nephrotic syndrome (defined by the presence of a serum albumin <2.5 g/dL [if measured by bromocresol green methods] or <2.0 g/dL [if measured by bromocresol purple methods] and refractory edema, or a thromboembolic event)

In such patients, we recommend treatment with immunosuppressive therapy, and treatment with immunosuppressive agents should not be delayed unless contraindicated. (See 'High or very high risk of progression' below.)

●In patients with primary MN who present with normal kidney function and do not have severe, disabling, or life-threatening nephrotic syndrome, we recommend treating with general supportive measures (including renin-angiotensin system inhibition) only for three to six months and monitoring for signs of disease progression (see 'General measures in all patients' above). During this observation period, we measure 24-hour urine protein excretion, serum creatinine (with determination of eGFR), and serum albumin at baseline and every one to three months depending upon the clinical severity at presentation. If the patient has PLA2R-associated primary MN (ie, serum anti-PLA2R antibody positive or positive immunostaining for PLA2R on kidney biopsy), we monitor serum anti-PLA2R antibody titers by enzyme-linked immunosorbent assay (ELISA) to evaluate the immunologic activity of the disease. We typically monitor antibody levels every two to three months and consider more frequent monitoring in patients with high antibody levels (arbitrarily defined as ≥150 RU/mL).

After three to six months of observation (or sooner if the patient has signs or symptoms of rapid disease progression), we stratify patients into the following categories of risk based upon an assessment of the temporal trends of these clinical and serologic parameters:

•High risk – Patients with two or more of the following are considered to be at high risk for progressive disease:

-Decrease in eGFR of ≥25 percent, not explained by other causes, at any time during the observation period

-Proteinuria >8 g/day at the end of the observation period or persistent nephrotic syndrome (proteinuria >3.5 g/day and serum albumin <3.5 g/dL [if measured by bromocresol green methods] or <3.0 g/dL [if measured by bromocresol purple methods])

-If the patient is anti-PLA2R antibody positive, serial anti-PLA2R antibody titers are high (arbitrarily defined as ≥150 RU/mL by the ELISA method) and not declining or are increasing to ≥150 RU/mL

Patients at high risk of progression should be treated with immunosuppressive therapy without further delay, unless contraindicated. (See 'High or very high risk of progression' below.)

•Moderate risk – Patients with two or more of the following are considered to be at moderate risk for progressive disease:

-Normal or stable (ie, <25 percent decrease) eGFR over the three-to-six-month period

-Proteinuria persistently between 4 and 8 g/day at the end of observation period

-If the patient is anti-PLA2R antibody positive, serial anti-PLA2R antibody titers are <150 RU/mL (by the ELISA method) and stable or increasing by <25 percent over the six-month period

Patients at moderate risk of progression may or may not require treatment with immunosuppressive therapy. Our approach to initial therapy varies depending upon the course of disease during the initial three to six-month observation period. (See 'Moderate risk of progression' below.)

•Low risk – Patients with two or more of the following are considered to be at low risk for progressive disease:

-Normal or stable eGFR (ie, <25 percent decrease) over the three-to-six-month period

-Proteinuria <4 g/day at the end of the observation period

-If the patient is anti-PLA2R antibody positive, serial anti-PLA2R antibody titers are low (arbitrarily defined as <50 RU/mL by the ELISA method) or decreasing by ≥25 percent at three to six months

Patients at low risk of progression do not require immunosuppressive therapy and can generally be managed with supportive measures only. (See 'Low risk of progression' below.)

In general, we do not give immunosuppressive therapy to patients if they have evidence of severe and irreversible kidney damage (ie, history of chronic kidney disease with a serum creatinine >3.5 mg/dL [309 micromol/L] or eGFR <30 mL/min/1.73 m² for >3 months; kidney size less than 8 cm on kidney ultrasound; or evidence of severe interstitial fibrosis, tubular atrophy, or glomerulosclerosis on kidney biopsy), since immunosuppressive therapy is not likely to be effective in such patients. In such patients, we continue to administer general supportive measures and discuss future options for kidney replacement therapy. We also do not give immunosuppressive therapy to patients with concomitant severe or potentially life-threatening infections. (See 'Patients with active infection' below.)

While most patients with primary MN will be properly classified into one of the above risk categories based upon their clinical and immunologic parameters, there may be some exceptions that are not perfectly represented by this stratification system. In such cases, clinical judgment is required to assess the risk of disease progression. In general, the three parameters (proteinuria, eGFR, and anti-PLA2R antibody levels) evolve in concert, with changes in anti-PLA2R antibody levels typically preceding changes in the clinical parameters. If this paradigm does not apply (eg, a decline in eGFR is accompanied by a decline in anti-PLA2R antibody titer), then alternative causes (eg, superimposed disease, hemodynamic effects of renin-angiotensin system inhibitors, renal vein thrombosis) should be excluded first.

This risk stratification applies only to the likelihood of progressive kidney failure. It does not take into account the risks that may be associated with other complications of the nephrotic syndrome, such as worsening atherosclerosis due to hypercholesterolemia and thromboemboli due to the associated hypercoagulable state. Thus, in addition to preservation of kidney function, immunosuppressive therapy may be beneficial by shortening the overall exposure to the nephrotic state, thereby minimizing the risk of other complications. (See 'General measures in all patients' above.)

The rationale for our approach is based in part upon studies showing the importance of persistent proteinuria and change in kidney function over a six-month observation period in predicting the probability of progression to chronic kidney disease [11,12]. Models incorporating these clinical variables have estimated the risk of chronic kidney disease (defined as creatinine clearance ≤60 mL/min/1.73 m²) over five years to be 75, 50, and less than 8 percent for patients defined as high, moderate, and low risk, respectively [12]. However, such models do not apply to algorithms that also incorporate anti-PLA2R antibody levels.

Clinical utility of anti-PLA2R antibody levels — The additive value of monitoring the trend of anti-phospholipase A2 receptor (PLA2R) antibody levels during the three-to-six-month observation period remains uncertain, and there are insufficient data to guide the use of anti-PLA2R antibody levels in making initial treatment decisions. However, in our practice, we find that serum anti-PLA2R antibody levels measured by ELISA are valuable to inform initial treatment decisions when used in combination with clinical and laboratory parameters. A clinical case followed by four different anti-PLA2R antibody scenarios illustrates the range of decision making based upon the opinions of the authors and editors of this topic:

●A 60-year-old patient with newly diagnosed MN and nephrotic syndrome presents with proteinuria of 7 g/day, serum creatinine of 1.0 mg/dL (88 micromol/L), and serum albumin of 2.5 g/dL (measured by bromocresol purple methods) or 3.0 g/dL (measured by bromocresol green methods). Mild to moderate peripheral edema is present. The patient receives general supportive measures (including renin-angiotensin system inhibition) over a six-month observation period. Over the course of six months, proteinuria does not change appreciably (from 7 to 5 to 8 g/day), serum creatinine remains relatively stable (around 1.0 to 1.1 mg/dL [88.4 to 97 micromol/L]), and serum albumin remains stable (around 2.5 to 2.6 g/dL).

•Settings where we would start immunosuppression:

-Anti-PLA2R antibody persistently high – At baseline, serum anti-PLA2R antibody level is 250 RU/mL. At three and six months, the anti-PLA2R antibody level is 275 and 250 RU/mL, respectively. In this scenario, we would initiate immunosuppressive therapy based upon the combination of persistent nephrotic syndrome and persistently high anti-PLA2R antibody levels.

-Anti-PLA2R antibody increasing – At baseline, serum anti-PLA2R antibody level is 70 RU/mL. At three and six months, the anti-PLA2R antibody level increases to 130 and 200 RU/mL, respectively. In this scenario, we would initiate immunosuppressive therapy based upon the combination of persistent nephrotic syndrome and the progressively rising anti-PLA2R antibody titers.

•Settings where we would treat expectantly:

-Anti-PLA2R antibody low and decreasing – At baseline, serum anti-PLA2R antibody level is 70 RU/mL. At three and six months, the anti-PLA2R antibody level decreases to 60 and 35 RU/mL, respectively. In this scenario, it would be reasonable to continue general supportive measures and defer starting immunosuppressive therapy. Although the patient has persistent nephrotic syndrome, kidney function remains stable, and progressively decreasing anti-PLA2R antibody levels suggest that the patient may achieve a spontaneous clinical remission.

-Anti-PLA2R antibody high but decreasing – At baseline, serum anti-PLA2R antibody level is 250 RU/mL. At three and six months, the anti-PLA2R antibody level is 130 RU/mL and 80 RU/mL, respectively. In this scenario, we would not initiate immunosuppressive therapy but continue close, monthly monitoring.

Choice of initial therapy based upon risk

High or very high risk of progression — In patients who are at high or very high risk of progression (see 'Which patients need treatment' above), we recommend treatment with immunosuppressive therapy and continued general supportive measures, rather than continued general supportive measures alone. As discussed above, patients at very high risk do not require a three-to-six-month observation period before immunosuppressive agents are initiated. (See 'General measures in all patients' above.)

The choice of immunosuppressive therapy depends upon how the patient is defined as high or very high risk:

●In patients who are classified as high or very high risk because of abnormal or declining kidney function at presentation or rapidly declining kidney function due to MN, we suggest combination treatment with glucocorticoids and a cytotoxic agent (preferably cyclophosphamide) rather than rituximab or other therapies. Such patients have a higher urgency for initiating treatment, and cytotoxic therapy appears to provide the best protection against progressive kidney disease. We prefer cyclophosphamide, rather than chlorambucil, because it is associated with fewer adverse effects. In patients who wish to avoid cytotoxic therapy, treatment with rituximab may be a reasonable alternative. (See 'Cytotoxic therapy plus glucocorticoids' below and 'Rituximab' below.)

In addition, patients with MN who experience a rapid decline in kidney function should be evaluated for other potential causes of worsening kidney function, such as crescentic glomerulonephritis (GN), acute hypersensitivity interstitial nephritis, or acute bilateral renal vein thrombosis.

●In patients who are classified as high or very high risk and have stable kidney function, we suggest treatment with rituximab rather than cytotoxic therapy or other therapies. Such patients have a lower urgency for initiating treatment than those with progressive kidney failure, and rituximab, which has less treatment-related toxicity, may be preferable. We prefer rituximab to a calcineurin inhibitor (CNI; cyclosporine or tacrolimus) because of the more prolonged remission with rituximab and higher relapse rate of CNIs after therapy has been discontinued. However, either a CNI or combination therapy with glucocorticoids plus a cytotoxic agent is a reasonable alternative to rituximab in patients who are anti-PLA2R antibody negative [37]. (See 'Rituximab' below and 'Calcineurin inhibitors' below and 'Cytotoxic therapy plus glucocorticoids' below.)

●We do not give glucocorticoids alone to high-risk patients, as this has not been shown to be effective.

The rationale for our approach in high-risk patients is based primarily upon the results of the Membranous Nephropathy Trial of Rituximab (MENTOR) and the United Kingdom Membranous Trial [37,38]. In the MENTOR trial, which included patients with primary MN who had heavy proteinuria but mostly preserved kidney function, treatment with rituximab was more effective than cyclosporine at maintaining complete or partial remission of proteinuria at 24 months [37]. The United Kingdom Membranous Trial, which specifically enrolled patients with primary MN who had deteriorating kidney function, found that the combination of glucocorticoids and cytotoxic therapy was superior to cyclosporine and to supportive therapy alone at preventing further loss of kidney function and end-stage kidney disease (ESKD) [38]. Additional data supporting our approach come from randomized trials in moderate-risk patients that found that cytotoxic therapy plus glucocorticoids prevented long-term progression to ESKD compared with supportive therapy alone (see 'Moderate risk of progression' below). There are no randomized trials that have directly compared rituximab with cytotoxic therapy in high-risk patients with primary MN.

●In the open-label, multicenter MENTOR trial, 130 patients with proteinuria ≥5 g/day and 24-hour creatinine clearance ≥40 mL/min/1.73 m² after at least three months of supportive care (angiotensin inhibition, blood pressure control, dietary sodium and protein restriction) were randomly assigned to rituximab (two infusions of 1 g administered 14 days apart; repeated at six months in case of partial response) or oral cyclosporine (3.5 mg/kg per day for six months [or 12 months in case of partial response at six months], with dose adjustment to maintain a 12-hour trough concentration of 125 to 175 ng/mL) [37]. At baseline, median proteinuria was 8.9 g/day and median serum albumin was 2.5 g/dL in both treatment arms; median creatinine clearance was 84.9 to 87.4 mL/min/1.73 m². Seventy-four percent of patients were anti-PLA2R antibody positive, and median anti-PLA2R antibody levels at baseline were similar between the two groups. The primary endpoint was a composite of complete (proteinuria <0.3 g/day and serum albumin ≥3.5 g/dL) or partial (reduction in proteinuria of ≥50 percent and final proteinuria between 0.3 and 3.5 g/day) remission at 24 months. The following results were noted:

•At 12 months, rates of complete or partial remission were similar between the two groups (60 and 52 percent for rituximab and cyclosporine, respectively). However, at 24 months, patients receiving rituximab had a higher rate of complete or partial remission than those receiving cyclosporine (60 versus 20 percent, respectively); only patients receiving rituximab achieved complete remission (35 versus 0 percent in the cyclosporine group).

•Among the patients who achieved complete or partial remission at 24 months, the mean creatinine clearance was higher in the rituximab group than in the cyclosporine group (96 versus 72 mL/min/1.73 m² at 12 months and 100 versus 87 mL/min/1.73 m² at 24 months, respectively).

•Among anti-PLA2R antibody-positive patients who achieved complete or partial remission at 24 months, those who received rituximab had a greater, more rapid, and more sustained decrease in anti-PLA2R antibody levels than those who received cyclosporine.

•Patients who did not respond to rituximab had higher anti-PLA2R antibody levels at baseline compared with those who achieved a partial or complete remission with rituximab.

•Among anti-PLA2R antibody-negative patients, rates of complete or partial remission at 12 months were comparable with rituximab or cyclosporine (67 versus 68 percent, respectively).

•Serious adverse events occurred in 17 percent of patients receiving rituximab and 31 percent of those receiving cyclosporine, although this difference was not statistically significant.

These findings show that rituximab is more effective than cyclosporine at maintaining clinical remission and may be better tolerated. However, patients in both study arms with high anti-PLA2R antibody titers at baseline may be less responsive to therapy, and it has been suggested, at least for rituximab, that patients might benefit from additional treatments [39] or an alternative treatment strategy to achieve remission. Since a placebo control group was not included in this trial, rates of remission with treatment cannot be formally compared with the rate of spontaneous remission. Longer-term follow-up data are needed to determine rates of relapse among rituximab-treated patients.

●The United Kingdom membranous trial randomly assigned 108 patients with a baseline serum creatinine less than 3.4 mg/dL (300 micromol/L) and at least a 20 percent decline in eGFR during the preceding two years to one of three therapies: supportive treatment plus alternating cycles of glucocorticoids and oral chlorambucil for six months, supportive treatment plus cyclosporine for 12 months, or supportive treatment alone [38]. Rates of the primary endpoint (an additional decline in eGFR of at least 20 percent) were lower with glucocorticoids/chlorambucil as compared with supportive treatment alone (58 versus 84 percent). In addition, cyclosporine did not reduce the incidence of kidney function loss as compared with supportive treatment (81 versus 84 percent). Fewer patients who received glucocorticoids/chlorambucil developed ESKD (one patient compared with four patients receiving supportive treatment and six patients receiving cyclosporine). The frequency of adverse events was similar in the glucocorticoid/chlorambucil and cyclosporine groups, although patients receiving glucocorticoids/chlorambucil had more serious adverse events than those receiving supportive treatment only (primarily hematological events such as leukopenia and anemia).

Although the United Kingdom Membranous Trial used chlorambucil, a cyclophosphamide-based regimen is thought to be equally effective and less toxic, as noted in a randomized head-to-head comparative trial that primarily enrolled moderate-risk patients (mean protein excretion 7 to 8 g/day and mean serum creatinine 1.05 mg/dL [93 micromol/L]) [40]. Chlorambucil also has more side effects; therefore, we prefer administering a cyclophosphamide-based regimen. (See "General toxicity of cyclophosphamide in rheumatic diseases".)

Moderate risk of progression — Our approach to initial therapy in patients at moderate risk of progression varies depending upon the course of disease during an initial three-to-six-month observation period with maximal general supportive measures (see 'Which patients need treatment' above and 'General measures in all patients' above):

●In moderate-risk patients who show a progressive increase in proteinuria over the observation period, we recommend treatment with immunosuppressive therapy and continued general supportive measures, rather than continued general supportive measures alone. (See 'General measures in all patients' above and 'Clinical utility of anti-PLA2R antibody levels' above.)

●In moderate-risk patients who show stable proteinuria over the observation period, we suggest immunosuppressive therapy and continued general supportive measures, rather than continued general supportive measures alone. However, some clinicians would continue to withhold immunosuppressive therapy beyond six months in such patients if they are doing well, especially if serum albumin is increasing, anti-PLA2R antibody levels (if initially positive) are low or decreasing, or if the patients are at high risk of having an adverse event with immunosuppressive therapy.

●In moderate-risk patients who show a progressive decline in proteinuria over the observation period, we withhold immunosuppression and continue general supportive measures.

Preferred first-line immunosuppressive therapies for moderate-risk patients with primary MN include rituximab, combination therapy with glucocorticoids plus a cytotoxic agent (preferably cyclophosphamide), or a CNI (cyclosporine or tacrolimus). The choice among these regimens depends upon several factors, including clinician and patient preference, drug availability and cost, efficacy, toxicity, and tolerability:

●In most moderate-risk patients who require immunosuppressive therapy, we suggest rituximab rather than glucocorticoids plus cytotoxic therapy or a CNI. (See 'Rituximab' below.)

●If rituximab is unavailable, either combination therapy with glucocorticoids plus a cytotoxic agent or CNI monotherapy is a reasonable alternative. Some experts prefer cytotoxic therapy in moderate-risk patients with evidence of disease progression (eg, decreasing eGFR, increasing proteinuria, or decreasing serum albumin), given the higher relapse rates and potential for nephrotoxicity with CNIs [41-43]. In patients with more stable disease, however, some experts prefer CNIs over cytotoxic therapy given their comparable short-term efficacy and better overall safety profile. If cytotoxic therapy is chosen, we prefer cyclophosphamide over chlorambucil since chlorambucil has more side effects. If a CNI is selected, either cyclosporine or tacrolimus is a reasonable choice. (See 'Cytotoxic therapy plus glucocorticoids' below and 'Calcineurin inhibitors' below.)

●We do not routinely use mycophenolate mofetil (MMF), natural adrenocorticotropic hormone (ACTH) gel, or synthetic ACTH as initial therapy in moderate-risk patients. However, such agents may be considered in patients who do not respond to all of the first-line therapies. (See 'Treatment of resistant disease' below.)

●We do not give glucocorticoids alone to moderate-risk patients, as this has not been shown to be effective.

Our preference for rituximab is based upon several studies demonstrating its efficacy and safety as initial therapy and as therapy for patients who have not responded to previous treatment with CNIs [37,44-54]. Although there have been no large, randomized trials directly comparing rituximab with cytotoxic therapy in patients with MN, we prefer rituximab to cytotoxic therapy based upon data from other trials and observational studies that suggest comparable short-term outcomes (60 to 70 percent remission rates) and a better safety profile with rituximab [37,44-54]. We prefer rituximab to a CNI based upon data from the MENTOR trial (which included patients who would be classified as moderate or high risk (see 'High or very high risk of progression' above)) that suggested that rituximab was more effective than cyclosporine at sustaining complete or partial remission of proteinuria beyond the 12 months of active treatment and had better patient adherence to therapy [37]. However, certain moderate-risk patients with higher serum anti-PLA2R antibody levels (eg, ≥150 to 200 RU/mL) [37,48] may require additional treatments with rituximab or an alternative regimen (such as combination therapy with glucocorticoids and cyclophosphamide) to achieve remission [39]. (See 'Rituximab' below.)

Several randomized trials have demonstrated the benefit of immunosuppressive therapy compared with supportive therapy alone in patients with primary MN who would be classified as having moderate risk for progression:

●In the Evaluate Rituximab Treatment for Idiopathic Membranous Nephropathy (GEMRITUX) trial, 75 patients with persistent proteinuria greater than 3.5 g/day after six months of treatment with angiotensin inhibition, diuretics, and a statin (general supportive measures) were randomly assigned to rituximab (two infusions of 375 mg/m² administered one week apart) or no rituximab [44]. General supportive measures were continued in all patients. At six months, there was no significant difference in the primary composite endpoint of complete or partial remission of proteinuria between patients treated with or without rituximab. Proteinuria, serum creatinine, and eGFR at six months were also similar between the two groups, but serum albumin levels were higher in those treated with rituximab (3.0 versus 2.4 g/dL). Anti-PLA2R antibodies, which were present in 73 percent of patients at baseline, disappeared in a greater proportion receiving rituximab (50 versus 12 percent). Serious adverse events were similar between the two groups.

The lack of benefit on proteinuria reduction from rituximab at six months may be attributed in part to the short duration to the preset primary endpoint. In a posttrial observational phase that followed patients for an additional 12 months, the rate of complete or partial remission was higher among patients treated with rituximab (65 versus 34 percent) [44]. In addition, patients treated with rituximab had less proteinuria (2195 versus 4701 mg/g) and higher serum albumin levels (3.2 versus 2.7 g/dL). These findings are consistent with observational studies that demonstrate a maximal reduction in proteinuria at 18 to 24 months after treatment with rituximab [45,55].

●Three randomized trials found that cyclophosphamide and chlorambucil, both given with glucocorticoids, are effective in inducing remission of proteinuria [5,6,40]. Two of the trials, mostly of patients with moderate disease (mean proteinuria 6 g/day in one), reported 10-year follow-up and showed that cytotoxic therapy also prevented progression to ESKD [5,6]. The patients receiving immunosuppressive therapy had significantly higher rates of complete or partial remission (88 versus 47 percent and 72 versus 34 percent with symptomatic therapy alone) and a higher rate of surviving without ESKD (92 versus 60 percent and 89 versus 65 percent).

Cyclophosphamide- and chlorambucil-based regimens are equally effective, as noted in a randomized head-to-head comparative trial that primarily enrolled moderate-risk patients [40]. However, since chlorambucil has more side effects, we prefer administering the cyclophosphamide-based regimen. Most studies evaluating cytotoxic therapy have included prednisone, and it is the clinical impression of many clinicians that cyclophosphamide alone is likely to be less beneficial.

●Both cyclosporine and tacrolimus have been shown in randomized trials to be effective in lowering proteinuria in moderate-risk patients with primary MN [56,57]. The best data are from a randomized trial of 51 patients (mean proteinuria 9.3 g/day and mean creatinine 1.2 mg/dL [106 micromol/L]) who were unresponsive to at least an eight-week course of daily glucocorticoid therapy [56]. The patients were assigned to prednisone (0.15 mg/kg up to a maximum dose of 15 mg/day) plus either placebo or cyclosporine (dose adjusted to maintain trough blood level of 125 to 225 microgram/L) for 26 weeks. The cyclosporine-treated group had a significantly higher rate of complete (≤300 mg/day) or partial (<3.5 g/day plus a ≥50 percent reduction from baseline) remission of proteinuria. Kidney function was the same in both groups. One year after the cessation of therapy, relapse of proteinuria was common, but 39 percent of treated patients were still in remission, compared with 13 percent with placebo.

The efficacy of tacrolimus (without glucocorticoids) was demonstrated in a randomized trial of 48 patients with MN who were treated with tacrolimus (0.05 mg/kg per day for 12 months with a six-month taper) or placebo [57]. The rate of complete or partial remission was significantly higher with tacrolimus at both 12 months (82 versus 24 percent) and 18 months (94 versus 35 percent). However, almost one-half of the patients who were in remission by the 18^th month developed a relapse of nephrotic syndrome after tacrolimus withdrawal.

There have been several trials and meta-analyses comparing the efficacy of different immunosuppressive regimens in moderate-risk patients with primary MN. Most trials have compared a CNI-based regimen with cytotoxic therapy and have found comparable rates of short-term remission but higher rates of relapse with CNIs after treatment withdrawal [41,58-61]. There are no large trials that have directly compared rituximab alone with cytotoxic therapy. One small pilot trial comparing rituximab with cyclical cyclophosphamide and glucocorticoids in moderate-risk patients suggested comparable rates of complete or partial remission at 24 months and similar rates of short-term adverse events [54]. Another randomized trial (Sequential Treatment with Tacrolimus and Rituximab versus Alternating Corticosteroids and Cyclophosphamide in PMN study [STARMEN]) that compared cyclical cytotoxic therapy with the combination of a CNI plus a single dose of rituximab at six months reported a higher rate of complete or partial remission of nephrotic syndrome at 24 months with cytotoxic therapy [43].

Low risk of progression — In patients at low risk of progression (see 'Which patients need treatment' above), we recommend continued observation with general supportive measures rather than administering immunosuppressive therapy. Such patients generally have excellent long-term prognosis and often undergo spontaneous partial or complete remission [4-6,8]. (See 'General measures in all patients' above.)

Low-risk patients should be monitored periodically for disease progression that might warrant therapy. We suggest clinical assessment, measurement of serum creatinine and 24-hour urine protein excretion, and measurement of anti-PLA2R antibody levels every three months for two years and twice yearly thereafter since the risk of developing progressive disease falls significantly after two years. The rationale for prolonged monitoring for disease progression was provided by a review of 395 patients with MN, 108 of whom (27 percent) presented with subnephrotic proteinuria [8]. Among these patients, 66 (61 percent) progressed to nephrotic syndrome; of these, 46 patients (70 percent) progressed within the first year, 11 (18 percent) progressed between years 1 to 4, and 9 patients (13 percent) progressed after year 4 (19 percent). (See 'Monitoring and modifying therapy' below.)

Dosing, duration, and toxicity of first-line therapies

Rituximab — Rituximab is an anti-CD20 monoclonal antibody that depletes CD20-positive B cells. Rituximab is the preferred first-line immunosuppressive therapy for high- or very-high-risk patients who have normal or near-normal kidney function and moderate-risk patients. (See 'High or very high risk of progression' above and 'Moderate risk of progression' above.)

The optimal dosing regimen for rituximab is uncertain. We administer rituximab using the dose that was given in the MENTOR trial [37], specifically 1 g initially followed 14 days later by another 1 g dose. Some experts use an alternative regimen, administering 375 mg/m² weekly for four weeks or a B cell-driven approach [50] in which a second dose of 375 mg/m² is given if ≥5 circulating B cells/microL are detected by flow cytometry one week after treatment. The rationale for two additional doses is to induce a more profound B cell depletion. However, there are no controlled trials that have compared the efficacy of the various rituximab regimens.

In patients treated with an initial dose of rituximab of 2 g, we do not routinely check peripheral B cell counts. However, in centers that are used to administering lower doses (eg, 375 mg/m², one or two doses), we suggest checking peripheral blood B lymphocyte counts (by monitoring CD19-positive cells) one week after the last rituximab dose to ascertain B cell depletion. We administer an additional dose of rituximab (1 g) if B cell depletion is not complete.

Rituximab is generally well tolerated. Short-term adverse events include infusion reactions, serum sickness, hypogammaglobulinemia, opportunistic infections, and, rarely, anaphylaxis. The adverse effects of rituximab are discussed in more detail elsewhere:

●(See "Rituximab: Principles of use and adverse effects in rheumatoid arthritis".)

●(See "Infusion-related reactions to therapeutic monoclonal antibodies used for cancer therapy", section on 'Rituximab'.)

Cytotoxic therapy plus glucocorticoids — Combination therapy with a cytotoxic agent (cyclophosphamide or chlorambucil) and glucocorticoids is the preferred initial therapy for patients classified as high or very high risk because of deteriorating kidney function due to MN and for whom immunosuppressive therapy is felt to be appropriate. In addition, cytotoxic therapy can be used as initial therapy in patients at moderate risk of progression if rituximab is unavailable. (See 'High or very high risk of progression' above and 'Moderate risk of progression' above.)

We prefer cyclophosphamide since chlorambucil has more side effects. An acceptable regimen is the "modified Ponticelli" regimen [6,40]:

●During months 1, 3, and 5, treat with intravenous (IV) methylprednisolone (1 g/day for three days), followed by oral prednisone (0.5 mg/kg per day for 28 days). At the end of months 1, 3, and 5, oral prednisone can generally be discontinued without a taper.

●During months 2, 4, and 6, treat with oral cyclophosphamide (2 mg/kg per day, maximum daily dose of 200 mg). We typically reduce the dose by 25 percent in older adults (age >75 years) and adjust the dose appropriately in patients with impaired kidney function (table 2). We titrate the dose to achieve a peripheral blood lymphocyte count of <700 cells/microL and a white blood cell count >3500 cells/microL.

Some centers may prefer daily (rather than cyclical) cyclophosphamide at a dose of 1.5 mg/kg per day combined with daily oral glucocorticoids (prednisone or equivalent 0.25 mg/kg per day or 0.5 mg/kg every other day). In patients receiving daily cyclophosphamide, we limit the maximum cumulative dose to 25 g, given the increased risk of malignancy with longer duration of therapy. There are insufficient data to advocate the use of IV cyclophosphamide. Upon completion of six months of cyclophosphamide and glucocorticoids, we taper the glucocorticoids over a period of eight weeks.

In patients who are planning to conceive, we suggest avoidance of cyclophosphamide. If cyclophosphamide is the only option, we suggest a maximum cumulative dose of 10 g. Patients must be clearly informed of the significant risk for infertility and should be referred to a fertility clinic to discuss egg/sperm harvesting. (See "General toxicity of cyclophosphamide in rheumatic diseases", section on 'Malignancy' and "General toxicity of cyclophosphamide in rheumatic diseases", section on 'Gonadal toxicity'.)

The use of cyclophosphamide and glucocorticoids is associated with significant toxicity. Earlier trials may have underreported rates of serious adverse events associated with cyclical immunosuppression therapies [5,6,62]. A subsequent study clearly demonstrated the significant toxicity associated with daily glucocorticoid/cyclophosphamide regimens, even among patients treated with lower doses (1.5 mg/kg per day of cyclophosphamide) [63]. A more detailed discussion of the toxicities of cyclophosphamide and glucocorticoids is presented elsewhere:

●(See "General toxicity of cyclophosphamide in rheumatic diseases".)

●(See "Major adverse effects of systemic glucocorticoids".)

Calcineurin inhibitors — Both cyclosporine and tacrolimus have proven short-term efficacy in patients with primary MN and can be used as immunosuppressive therapy in moderate-risk patients if rituximab is unavailable and the patient wishes to avoid cytotoxic therapy. (See 'Moderate risk of progression' above.)

●If a cyclosporine-based approach is chosen, the preferred regimen is treatment for at least six months at a dose of 3 to 5 mg/kg per day in two divided doses to maintain whole blood trough levels of 120 to 200 ng/mL; some clinicians would also initiate therapy with prednisone given every other day (maximum 10 mg every other day).

●If a tacrolimus-based approach is chosen, the preferred regimen is 0.05 to 0.1 mg/kg per day for at least six months in two divided doses to maintain whole blood trough levels between 3 and 5 ng/mL. The dose may be increased to achieve a higher trough level between 5 and 8 ng/mL if there is no reduction in proteinuria by two months, provided that kidney function has not worsened.

Among patients who attain at least a partial remission (protein excretion <3.5 g/day plus a ≥50 percent reduction from baseline), we continue the dose of the CNI for an additional six months. In patients on cyclosporine, we then reduce the dose to 1.5 to 2.5 mg/kg per day to minimize the risk of nephrotoxicity and administer this dose for at least one to two years. In patients on tacrolimus, we reduce the dose by 25 to 50 percent over the same time frame as with cyclosporine. The duration of therapy is based upon an ongoing assessment of the benefits and risks. Some clinicians may choose to withdraw therapy earlier and more rapidly in patients who achieve a disappearance of anti-PLA2R antibodies or a complete clinical remission (protein excretion ≤300 mg/day). Conversely, some clinicians may choose to continue the CNI indefinitely if achieving remission was difficult, provided that there are no signs of nephrotoxicity. Relapse of proteinuria can occur after the CNI is discontinued. (See 'Treatment of relapsing disease' below.)

In general, patients who do not respond to one of these CNIs are unlikely to respond to the other.

The adverse effects of cyclosporine and tacrolimus, including their potential nephrotoxicity, are discussed in detail elsewhere:

●(See "Pharmacology of cyclosporine and tacrolimus".)

●(See "Cyclosporine and tacrolimus nephrotoxicity".)

Monitoring and modifying therapy — All patients who are receiving treatment for primary MN should be closely monitored for clinical response to therapy. In patients with PLA2R-associated primary MN, serial assessment of anti-PLA2R antibody levels may be useful for monitoring the immunologic activity of disease and guiding treatment decisions.

Goals of therapy — In patients with primary MN who are receiving immunosuppressive therapy, the goals of therapy are to induce a clinical remission and, in those who have positive serum anti-PLA2R antibodies, to achieve an immunologic remission. We use the following definitions of clinical and immunologic response:

●Clinical response – Clinical response in patients with MN is reported in terms of degree of reduction in proteinuria. We use the following definitions:

•A complete clinical remission is a reduction in proteinuria to ≤300 mg/day

•A partial clinical remission in patients presenting with nephrotic-range proteinuria is a reduction in proteinuria of ≥50 percent from baseline and to >300 mg/day and <3.5 g/day

•No clinical remission is a less than 50 percent reduction in proteinuria from baseline or proteinuria ≥3.5 g/day

Most patients with primary MN who successfully achieve a clinical remission will demonstrate a reduction in proteinuria within 6 to 24 months of immunosuppressive therapy.

●Immunologic response – In patients with PLA2R-associated primary MN, serial assessment of serum anti-PLA2R antibody levels can be used to monitor the immunologic activity of the disease. Immunologic remission is defined as depletion of anti-PLA2R antibodies, as evidenced by anti-PLA2R antibody titers below the cut-off value for a positive result (<14 RU/mL by ELISA, although some experts prefer to use a lower cut-off of <2 RU/mL to define true absence of anti-PLA2R antibody) and/or a negative indirect immunofluorescence test.

A decline in anti-PLA2R antibody levels may predict the clinical response to treatment. Several studies have shown that a decrease in anti-PLA2R antibody levels consistently precedes a decrease in proteinuria (figure 1) [23,26,48,64-68]. Among patients who achieve an immunologic remission, a remission in proteinuria usually follows within 12 to 24 months. Clinicians should be aware of this characteristic delay between immunologic and clinical remission and should not necessarily modify the treatment plan if the patient has achieved an immunologic remission but has not yet experienced a significant reduction in proteinuria.

Some [2,37,67,68], but not all [23,26,64], studies have shown that the rate of reduction in anti-PLA2R antibody levels may depend upon the type of immunosuppressive therapy given.

Anti-PLA2R antibody levels at the completion of therapy may predict long-term treatment outcomes. In one study, patients with undetectable antibody levels after treatment were more likely to remain in remission at five years than those with detectable antibodies (58 versus 0 percent, respectively) [67]. (See 'Prognosis' below.)

A re-emergence of or an increase in anti-PLA2R antibody levels in a patient who previously achieved clinical remission may predict clinical relapse [48]. (See 'Treatment of relapsing disease' below.)

Non-PLA2R-associated MN — In patients with non-PLA2R-associated primary MN (see 'Terminology' above), we perform a clinical assessment and measure serum creatinine, serum albumin, and 24-hour urine protein excretion every two to three months. Based upon these parameters, we determine the clinical response to treatment after six months of therapy and modify treatment and monitoring as follows:

●Patients with complete clinical remission – In patients who achieve a complete clinical remission (see 'Goals of therapy' above) at six months, we reduce or discontinue therapy depending upon the agent(s) being used. If the patient was treated with rituximab, we do not administer any additional doses of rituximab. If the patient was treated with cyclophosphamide plus glucocorticoids, we discontinue these agents. If the patient was treated with a CNI, we start tapering and withdraw the agent after three to six months. In all patients, we continue general supportive measures until complete clinical remission has been maintained for 6 to 12 months (see 'General measures in all patients' above). We reduce the frequency of laboratory monitoring to every three to six months for the first year after remission and then every 6 to 12 months thereafter. The frequency of monitoring should increase if the patient shows signs of increasing proteinuria, decreasing serum albumin, or increasing serum creatinine; such patients should be evaluated for disease relapse. (See 'Treatment of relapsing disease' below.)

●Patients without complete clinical remission – In patients who do not achieve a complete clinical remission at six months, we take the following approach:

•Patients who have unchanged or increasing proteinuria with no increase in serum albumin or who have a ≥25 percent reduction in eGFR are considered to have an unsatisfactory initial response to therapy. In such patients, we modify the treatment regimen depending upon the agent(s) being used. If the patient was treated with rituximab, we would switch to combination therapy with cyclophosphamide plus glucocorticoids. If the patient was treated with cyclophosphamide plus glucocorticoids, we add rituximab (two infusions of 1 g administered 14 days apart). If the patient was treated with a CNI, we switch to rituximab or cyclophosphamide plus glucocorticoids.

•Patients who have decreasing proteinuria (≥25 percent reduction) and stable (<25 percent reduction) eGFR are considered to have a satisfactory initial response to therapy. However, because it remains uncertain if such patients will achieve a partial or complete clinical remission, we continue or modify immunosuppressive therapy depending upon the agent(s) being used. If the patient was treated with rituximab, we give an additional course of rituximab (two infusions of 1 g administered 14 days apart) [37]. If the patient was treated with cyclophosphamide plus glucocorticoids, we discontinue these agents to reduce the risk of toxicity and continue monitoring since it can take at least 12 to 18 months to achieve a clinical remission. If the patient was treated with a CNI, we continue the same dose for another six months after the proteinuria nadir has been achieved and then begin to reduce the dose, as described elsewhere in this topic. (See 'Calcineurin inhibitors' above.)

•We continue general supportive measures until complete remission has been achieved and maintained for 6 to 12 months (see 'General measures in all patients' above). We continue to perform a clinical assessment and laboratory monitoring every two to three months and reevaluate the patient's response after six months (ie, a total of 12 months of therapy).

Patients who do not achieve a reduction in proteinuria after treatment with rituximab-, cyclophosphamide-, or CNI-based regimens are considered to have resistant disease. The management of such patients is discussed below. (See 'Treatment of resistant disease' below.)

PLA2R-associated MN — In patients with PLA2R-associated primary MN (see 'Terminology' above) who are receiving treatment, we perform a clinical assessment and measure serum creatinine, serum albumin, 24-hour urine protein excretion, and serum anti-PLA2R antibody levels every two to three months. Based upon these parameters, we determine both the clinical and immunologic response to treatment after six months of therapy and modify treatment and monitoring as follows (algorithm 1):

●Patients with immunologic remission – In patients who achieve an immunologic remission (see 'Goals of therapy' above) at six months, we reduce or discontinue therapy depending upon the agent(s) being used. If the patient was treated with rituximab, we do not administer any additional doses of rituximab. If the patient was treated with cyclophosphamide plus glucocorticoids, we discontinue these agents [69]. If the patient was treated with a CNI, we start tapering and withdraw the agent after three to six months. In all patients, we continue general supportive measures until complete clinical remission has been achieved and maintained for 6 to 12 months (see 'General measures in all patients' above). We reduce the frequency of laboratory monitoring to every three to six months for the first year after immunologic remission and then every 6 to 12 months thereafter. The frequency of monitoring should increase if the patient shows signs of increasing proteinuria, decreasing serum albumin, increasing serum creatinine, or increasing anti-PLA2R antibody levels; such patients should be evaluated for disease relapse. (See 'Treatment of relapsing disease' below.)

In most patients who achieve an immunologic remission, a clinical remission in proteinuria usually follows within 12 to 24 months. Some patients who achieve an immunologic remission may not achieve a complete clinical remission within this time frame. This can be caused by incomplete remodeling of the glomerular filtration barrier (in which case proteinuria may eventually decrease further) or due to irreversible chronic damage (in which case some degree of proteinuria may persist), and additional immunosuppressive therapy is not warranted in these patients. Patients who achieve an immunologic remission at six months but have increasing proteinuria or decreasing eGFR are uncommon and should be evaluated for other causes of worsening kidney function and/or worsening proteinuria (eg, renal vein thrombosis). In such patients, serum anti-PLA2R antibody testing should also be repeated to confirm that immunologic remission has been maintained.

●Patients without immunologic remission – In patients who do not achieve an immunologic remission at six months, we take the following approach:

•Patients who have unchanged or increasing serum anti-PLA2R antibody levels, unchanged or increasing proteinuria (with no increase in serum albumin), or a ≥25 percent reduction in eGFR are considered to have an unsatisfactory initial response to therapy. In such patients, we modify the treatment regimen depending upon the agent(s) being used. If the patient was treated with rituximab, we would switch to combination therapy with cyclophosphamide plus glucocorticoids. If the patient was treated with cyclophosphamide plus glucocorticoids, we add rituximab (two infusions of 1 g administered 14 days apart). If the patient was treated with a CNI, we switch to rituximab or cyclophosphamide plus glucocorticoids.

•Patients who have decreasing anti-PLA2R antibody levels, decreasing proteinuria, and stable (<25 percent reduction) eGFR are considered to have a satisfactory initial response to therapy. However, because it remains uncertain if such patients will achieve a clinical remission, we continue or modify immunosuppressive therapy depending upon the agent(s) being used. If the patient was treated with rituximab, we give an additional course of rituximab (two infusions of 1 g administered 14 days apart) [37]. If the patient was treated with cyclophosphamide plus glucocorticoids, we discontinue these agents to reduce the risk of toxicity and continue monitoring since it can take at least 12 to 18 months to achieve a clinical remission. Some authors would switch to mycophenolate mofetil (1000 mg twice daily) for up to six months [69]. If the patient was treated with a CNI, we switch to rituximab; if rituximab is not available, we treat with cyclophosphamide plus glucocorticoids.

•We continue general supportive measures until complete remission has been achieved and maintained for 6 to 12 months (see 'General measures in all patients' above). We continue to perform a clinical assessment and laboratory monitoring every two to three months and reevaluate the patient's response after six months (ie, a total of 12 months of therapy).

Patients who do not achieve a clinical or immunologic remission after treatment with rituximab-, cyclophosphamide-, and CNI-based regimens are considered to have resistant disease. The management of such patients is discussed below. (See 'Treatment of resistant disease' below.)

TREATMENT OF RELAPSING DISEASE — In patients with primary MN, a relapse is a return of proteinuria to ≥3.5 g/day after an initial complete or partial response to immunosuppressive therapy. Patients with a relapse after an initial partial remission should also have a ≥50 percent increase in proteinuria from the proteinuria nadir in addition to proteinuria ≥3.5 g/day. Approximately one-third of patients who respond to immunosuppressive therapy eventually develop a relapse [6,40,70-72], which may adversely affect their long-term prognosis [73,74]. Relapse rates are higher in patients who achieve partial remission than in patients who achieve complete remission (approximately 50 versus 25 percent) [75,76]. Among patients with partial remission, the risk of relapse is higher in those who remain hypoalbuminemic compared with those who have normal serum albumin levels.

Relapses are mostly caused by activity of the MN, as is illustrated by the reappearance of anti-phospholipase A2 receptor (PLA2R) antibodies in patients with PLA2R-associated MN. In rare instances, a repeat kidney biopsy may be needed to ascertain the cause of relapsing proteinuria.

Relapsing disease may require repeat treatment with immunosuppressive therapy. However, some patients with relapsing disease may develop a spontaneous remission without additional treatment, even if they previously received immunosuppressive therapy for their initial presentation. Thus, in patients who develop a relapse, we suggest a six-month observation period with general supportive measures to reassess the risk for progressive disease, using the same approach as that for newly diagnosed patients with primary MN. (See 'Which patients need treatment' above.)

In patients who require immunosuppressive therapy, the choice of agent for relapse depends upon the initial immunosuppressive regimen:

●In patients who were initially treated with cytotoxic therapy plus glucocorticoids, we treat with either rituximab or a calcineurin inhibitor (CNI). Given the significant toxicity associated with cytotoxic therapy and glucocorticoids, we avoid using this regimen for relapse unless the patient cannot receive rituximab or a CNI. (See 'Rituximab' above and 'Calcineurin inhibitors' above.)

●In patients who were initially treated with rituximab or a CNI and have stable kidney function, we typically retreat with the same agent and regimen that was used to achieve the prior clinical response. In those who are experiencing a deterioration in kidney function, we prefer to use cytotoxic therapy (cyclophosphamide) plus glucocorticoids. (See 'Rituximab' above and 'Calcineurin inhibitors' above and 'Cytotoxic therapy plus glucocorticoids' above.)

●In patients who do not tolerate or cannot receive rituximab, a CNI, or cytotoxic therapy plus glucocorticoids, other treatment options such as mycophenolate mofetil (MMF) or adrenocorticotropic hormone (ACTH) may be considered. These therapies are discussed elsewhere in this topic. (See 'Treatment of resistant disease' below.)

Patients receiving treatment for relapsing disease should be monitored using the same approach as those receiving initial therapy. (See 'Monitoring and modifying therapy' above.)

The treatment of relapse in patients with primary MN has not been evaluated in controlled trials.

TREATMENT OF RESISTANT DISEASE — Patients with primary MN who do not respond to initial therapy with rituximab, cyclophosphamide plus glucocorticoids, or a calcineurin inhibitor (CNI) should be given an alternative regimen of one of these first-line therapies, as discussed elsewhere in this topic. (See 'Monitoring and modifying therapy' above.)

Patients who do not achieve a complete or partial remission after all of the first-line therapies have been tried are considered to have resistant disease. Among patients who are anti-phospholipase A2 receptor (PLA2R) antibody positive, persistence of serum anti-PLA2R antibodies in spite of immunosuppressive therapy is also indicative of resistant disease. The optimal approach to patients with resistant disease is not known, and in general, such patients should be referred to centers that have clinical experience in the treatment of patients with MN.

●Mycophenolate mofetil (MMF) [77-82], natural adrenocorticotropic hormone (ACTH) gel [62,83-87], and plasma exchange [88] are reasonable therapeutic options, although high-quality data demonstrating their efficacy are lacking.

●Patients who are resistant to rituximab may respond to obinutuzumab, a more potent B cell–depleting antibody [89,90].

●Plasma cell–directed therapies (such as bortezomib and daratumumab) have been used with variable efficacy in case reports [91,92].

In patients who are refractory to multiple immunosuppressive regimens, it may be difficult to distinguish between resistant disease and chronic, irreversible damage to the glomerular filtration barrier or the development of glomerulosclerosis. In such patients, a repeat kidney biopsy may be helpful to inform therapeutic decisions. Patients who are found on biopsy to have fresh immune deposits may potentially benefit from additional immunosuppressive therapy, whereas those with evidence of widespread, chronic histologic damage and no fresh immune deposits are unlikely to benefit from further immunosuppressive therapy.

SPECIAL POPULATIONS

Patients with active infection — In general, patients with primary MN who have an active infection should not be treated with immunosuppressive therapy. If the patient warrants immunosuppressive therapy but has not yet started treatment, we typically wait until the infection has resolved before initiating therapy. (See 'High or very high risk of progression' above.)

In patients with primary MN who develop an active infection while receiving immunosuppressive therapy, we temporarily withhold all immunosuppressive agents. We usually restart therapy when the patient has fully recovered from the infection.

Patients with crescentic GN — MN is infrequently associated with a crescentic glomerulonephritis (GN) that in some patients may be related to antineutrophil cytoplasmic antibodies (ANCA) or, less often, anti-glomerular basement membrane (GBM) antibodies. (See "Membranous nephropathy: Pathogenesis and etiology", section on 'Crescentic glomerulonephritis'.)

There is no high-quality evidence to guide the optimal therapeutic approach to patients with MN associated with a crescentic GN. Our approach in such patients is based on limited data from observational studies and our clinical experience:

●In patients with MN associated with crescentic GN who have evidence of ANCA or anti-GBM antibodies, we treat with immunosuppressive therapy using a similar approach as that for patients with crescentic GN associated with ANCA vasculitis or anti-GBM disease, respectively [93-98]. Treatment for these disorders is discussed in more detail elsewhere:

•(See "Granulomatosis with polyangiitis and microscopic polyangiitis: Induction and maintenance therapy".)

•(See "Anti-GBM (Goodpasture) disease: Treatment and prognosis", section on 'Initial therapy'.)

●In patients with MN associated with crescentic GN who do not have evidence of ANCA or anti-GBM antibodies, we treat with immunosuppressive therapy given the generally poor prognosis of these patients [99]. We administer rituximab or combination therapy with glucocorticoids plus cyclophosphamide using a regimen similar to that used for initial therapy in patients with ANCA vasculitis. However, there are no high-quality data to support this or any other approach. (See "Granulomatosis with polyangiitis and microscopic polyangiitis: Induction and maintenance therapy".)

Patients with primary MN and light chain-restricted deposits — Rare cases of primary MN with light chain-restricted deposits have been reported. (See "Membranous nephropathy: Pathogenesis and etiology", section on 'MN with light chain–restricted deposits'.)

The optimal management of patients with primary MN with light chain-restricted deposits is not known. In such patients, an evaluation for an underlying lymphoproliferative disorder should be performed, and, if a hematologic malignancy is diagnosed, appropriate treatment should be administered. Patients who have evidence of a pathologic B cell or plasma cell clone but do not meet criteria for a hematologic malignancy are considered to have monoclonal gammopathy of renal significance (MGRS). The treatment of patients with MGRS is discussed in detail elsewhere. (See "Diagnosis and treatment of monoclonal gammopathy of renal significance", section on 'Overview of our approach'.)

Patients who are pregnant — Rare case reports of primary MN occurring in pregnant women have been described [100,101].

The optimal approach to the treatment of MN in pregnant patients is not known. As in nonpregnant patients with primary MN, we suggest an initial assessment to determine the patient's risk of disease progression (see 'Which patients need treatment' above). However, in pregnant patients with nephrotic syndrome, most authors of this topic would start immunosuppressive therapy. We prefer a CNI (cyclosporine or tacrolimus) in those with normal or near-normal kidney function. In general, rituximab and cyclophosphamide should be avoided in pregnant patients. (See "Safety of rheumatic disease medication use during pregnancy and lactation".)

Given the rarity of this clinical scenario, we suggest that pregnant patients with MN be referred to centers with expertise in the management of glomerular disease in pregnancy. Pregnant women with kidney disease should be jointly managed by a nephrologist and by a maternal-fetal medicine specialist. (See "Pregnancy and contraception in patients with nondialysis chronic kidney disease", section on 'Management during pregnancy'.)

A detailed discussion of pregnancy in women with underlying kidney disease and the general management of nephrotic syndrome in pregnancy are presented separately:

●(See "Pregnancy and contraception in patients with nondialysis chronic kidney disease".)

●(See "Proteinuria in pregnancy: Diagnosis, differential diagnosis, and management of nephrotic syndrome", section on 'Management of nephrotic syndrome in pregnancy'.)

TREATMENT OF SECONDARY MN — Secondary MN has been attributed to a variety of agents and conditions (table 3). In general, the treatment of secondary MN focuses on cessation of the offending drug or effective treatment of the underlying disease, which is usually associated with improvement in the nephrotic syndrome. (See "Membranous nephropathy: Pathogenesis and etiology", section on 'Secondary MN'.)

Some patients with secondary causes of MN may test positive for anti-phospholipase A2 receptor (PLA2R) antibodies. In such patients, the possibility of two separate disease entities, primary PLA2R-associated MN and a coincidental secondary disease (eg, hepatitis B virus [HBV] infection, malignancy), must be considered [102]. If the patient develops signs of progressively worsening MN (eg, declining kidney function, increasing proteinuria), we would administer immunosuppressive therapy, especially if the underlying secondary disease has already been successfully treated. Careful monitoring for recurrence or reactivation of the secondary disease should be performed. (See 'Initial therapy for primary MN' above.)

MN is the most common cause of the nephrotic syndrome that occurs following hematopoietic cell transplantation (HCT). Most patients are negative for anti-PLA2R antibodies; however rare cases of PLA2R-positive MN after HCT have been reported [103]. The treatment of MN after HCT is discussed in more detail elsewhere. (See "Kidney disease following hematopoietic cell transplantation", section on 'Membranous nephropathy'.)

PROGNOSIS — Patients with primary MN who undergo spontaneous or drug-induced remission generally have a good long-term prognosis. Thirty to 40 percent of patients who receive cytotoxic therapy plus glucocorticoids undergo complete remission of proteinuria, 30 to 50 percent undergo partial remission, and the incidence of progressive kidney function impairment is only approximately 10 percent [5,6,40].

The prognosis after complete remission (whether treatment associated or spontaneous) was reported in 82 adults with biopsy-confirmed primary MN [70]. The mean total observation time was 101 months, 69 months of which was in the postremission period. Three distinct groups of patients were identified:

●Those who remained in remission (67 percent)

●Those with a remitting, relapsing course but without kidney function impairment (20 percent)

●Those who developed kidney function impairment (13 percent)

No patient subsequently developed end-stage kidney disease (ESKD). Multivariate analysis revealed that female sex and persistently low levels of proteinuria were associated with a durable remission. These parameters are also associated with a better long-term prognosis [13]. By comparison, progressive disease is characterized by persistent heavy proteinuria and a gradual elevation in the serum creatinine concentration over a period of years. As noted above, however, effective therapy can improve outcomes in such patients. (See 'High or very high risk of progression' above.)

Patients who attain partial remission have a much better outcome than those with stable or progressive disease. This was illustrated in an observational study that included 136 patients with primary MN who had partial remission and 110 who had no remission [75]. At a median of five years, partial remission was independently associated with a slower decrease in kidney function over time (-0.17 versus -0.86 mL/min per month with no remission) and a lower incidence of kidney failure (9 versus 29 percent, adjusted hazard ratio [HR] 0.17).

The durability of remission (spontaneous or treatment associated) is associated with improved kidney survival. One prospective study examined this relationship among 376 patients with primary MN who achieved complete or partial remission after a period of nephrotic-range proteinuria [74]. Persistent remission at 3, 6, 12, and 24 months after remission onset, compared with disease relapse at each of these time points, was associated with a significantly lower risk of the primary endpoint (ESKD or 50 percent reduction in estimated glomerular filtration rate [eGFR], HRs 0.48, 0.35, 0.37, and 0.40, respectively).

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Glomerular disease in adults".)

SUMMARY AND RECOMMENDATIONS

●General measures – General supportive measures in all patients with membranous nephropathy (MN) include blood pressure control, minimization of proteinuria with renin-angiotensin system inhibition, treatment of dyslipidemia, and, in select patients, anticoagulation. Other aspects of therapy include dietary sodium restriction, diuretics to control edema, and maintenance of adequate nutrition. (See 'General measures in all patients' above.)

●Risk assessment – Since many patients with mild to moderate disease undergo spontaneous remission (figure 2) and immunosuppressive agents have appreciable toxicity, most experts agree that treatment with immunosuppressive therapy should be reserved for those who are at highest risk for developing progressive kidney failure. In patients with primary MN, our approach to initial therapy is based upon an initial assessment of the patient's risk of progressive disease (eg, very high, high, moderate, or low), which we determine using both clinical and immunologic parameters (table 1). (See 'Which patients need treatment' above.)

●High or very high risk of progression – In patients who are at high or very high risk of progression, we recommend treatment with immunosuppressive therapy and continued general supportive measures, rather than continued general supportive measures alone (Grade 1B). The choice of immunosuppressive therapy depends upon how the patient is defined as high or very high risk (see 'High or very high risk of progression' above):

•In patients who are classified as high or very high risk and have stable kidney function, we suggest treatment with rituximab rather than cytotoxic therapy or other therapies (Grade 2C). However, a calcineurin inhibitor (CNI) is a reasonable alternative to rituximab in patients who are anti-phospholipase A2 receptor (PLA2R) antibody negative. (See 'Rituximab' above and 'Cytotoxic therapy plus glucocorticoids' above and 'Calcineurin inhibitors' above.)

•In patients who are classified as high or very high risk because of abnormal or declining kidney function at presentation or rapidly declining kidney function due to MN, we suggest combination treatment with glucocorticoids and a cytotoxic agent (preferably cyclophosphamide) rather than rituximab or other therapies (Grade 2C). In patients who wish to avoid cytotoxic therapy, treatment with rituximab may be a reasonable alternative.

●Moderate risk of progression – In patients who are at moderate risk of progression, our approach to initial therapy varies depending upon the course of disease during an initial three-to-six-month observation period with maximal general supportive measures (including renin-angiotensin system inhibition) (see 'Moderate risk of progression' above):

•In moderate-risk patients who show a progressive increase in proteinuria over the observation period, we recommend treatment with immunosuppressive therapy and continued general supportive measures, rather than continued general supportive measures alone (Grade 1B).

•In moderate-risk patients who show stable proteinuria over the observation period, we suggest immunosuppressive therapy and continued general supportive measures, rather than continued general supportive measures alone (Grade 2C). However, some clinicians would continue to withhold immunosuppressive therapy beyond six months in such patients if they are doing well, serum albumin is increasing, anti-PLA2R antibody levels (if initially positive) are low or decreasing, or if the patients are at high risk of having an adverse event with immunosuppressive therapy.

•In moderate-risk patients who show a progressive decline in proteinuria over the observation period, we withhold immunosuppression and continue general supportive measures.

•In most moderate-risk patients who require immunosuppressive therapy, we suggest rituximab rather than glucocorticoids plus cytotoxic therapy or a CNI (Grade 2C). If rituximab is unavailable, either combination therapy with glucocorticoids plus a cytotoxic agent or CNI monotherapy is a reasonable alternative. If cytotoxic therapy is chosen, we prefer cyclophosphamide over chlorambucil, since chlorambucil has more side effects. If a CNI is selected, either cyclosporine or tacrolimus is a reasonable choice. (See 'Rituximab' above and 'Cytotoxic therapy plus glucocorticoids' above and 'Calcineurin inhibitors' above.)

●Low risk of progression – Patients at low risk of progression do not require immunosuppressive therapy and can be managed with continued observation and general supportive measures only. Such patients generally have excellent long-term prognosis and often undergo spontaneous partial or complete remission. (See 'Low risk of progression' above.)

●Monitoring – All patients who are receiving treatment for primary MN should be closely monitored for clinical response to therapy. In patients with PLA2R-associated primary MN, serial assessment of anti-PLA2R antibody levels may be useful for monitoring the immunologic activity of disease and guiding treatment decisions. (See 'Monitoring and modifying therapy' above.)

●Relapsing disease – In patients who develop a relapse, we suggest a six-month observation period with general supportive measures to reassess the risk for progressive disease, using the same approach as that for newly diagnosed patients with primary MN. In patients who require immunosuppressive therapy, the choice of agent for relapse depends upon the initial immunosuppressive regimen. (See 'Treatment of relapsing disease' above.)

●Resistant disease – The optimal approach to patients with resistant disease is not known. Mycophenolate mofetil (MMF), natural adrenocorticotropic hormone (ACTH) gel, and plasma exchange are reasonable therapeutic options, although high-quality data demonstrating their efficacy are lacking. In general, patients with resistant disease should be referred to centers that have clinical expertise in the treatment of patients with MN. (See 'Treatment of resistant disease' above.)