This multi-center, randomized controlled trial was conducted from June 2013 to May 2016. Patients with biopsy-proven idiopathic MN were assessed for eligibility for this study at multiple centers in the Republic of Korea. We also included patients who underwent a biopsy within the preceding 12 months, experienced worsening proteinuria, and exhibited deteriorating renal function, but satisfied the following inclusion criteria. Patients who were ≥ 18 years old were enrolled in this study if they had proteinuria > 8 g/day. Patients with proteinuria < 8 g/day were also enrolled if they met 3 or more of the following criteria: 1) eGFR < 60 mL/min/1.73 m², 2) hypertension (blood pressure ≥ 140/90 mmHg or ≥ 120/80 with anti-hypertensive drugs), 3) 24-hour urinary protein > 5.0 g/day or spot urine protein to creatinine ratio > 5.0 g/g, 4) serum albumin < 3.0 g/dL, and 5) selectivity index > 0.2. eGFR was calculated by the modification of diet in renal disease (MDRD) equation. The selectivity index was calculated using the following equation: urine IgG × serum albumin/serum IgG × urine albumin. Exclusion criteria included the presence of moderate to severe gastrointestinal disorder at screening; a history of allergy to mycophenolate mofetil or cyclosporine; acute or chronic allergy within 4 weeks; presence of serious life-limiting comorbid disorders such as malignancy or uncontrollable active infection; drug or alcohol addiction within 6 months; uncontrolled high blood pressure (≥ 160/100 mmHg); eGFR ≤ 30 mL/min/1.73 m² at screening; absolute neutrophil count < 1,500/mm³ or white blood cell (WBC) < 2,500/mm³; platelets < 100,000/mm³; three-times greater than normal liver function test values; pregnancy; or lactation. Patients who had received immunosuppressive agents within 6 months for secondary MN with systemic disorder, or had a life expectancy of less than 1 year were also excluded.

All patients were assigned using a block randomization technique to either the MMF group or the CsA group. The table of random numbers was generated using the SAS randomization program, which was managed by the Department of Clinical Statistics in the clinical trial center at Kyungpook National University Hospital in Daegu, Korea. Oral MMF or CsA were provided as prepacked drugs in identical bottles. Allocation concealment was done by sealed sequentially numbered opaque envelopes. They were consecutively numbered and bottles were provided to the patients according to the number allocated. Treatment consisted of oral prednisolone 0.15 mg/kg up to a maximum dose of 15 mg/day for patients in both groups. Prednisolone was maintained at a minimum of 5 mg/day during the study period. In the MMF group, oral MMF (Myconol^®, Hanmi Pharmaceutical, Seoul, Korea) was added to corticosteroids. The MMF treatment regimen consisted of 500 mg twice daily in patients weighing less than 50 kg, or 750–1,000 mg twice daily in patients weighing more than 50 kg. The dose of MMF was adjusted in a range of 500–1,000 mg twice daily based on laboratory findings, at the discretion of the attending physician. The dose of MMF was reduced by 25%–33% of the previous dose when a patient had moderate to severe diarrhea. MMF was withheld when the WBC count was less than 4,000/mm³ or the patient had intolerable gastrointestinal symptoms, and was restarted at a 50% dose at least 2 weeks after recovery. MMF was discontinued when the patient showed severe adverse events, more than doubled values in liver function tests, newly developed malignancy, or increased serum creatinine level by over 50%. In the CsA group, oral CsA (Implanta soft cap^®, Hanmi Pharmaceutical) was added to corticosteroids. CsA was started at a dose of 4 mg/kg and the dose was adjusted to maintain a 100 ± 50 ng/mL trough blood level. The trough level of CsA was not an absolute target and the dose could be adjusted at the physician's discretion. The dose was decreased in cases in which serum CsA level ≥ 250 ng/mL, there was an increase in serum creatinine level of more than 0.3 mg/dL compared to baseline, elevated serum aspartate aminotransferase (AST) or alanine aminotransferase (ALT) level, or increased serum bilirubin level ≥ 2 mg/dL. CsA was discontinued when serum creatinine level did not improve 4 weeks after the dose was reduced in cases of increased serum creatinine ≥ 30% over baseline. CsA was also discontinued when the patient had severe adverse events or uncontrolled high blood pressure despite the use of three or more anti-hypertensive drugs at maximally-tolerated doses. Concurrent treatment was not standardized; however, blood pressure and proteinuria were primarily treated with angiotensin-converting enzyme inhibitors or angiotensin receptor blockers. Statins were used to decrease serum cholesterol levels. Patients were assessed 2 weeks after starting the treatment regimen, every 4 to 6 weeks for the next 24 weeks, and then every 12 weeks thereafter. Patients were followed up for a total of 48 weeks. Vital signs, side effects of treatment, and laboratory data, including spot urine protein to creatinine ratio, complete blood cell count, serum albumin, and serum creatinine, were prospectively collected at each visit. Lipid profile and 24-hour urinary protein were monitored at baseline, and at 12, 24, and 48 weeks.

The primary outcome was the rate of complete or partial remission of proteinuria at 48 weeks in the MMF and CsA groups. To assess remission of proteinuria at 48 weeks, level of 24-hour urinary protein was used. Complete remission was defined as a decrease in proteinuria to ≤ 200 mg/day and a sustained serum albumin level ≥ 3.5 g/dL. Partial remission was defined as a decrease in proteinuria to > 200 and < 3,500 mg/day or a decrease greater than 50% compared to baseline.19 As secondary outcomes, eGFR, relapse rate, and improvement of hypoalbuminemia and hypercholesterolemia at 48 weeks were compared between the 2 groups. A relapse was defined as proteinuria ≥ 3,500 mg/day after achievement of partial or complete remission or an increase in proteinuria greater than 50% in patients in whom proteinuria had improved initially by more than 50%.

To assess safety, any adverse events, such as leukopenia, infectious episodes, malignancy, and gastrointestinal symptoms, including diarrhea and epigastric discomfort, were recorded. To compare gastrointestinal symptoms in the 2 groups, patients were asked to fill in a questionnaire including the Gastrointestinal Symptom Rating Scale (GSRS)2021 and Gastrointestinal Quality of Life Index (GIQLI)22 at baseline and at 48 weeks.

Serum samples from the patients at baseline and at 48 weeks were obtained and stored at −80°C. Levels of anti-phospholipase A2 receptor (anti-PLA2R) antibodies (Ab) were measured using a commercial ELISA kit (Euroimmun AG, Lubeck, Germany) in accordance with the manufacturer's protocol, as described in previous studies.2324 A value higher than 20 RU/mL was regarded as a positive result.

It was estimated that at least 28 patients in each group would be needed for 80% power assuming a 5% significance level. As we estimated a 10% screening failure and dropout rate, 31 patients would finally need to be included in each group.

The intention-to-treat (ITT) analyses included all randomly assigned patients. We compared complete or partial remission rates of proteinuria in the 2 treatment groups using a non-inferiority test: the lower limit of the confidence interval (CI) of the difference in proportions of patients who achieved complete or partial remission of proteinuria in the MMF and CsA groups could not exceed the threshold of −20%, which is 50% of the difference in remission rate between spontaneous remission (−30%)34 and CsA treatment (−70%).2526 Mean values and frequencies of the parameters were compared by independent sample t-tests or χ² tests, as appropriate. For non-normally distributed variables (body mass index, serum creatinine, and high-sensitivity C-reactive protein [hsCRP]), the Mann-Whitney U test was performed after a test for normality. If patients discontinued treatment due to a lack of efficacy, their last value was carried forward for the analysis. Cumulative probabilities of complete or partial remission were estimated using the Kaplan-Meier method. SPSS version 19.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analyses. A P value of < 0.05 was considered statistically significant.

Approval was obtained from the Institutional Review Board of Kyungpook National University Hospital (KNUH_10-1096). Written informed consent was obtained from all patients prior to randomization. The trial was registered at ClinicalTrials.gov site (www.ClinicalTrials.gov NCT01282073).

Of 43 MN patients with severe proteinuria screened for this study, 1 was not eligible, 2 refused to participate, and 1 was in poor general condition and had a rapidly rising serum creatinine level. Thus, 39 patients were included in the study. Of these, 21 and 18 were allocated to the MMF and CsA groups, respectively (Fig. 1). The baseline characteristics of patients were comparable between the 2 groups (Table 1).

A total of 16 patients (76.1%) in the MMF group and 12 (66.7%) in the CsA group achieved complete or partial remission of proteinuria at 48 weeks (Table 2; P = 0.805). The absolute difference in complete or partial remission of proteinuria was 9.4% (95% CI, −0.18 to 0.38), which did not exceed the margin. The cumulative incidence of complete or partial remission of proteinuria at 48 weeks was 82.8% in the MMF group and 70.9% in the CsA group, which was not significantly different between the groups (Fig. 2; P = 0.929).

Rate of complete or partial remission was compared based on the categories of proteinuria at baseline. The results were presented in Table 3. Remission rate of proteinuria was higher in patients with proteinuria less than 8 g/day compared to patients with proteinuria > 8 g/day. However, there was no significant difference between the MMF and CsA group in patients with each categories of proteinuria (Table 3).

Proteinuria at baseline and 48 weeks measured 8.9 ± 5.9 and 2.1 ± 3.1 g/day, respectively, in the MMF group compared to 8.4 ± 3.5 and 3.2 ± 5.7 g/day, respectively, in the CsA group. Changes in proteinuria from baseline were comparable between the 2 groups at 12 (−57.0% CsA vs. −43.3% MMF group; P = 0.330), 24 (−76.7% CsA vs. −57.6% MMF group; P = 0.174), 36 (−75.9% CsA vs. −58.5% MMF group; P = 0.326), and 48 weeks (−63.9% CsA vs. −69.0% MMF group; P = 0.745) (Fig. 3). Proteinuria was significantly decreased at each time point compared to baseline in both groups. In the CsA group, there was no significant difference in reduction of proteinuria in each time point (−57.0% at 12 weeks vs. −76.7% at 24 weeks, P = 0.095; −57.0% at 12 weeks vs. −75.9% at 36 weeks, P = 0.145; −57.0% at 12 weeks vs. −63.9% at 48 weeks, P = 0.665; −76.7% at 24 weeks vs. −75.9% at 36 weeks, P = 0.947; −76.7% at 24 weeks vs. −63.9% at 48 weeks, P = 0.398; −75.9% at 36 weeks vs. −63.9% at 48 weeks, P = 0.472). In the MMF group, there was also no significant difference in reduction of proteinuria in each time point (−43.3% at 12 weeks vs. −57.6% at 24 weeks, P = 0.376; −43.3% at 12 weeks vs. −58.5% at 36 weeks, P = 0.414; −43.3% at 12 weeks vs. −69.0% at 48 weeks, P = 0.070; −57.6% at 24 weeks vs. −58.5% at 36 weeks, P = 0.963; −57.6% at 24 weeks vs. −69.0% at 48 weeks, P = 0.427; −58.5% at 36 weeks vs. −69.0% at 48 weeks, P = 0.531).

Serum albumin was significantly increased at each time point in both groups compared to baseline except 12 weeks in CsA group (Table 4). Four patients (19.0%) in the MMF group and 4 (22.2%) in the CsA group had relapse of proteinuria during the study period (P = 1.000).

eGFRs at baseline and at 48 weeks were 73.9 ± 31.0 and 70.1 ± 19.2 mL/min/1.73 m², respectively, in the MMF group compared to 84.9 ± 25.2 and 84.0 ± 32.6 mL/min/1.73 m², respectively, in the CsA group. There was no significant difference in eGFR between the 2 groups at each visit. Hypercholesterolemia were improved at 48 weeks compared to baseline in both groups, but did not show significant differences between the groups at each visit (Table 4).

The trough CsA level in the CsA group was 134.6 ± 55.8 and 131.7 ± 36.4 ng/mL at 18 and 36 weeks, respectively. The dose of MMF and CsA was not significantly different at each time point in both groups (Supplementary Table 1).

Serum samples could be obtained from 33 of the 39 patients (18 patients in the MMF group and 15 in the CsA group) for measurement of anti-PLA2R Ab levels at baseline and 48 weeks. Anti-PLA2R Ab levels from the samples in baseline and 48 weeks were strongly correlated with 24-hour urinary protein (r = 0.546, P = 0.000, Fig. 4A). Twenty-one patients (63.6%) showed anti-PLA2R Ab positivity. No significant difference in anti-PLA2R Ab level was found between the MMF and CsA groups at baseline and 48 weeks. Anti-PLA2R Ab levels at baseline significantly decreased at 48 weeks in the complete or partial remission group (P = 0.001), but showed no significant difference between baseline and 48 weeks in the no-response group (P = 0.679, Fig. 4B). In 10 patients (5 patients in the MMF group and 5 in the CsA group), positive anti-PLA2R Ab at baseline turned negative at 48 weeks. Of the patients, 9 had complete or partial remission.

Adverse events occurred in 12 (57.1%) and 9 (50.0%) patients in the MMF and CsA groups, respectively (Table 5; P = 0.901). Adverse events, such as leukopenia and thrombocytopenia, did not occur in either group during treatment. One patient in each group developed anaemia. Two patients in each group had diarrhea, which was mild and transitory. Three patients in the MMF group and 4 in the CsA group developed epigastric discomfort (P = 0.742). Infectious episodes occurred with a similar frequency in both groups. One patient in each group developed malignancy. The patient in the MMF group was diagnosed with early gastric cancer, and the patient in the CsA group was diagnosed with a bladder tumour. There were no significant differences between the two groups in changes in the GSRS and GIQLI scores from baseline to 48 weeks (Table 6; P = 0.660 for GSRS and 0.221 for GIQLI).