Multitarget therapy with a corticosteroid, cyclosporine and mycophenolate mofetil for idiopathic membranous nephropathy: a prospective randomized controlled trial

ABSTRACT

Graphical Abstract

Open in new tabDownload slide

INTRODUCTION

Idiopathic membranous nephropathy (IMN) is one of the most common causes of nephrotic syndrome in adults [1]. Immunosuppressive therapy is mandatory in patients with persistent nephrotic proteinuria, impaired renal function and high titres of anti-phospholipase A₂ receptor (PLA2R) antibodies [2].

Immunological dysregulation is fundamental to IMN pathogenesis, with evidence of B- and T-cell involvement [3]. Combination immunosuppressants that address different aspects of the immune response may be more effective than monotherapy in patients with IMN. Multitarget therapy [i.e. a corticosteroid combined with cyclosporine and mycophenolate mofetil (MMF)] has been the basis of maintenance therapy following solid organ transplantation; these agents have additive inhibitory effects on lymphocytes [4–7]. Cyclosporine binds a cytoplasmic receptor and inhibits cell phosphatase, thus blocking activation of T-cell-specific transcription factors [8]. Furthermore, cyclosporine enhances the effect of corticosteroids by competitively inhibiting excretion of intracellular corticosteroids through P-glycoprotein on lymphocytes [9]. MMF is hydrolyzed to the active drug mycophenolic acid, which is a potent inhibitor of T- and B-lymphocyte proliferation via reversible inhibition of inosine 5-monophosphate dehydrogenase [10].

Cyclosporine plus a corticosteroid has short-term efficacy and is safe for induction of remission of IMN; however, relapse rates are high (40%–50%) after discontinuation [11–13]. Some small observational studies have found that MMF (usually given in combination with prednisone) reduces proteinuria in patients with IMN [14, 15]. Data from a small study in China showed that, at 6 months, multitarget therapy (a combination of corticosteroid, tacrolimus and MMF) is more effective and safe for treating refractory IMN than is cyclophosphamide plus a corticosteroid [16]. However, data comparing the efficacy of multitarget therapy (a combination of corticosteroid, cyclosporine and MMF) and a cyclical corticosteroid–cyclophosphamide regimen against IMN are lacking.

Here, we designed and conducted a study to compare the efficacy and safety of multitarget therapy (a combination of corticosteroid, cyclosporine and MMF) and cyclically alternating corticosteroid–cyclophosphamide.

MATERIALS AND METHODS

Trial design

This single-centre, prospective, randomized controlled trial was conducted at Beijing Friendship Hospital, Capital Medical University (Beijing, China). The study protocol was approved by the Bioethics Committee of Beijing Friendship Hospital on 29 May 2020 (2020-P2-104-02) and registered at ClinicalTrials.gov (NCT04424862). Patients provided written informed consent to participate in this trial.

Patient cohort

Patients were eligible if the diagnosis of IMN had been confirmed by renal biopsy. All renal biopsies were examined by light, immunofluorescence and electron microscopy, and reviewed by two principal investigators and two renal pathologists. All eligible patients were followed for an observational period of at least 3 months.

The inclusion criteria were: (i) age 18–70 years; (ii) estimated glomerular filtration rate (eGFR; according to the Chronic Kidney Disease Epidemiology Collaboration formula) ≥60 mL/min/1.73 m²; (iii) serum albumin concentration <30 g/L; and (iv) nephrotic-range proteinuria (>3.5 g/24 h, without a decrease of >50% during the observational period). During at least the previous 3 months before randomization, all patients received the standard of care that included blockers of renin–angiotensin system, blood pressure management targeting a value of less than 130/80 mmHg, with exceptions in case of intolerance, contraindications or low blood pressure, before the screening period.

The exclusion criteria were: (i) secondary causes of membranous nephropathy (i.e. hepatitis B and C, systemic lupus erythematosus, medications, malignancies); (ii) being pregnancy or breastfeeding; (iii) uncontrollable active infectious disease; and (iv) previous treatment with immunosuppressive agent (3 months before screening), rituximab or any other biologic agent (6 months before screening), and patients who were non-responders to previous immunosuppressants.

Randomization

Randomization and concealment of allocated treatment were achieved using computer-generated random numbers and sealed envelopes. Eligible patients were allocated to the study groups in a ratio of 1:1. Randomization, allocation and concealment were supervised by an independent researcher who was blinded to the study protocol.

Interventions and follow-up

Multitarget therapy group

Patients received oral prednisone at 1 mg/kg/day (maximum dose = 60 mg/day) for 2 weeks, then tapered by 5 mg/week. After the dose had been reduced to 30 mg/day, it was maintained there for 1 month and then tapered by 5 mg/month. After the dose had been reduced to 20 mg/day, it was maintained there for 2 months and then tapered by 2.5 mg every 2 months. After reduction of the dose to 5 mg/day, it was maintained there for 4 months. The initial dosage of cyclosporine was 100 mg/day, this dose being adjusted to maintain a trough concentration of 80–120 ng/mL. MMF (1 g/day) was given for 6 months, tapered to 0.75 g/day for 12 months, and tapered to 0.5 g/day for another 12 months.

Cyclosporine doses were adjusted monthly according to efficacy or to achieve the target blood concentration of 80–120 ng/mL. If complete remission had been achieved, we maintained the cyclosporine dose for 2 weeks, then tapered it by 25 mg/day at 3-monthly intervals. After reduction of the dose to 50 mg/day, it was maintained for 12 months, then tapered to 25 mg/day for another 12 months. If partial remission was achieved, we maintained the target dose for 12 months, and then adjusted the dose according to the regimen outlined above for patients in complete remission. If there was no response to treatment, we maintained the target dose for 6 months. Thereafter, if the total urinary protein had decreased by <25%, we suggested changing the regimen. If the total urinary protein had decreased by ≥25%, we maintained the dose for another 6 months, then followed the regimen described above for patients in partial remission.

Corticosteroid–cyclophosphamide group

These patients received a corticosteroid during Months 1, 3 and 5 (methylprednisolone 0.5 g intravenously on Days 1, 2 and 3, then prednisone 0.5 mg/kg/day orally from Days 4 to 30). During Months 2, 4 and 6, the patients received oral cyclophosphamide adjusted for age and renal function (1.0–2.0 mg/kg/day for 30 days, maximum dose 100 mg/day).

Outcomes

The primary outcome was the composite of complete or partial remission at 12 months. Secondary outcomes included the complete remission at 12 months and adverse events.

Complete remission was defined as reduction of proteinuria to ≤0.3 g/24 h. Partial remission was defined as reduction of proteinuria by >50% of baseline, and <3.5 g/24 h. No response was defined as a reduction in proteinuria of <50% of the baseline amount. Patients were considered positive for anti-PLA2R antibodies if the serum concentration at baseline was >20 RU/mL, as determined by a standardized commercial enzyme-linked immunosorbent assay (Euroimmune, Lubeck, Germany). Adverse events were defined as any untoward medical occurrence. Serious adverse events were defined as any untoward medical occurrence that resulted in death, was life-threatening, required hospitalization or resulted in persistent or significant disability/incapacity.

Statistical analysis

On the basis of results of previous studies, we hypothesized a remission rate of 78.3% at 1 year for the corticosteroid–cyclophosphamide group and 85.7% for the multitarget therapy group. We planned to enrol 78 patients, assuming a statistical power of 80%. The primary outcome (complete/partial remission at 12 months) was analysed by intention-to-treat and per-protocol estimating the relative risk (RR) with a 95% confidence interval (CI) and compared between the study groups using Fisher's exact test. The hazard ratios for complete remission at 12 months were estimated to evaluate secondary outcomes. Differences in continuous variables between the two groups were analysed with an unpaired Student's t-test or Mann–Whitney U-test, as appropriate. Differences between categorical variables were analysed with Fisher's exact test, as appropriate. Longitudinal data (i.e. serum concentrations of albumin and creatinine, proteinuria, anti-PLA2R antibody concentrations and other repeated measures) from randomization until 12 months were analysed using a generalized estimating equation. Time-to-event analyses (time to nephrotic syndrome relapse) were carried out using the log-rank test. Patients who dropped out of the study without reaching the primary outcome were censored.

RESULTS

Patients

From June 2020 through January 2022, 78 patients were enrolled. They were assigned randomly to a multitarget therapy group (39 patients) or corticosteroid–cyclophosphamide group (39 patients) (Fig. 1). Their mean age was 52.6 years and 54 of them (69.2%) were men. No significant differences between the two groups were identified at baseline (Table 1).

Figure 1:

Enrollment, randomization and follow-up of study participants.

Open in new tabDownload slide

Eight patients in the multitarget therapy group and three in the corticosteroid–cyclophosphamide group discontinued the assigned treatment. The remaining 31 patients in the multitarget therapy group and 36 in the corticosteroid–cyclophosphamide group adhered completely to the assigned treatment regimen. All 78 enrolled patients completed follow-up.

The cyclosporine dose was increased in 14 of the patients in the multitarget therapy group, in the first 3 months in 11 of them. In the first 3 months, the maximum cyclosporine dose was 3.3 mg/kg/day and the mean cyclosporine dose 1.6 ± 0.3 mg/kg/day. These doses achieved a trough cyclosporin blood concentration of 68.8 ± 25.2 ng/mL.

Clinical outcomes

The primary outcome of complete or partial remission at 12 months was achieved by 31/39 patients (79%) in the multitarget therapy group and 34/39 (87%) in the corticosteroid–cyclophosphamide group (RR 0.93; 95% CI 0.72–1.21; P = .85; log-rank test) (Table 2; Fig. 2a) according to intention-to-treat analysis. Per-protocol analysis revealed no significant difference in the rate of primary outcome between the groups: 28 of 31 (90%) in the multitarget therapy group and 32 of 36 (89%) in the corticosteroid–cyclophosphamide group (RR 1.02; 95% CI 0.86–1.20) (Table 2).

Figure 2:

Probability of nephrotic syndrome remission during the 12 months of follow-up.

Open in new tabDownload slide

Complete remission at 12 months was achieved by 16 patients (41%) in the multitarget therapy group and 14 (36%) in the corticosteroid–cyclophosphamide group. The cumulative incidence of complete remission did not differ significantly between these groups (P = .47; log-rank test) (Fig. 2b). Amounts of urinary protein and serum concentrations of anti-PLA2R antibody decreased and the serum albumin concentration increased gradually in both groups during follow-up. There were no significant differences in the changes in amount of urinary protein, serum concentrations of anti-PLA2R antibody, or albumin concentrations between the two groups (P = .31, .88 and .39, respectively, using generalized linear models) (Fig. 3a and b). The proportions of anti-PLA2R antibody–positive patients who achieved complete or partial remission at 6 and 12 months were higher in the corticosteroid–cyclophosphamide group (83.3% and 86.7%, respectively) than in the multitarget therapy group (63.6% and 81.8%, respectively); however, this difference was not significant (Supplementary data, Table S1). Serum creatinine concentrations did not change significantly between baseline and 12 months in either group (Fig. 3c).

Figure 3:

Evolution of urinary protein, serum anti-PLA2R antibody, serum albumin and serum creatinine. Data are presented as mean ± standard deviation.

Open in new tabDownload slide

Adverse events

The prevalence of adverse events was significantly lower in the multitarget therapy group [46% (18 of 39) vs 74% (29 of 39); P < .05] (Table 3). No deaths occurred in either group. The most common adverse event was infection (most commonly of the upper respiratory tract). Acute kidney injury occurred more often in the multitarget therapy group; however, this difference was not significant (Table 3). The prevalence of hyperglycaemia was lower in the multitarget therapy than in the corticosteroid–cyclophosphamide group [0% (0 of 39) vs 28% (11 of 39); P < .001] (Table 3).

DISCUSSION

In our study, we found that a multitarget combination of corticosteroid, cyclosporine and MMF was noninferior to corticosteroid–cyclophosphamide in inducing remission at 12 months and had fewer adverse effects. The rate of remission of IMN in patients receiving multitarget therapy did not differ significantly from that in patients receiving corticosteroid–cyclophosphamide.

Multitarget immunosuppression with corticosteroids, calcineurin inhibitors (CNIs; including cyclosporine and tacrolimus) and MMF has been reported in a few studies of treatment of patients with refractory IMN. We searched PubMed and MEDLINE for articles published in English and found only one pilot study, which was conducted by Ballarin et al. [17]. These authors suggested that multitarget therapy (combination of steroids, tacrolimus and MMF) helped to induce remission in refractory IMN. In this pilot study, MMF was added after the third month in nine patients with IMN whose proteinuria remained higher than 1 g/day despite treatment with prednisone and tacrolimus. Five (55.6%) of those patients achieved remission by the end of therapy. However, the value of this multitarget combination as induction therapy has not yet been determined. To our knowledge, the present study is the first to report on a multitarget combination regimen of corticosteroid, cyclosporine and MMF as induction therapy for IMN.

Cyclosporine has been used to treat IMN for decades. One meta-analysis found that the efficacy of cyclosporine (usually 2–5 mg/kg/day) plus a corticosteroid in Asian individuals is greater than that of cyclophosphamide plus corticosteroid, whereas stopping cyclosporine treatment is correlated with high relapse rates [18], consequently necessitating its long-term use. Moreover, the potential nephrotoxicity of cyclosporine (which is dose- and time-dependent) raises concerns regarding this treatment option [19, 20]. Our decision to add MMF to a multitarget combination was based on reports that MMF reduces proteinuria in some patients with IMN [14, 15, 21, 22]. This enables lowering the CNI dose and thus reducing the risk of iatrogenic nephrotoxicity; furthermore, lower doses of multiple drugs may maximize efficacy and minimize adverse effects [23]. One study of Chinese patients with IMN reported a 68.8% remission rate at a cyclosporine dose of 2.1 ± 0.4 mg/kg/day (trough cyclosporine blood concentration 92.5 ± 23.5 ng/mL) [24]. The dosage of cyclosporine in our study was 1.6 ± 0.3 mg/kg/day with a trough cyclosporine blood concentration of 68.8 ± 25.2 ng/mL, which is significantly lower than in previous studies [24, 25]. However, in our study the remission rate was higher, suggesting that the combined effect of a corticosteroid, cyclosporine and MMF results in mixed humoral and cellular immunosuppression, making this combination more effective than a single agent.

In our study, adverse events occurred significantly less frequently in the multitarget therapy group than in the corticosteroid–cyclophosphamide group. The most common adverse event was infection and the most common site of infection was the respiratory tract. We noted a higher prevalence of infection in the corticosteroid–cyclophosphamide than in the multitarget therapy group (36% vs 23%); however, this difference was not significant. Corticosteroids increase the risk of hyperglycaemia; other immunosuppressive agents can also affect glycaemic control. Compared with tacrolimus, cyclosporine is less expensive, more widely available and has a better profile regarding glucose metabolism [26]. In patients with systemic lupus erythematosus on high-dose corticosteroid therapy, development of diabetes mellitus has been associated with concurrent use of MMF; this may be attributable to impaired insulin secretion as a result of increased stress on pancreatic beta cells [27]. The prevalence of hyperglycaemia was significantly lower in the multitarget therapy than in the corticosteroid–cyclophosphamide group, possibly because lower doses of corticosteroids, CNIs and MMF were used in the multitarget therapy group.

Our study had two main limitations. First, it was a single-centre study. Further studies are required to ascertain whether the efficacy of multitarget therapy can be generalized. Second, the follow-up period was only 12 months. Future long-term studies are needed to evaluate the prevalence of relapse and of potential long-term toxic effects of these immunosuppressive schedules.

In conclusion, we found that multitarget therapy is noninferior to corticosteroid–cyclophosphamide in inducing proteinuria remission at 12 months. Additionally, multitarget therapy had a more favourable safety profile than did corticosteroid–cyclophosphamide.

FUNDING

This study was supported by the Beijing Municipal Administration of Hospitals Incubating Program (Code: PX2022003) and the Research Foundation of the Beijing Friendship Hospital, Capital Medical University (Code: YYQDKT2019-22).

AUTHORS’ CONTRIBUTIONS

Z.D. conceived this study. Z.D. and W.L. designed the study. Y.D., Y.B., We.Guo, L.W., W.D., Wa.Guo and H.H. recruited patients and collected the clinical data. Z.D. and Y.D. analysed the data. Y.D. and Y.B. wrote the manuscript. All authors reviewed and revised the article and approved the final version.

CONFLICT OF INTEREST STATEMENT

None declared.

DATA AVAILABILITY STATEMENT

The data underlying this article are available in the article.

REFERENCES

Author notes