Leflunomide therapy for adult patients with steroid-dependent minimal change disease or primary focal segmental glomerulosclerosis

To the Editor: Minimal change disease (MCD) and focal segmental glomerulosclerosis (FSGS) are common histological subtypes of nephrotic syndrome in adults and children.^[1,2] Corticosteroids are the first-line treatment, whereas some patients present with steroid-dependent nephrotic syndrome. Repeated treatment with steroids is required, accompanied by serious adverse effects related to prednisone monotherapy in the long term.^[3] Leflunomide is a competitive and reversible inhibitor of dihydroorotate dehydrogenase that inhibits protein tyrosine kinases in proliferating T cells and B cells and has anti-inflammatory and antiproliferative effects.^[4] We previously reported that leflunomide therapy reduced the rate of relapse and the amount of prednisone to maintain remission in 16 adult patients with MCD and steroid-dependent or steroid-resistant nephrotic syndrome.^[5] We performed a retrospective study to investigate the efficacy and safety of leflunomide in steroid-dependent MCD and FSGS.

From 2012 to 2018, a total of 49 adult patients with steroiddependent nephrotic syndrome of biopsy-proven MCD or primary FSGS were enrolled. Among them, 40 patients had MCD, and nine patients had FSGS. The clinical parameters are shown in Supplementary Table 1. Data were shown as percentage or median (Q1, Q3). All patients received corticosteroid monotherapy without any immunosuppressive drugs and presented with steroid-dependent nephrotic syndrome with a previous threshold dose of prednisone of 30.0 [interquartile range (IQR), 20.0–40.0] mg/day. After relapse, complete remission was induced by prednisone at a dose of 0.1–1.0 mg·kg⁻¹·d⁻¹ (not exceeding 60 mg/day). Then, 37 patients received leflunomide treatment; among them, 28 patients received leflunomide at a dose of 20 mg/day, and nine patients (>65 years old) received 10 mg/day. Twelve patients received cyclosporin therapy, including eight patients at a dose of 100 mg/day, three patients at a dose of 150 mg/day, and one patient at a dose of 200 mg/day, with a whole blood trough concentration of 117.6 ± 45.0 ng/mL. No significant difference was found between the two groups, including age, sex, and follow-up time [Supplementary Table 1].

The research was in compliance with the Declaration of Helsinki and was approved by the Ethics Committee of Peking University First Hospital. Statistical analysis was performed using SPSS software (version 22.0; IBM, New York, NY, USA). All statistical analyses were two-tailed, and a P value < 0.05 was considered statistically significant.

In the leflunomide group [Supplementary Figure 1], 24/37 (65%) patients maintained remission until the endpoint of follow-up, with a median time of 18.9 (11.1, 34.7) months. Of them, 10/24 (42%) patients withdrew the usage of steroids for 8.1 (4.3, 27.6) months. In the patients who still received steroids, the dose of prednisone decreased from 30.0 (18.8,40.0) mg/day to 4.4 (2.5, 10.0) mg/day. There were 13/37 (35%) patients who relapsed with nephrotic syndrome during follow-up. The median time for them to maintain clinical remission was 10.8 (5.4, 20.1) months. Before relapse, 4/13 (31%) patients withdrew the usage of steroids for 9.4 (5.2, 20.3) months. In the patients who still received steroids at the time of relapse, the dose of prednisone was 15.0 (6.3, 18.8) mg/day. In all, 5/37 (14%) patients stopped steroids and leflunomide after clinical remission was maintained for 22.6 (19.3, 27.6) months. Three of them remained in remission until the end of follow-up, with a remission time of 35.8 ± 17.7 months. Two patients relapsed at 31.0 months and 32.3 months, respectively, after the termination of treatment. After relapse, they were treated with steroids and leflunomide again and reached complete remission, which was maintained for another 41.0 months and 40.0 months, respectively, until the endpoint of follow-up.

In the cyclosporin group [Supplementary Figure 1], only 2/12 patients maintained remission until the endpoint of follow-up, with times of 18.9 months and 1.9 months, respectively. They both still received prednisone at doses of 5.0 mg/day and 20.0 mg/day, respectively. Ten of 12 patients relapsed with nephrotic syndrome. The median time for them to maintain clinical remission was 3.5 (2.0, 9.6) months. They all received prednisone at the time of relapse, with a median dose of 15.0 (8.8, 25.6) mg/day.

Leflunomide vs. cyclosporin: The rate of continuous clinical remission was significantly higher in the leflunomide group than in the cyclosporin group (64.9% vs. 16.7%, P = 0.004) [Supplementary Table 1]. The time for these patients to maintain clinical remission was longer in the leflunomide group (18.9 [11.1, 34.7] vs. 10.4 [1.9, 18.9] months), although the difference did not reach significance (P = 0.345). The rate of steroid withdrawal was higher in the leflunomide group (41.7% vs. 0), but the difference was not significant (P = 0.508). In the patients who still received steroids, the median dose of prednisone was comparable (4.4 [2.5, 10.0] mg/day vs. 12.5 [5.0, 20.0] mg/day, P = 0.267). In the patients who had a relapse, the duration of clinical remission before relapse was significantly longer in the leflunomide group than in the cyclosporin group (10.8 [11.1, 34.7] months vs. 3.5 [2.0, 9.6] months, P = 0.021). Before relapse, in the leflunomide group, four patients withdrew steroids, whereas all the patients in the cyclosporin group still received steroids (4/13 [31%] vs. 0, P = 0.081). Of the patients who still received steroids, the median dose of prednisone was comparable between the two groups (15.0 [6.3, 18.8] mg/d vs. 15.0 [8.8, 25.6] mg/d, P = 0.661). The overall remission rate without relapse differed remarkably between the leflunomide group and the cyclosporin group (P < 0.001) [Figure 1]. The estimated average duration of remission was longer in the leflunomide group (45.9 [95% CI, 32.8–59.1] months) as compared with that in the cyclosporin group (6.9 [95% CI, 3.4–10.4] months, P < 0.001). All these steroiddependent patients maintained kidney function during the follow-up of this study. In the leflunomide group, the estimated glomerular filtration rate (eGFR) was 89.8 ± 26.4 ml·min⁻¹·1.73 m⁻² at the beginning of treatment and 94.1 ± 23.3 ml·min⁻¹·1.73 m⁻² at the endpoint (P = 0.114). In the cyclosporin group, the eGFR was 103.0 ± 22.2 ml·min⁻¹·1.73 m⁻² and 91.6 ± 33.7 ml·min⁻¹·1.73 m⁻², respectively (P = 0.293).

Figure 1.

Kaplan-Meier curves showed the remission rates of patients with steroid-dependent nephrotic syndrome of MCD or primary FSGS, in different therapeutic groups. No. at Risk: Number of patients at high risk of relapse. FSGS: Focal segmental glomerulosclerosis; MCD: Minimal change disease.

Leflunomide was well tolerated, and six adverse events were reported in five patients (13.5%) [Supplementary Table 2]. Two events (skin rash and diarrhea) occurred in one old patient. Three patients developed skin rashes, and one of them stopped the medication. Two patients showed an elevated blood pressure of < 160/100 mmHg. They all recovered rapidly with supportive treatments.

In this study, we evaluated the efficacy and safety of leflunomide therapy in 37 patients with steroid-dependent MCD or FSGS, with comparisons to the patients receiving cyclosporin in the same period of follow-up. The results from this retrospective study showed that leflunomide therapy was effective and safe in maintaining clinical remission, reducing the relapse rate, achieving steroid withdrawal, and even the complete cessation of all immunosuppressive agents in the treatment of the steroid-dependent nephrotic syndrome. The major limitation of this study is the small number of participants and the retrospective study design, which hindered us from performing further statistical analysis. It is worth conducting randomized controlled clinical trials to compare the efficacy of leflunomide to cyclophosphamide or cyclosporin therapy.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patients have given their consent for their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, although anonymity cannot be guaranteed.

Acknowledgements

The authors thank Jinying Wang for valuable help in data collection.

Funding

This work is supported by grants from Natural Science Foundation of China (No. 81870486, No. 82070732).

Conflicts of interest

None.