OBJECTIVE： To systematically evaluate the efficacy and safety of cyclosporin combined with glucocorticoid versus cyclophosphamide combined with glucocorticoid in the treatment of membranous nephropathy （MN）. METHODS： Retrieved from Embase， Medline， CNKI， VIP and Wanfang database， RCTs about cyclosporin combined with glucocorticoid （trial group） versus cyclophosphamide combined with glucocorticoid （control group） in the treatment of MN were collected. Meta-analysis was conducted by using Rev Man 5.3 statistical software after literature screening， data extraction and quality evaluation with Jadad scale. RESULTS： Totally 6 RCTs were included， involving 312 patients in total. Results of Meta-analysis showed that remission rate 3 months after treatment [OR＝3.42，95％CI（2.05，5.71），P＜0.000 01] and relapse rate [OR＝3.12，95％CI（1.45，6.70），P＝0.004]， leukocyte count 12 months after treatment [MD＝1.77，95％CI（0.96，2.58），P＜0.000 1] in trial group were significantly higher than control group. There was no statistical significance in remission rate 6 months after treatment [OR＝2.06，95％CI（0.80，5.30），P＝0.13] and remission rate 12 months after treatment [OR＝1.30，95％CI（0.68，2.48），P＝0.42]， blood creatinine level 3 months after treatment [MD＝－1.55，95％CI（－6.72，3.62），P＝0.56] and blood creatinine level 6 months after treatment [MD＝－1.21，95％CI（－5.96，3.54），P＝0.62]， cholesterol level 12 months after treatment [MD＝－0.77，   95％CI（－1.81，0.28），P＝0.15] or ALT level[MD＝－0.40，95％CI（－4.38，3.58），P＝0.98] between 2 groups. ADR were reported in 5 RCTs， but their results were different. CONCLUSIONS： Long-term efficacy of cyclosporine combined with corticosteroid is similar to that of cyclophosphamide combined with corticosteroid in the treatment of MN. Cyclosporin combined with glucocorticoid has a faster effect， but a higher relapse rate.