Objective To assess the efficacy and safety of mycophenolate mofetil (MMF) versus cyclophosphamide (CTX) in the induction treatment for lupus nephritis (LN). Methods Such databases as MEDLINE, EMBASE, SCIE, The Cochrane Library, the Cochrane Controlled Trials Register, CBM, and CNKI were searched from their establishment date to August of 2010 to retrieve the randomized controlled trials (RCTs) about MMF versus CTX for LN. The methodology quality of included studies was evaluated. The efficacy indexes i.e. the clinical total remission (TR), complete remission (CR), partial remission (PR), pathological activity index, the chronicity index and complete induction therapy rate (CIR), and the safety indexes i.e. the rate of patient intolerance-to-drug, the incidence of infection, leukopenia and diarrhea, were abstracted. Finally the Meta-analyses were conducted by using Cochrane Collaboration’s RevMan 4.2. Results Eight RCTs involving 773 patients met the inclusive criteria. The results of meta-analyses showed that the total remission rate (OR=1.49, 95%CI 1.10 to 2.02) and complete remission rate (OR=1.67, 95%CI 1.08 to 2.57) were significantly higher in the MMF group than the CTX group. There was no significant difference in the rate of partial remission, the complete induction rate, the rate of patient intolerance-to-drug, the incidence of infection and leukopenia. However, the incidence of diarrhea was higher in the MMF group (OR=2.99, 95%CI 1.87 to 4.78). The results of meta-analyses for type IV LN were the same. Conclusion MMF is superior to CTX in the induction therapy to Lupus Nephritis (type III, IV, V), but the incidence of diarrhea is higher.