Treatment of lupus nephritis (LN) with cyclophosphamide (CYC) is effective but retains a certain severe adverse effect. Tacrolimus (TAC) may be a suitable treatment for LN. Forty patients with diffuse proliferative or membranous LN were recruited for this non-randomized open-label study - 67.5% (27/40) had nephrotic proteinuria (>3.5 g/day) and 50.0% (20/40) had low estimated glomerular filtration rate (eGFR) (<60 mL/min/1.73m(2)). We compared the efficacy and adverse effects of TAC (0.04-0.08 mg/kg/d)/prednisone for 12 months (TAC group, n = 20) with intravenous CYC (750 mg/m(2) per month)/prednisone for six months followed by azathioprine (Aza) (100 mg/day)/prednisone for six months (CYC group, n = 20). The TAC target concentration was 6-8 ng/mL or 4-6 ng/mL, respectively, when induction or maintenance therapy was required and 4.0 ng/mL for patient with renal insufficiency. In the TAC group, mean urinary protein excretion decreased significantly from 5.00 ± 1.91 g/day at baseline to 2.54 ± 1.68 g/day after two weeks of therapy (P < 0.001), compared with the CYC group (4.9 ± 19.4 g/day), P = 0.001, and 65.0% (13/20) achieved partial remission at one month, compared with the CYC group (0/20), P < 0.001. The incidence of complete remission (CR) was significantly higher in the TAC group than in the CYC group (55.0% vs.15.0% by five months, P = 0.008, and 75.0% vs.40.0% by 12 months, P = 0.025, respectively). The significant improvement in serum anti-dsDNA and systemic lupus erythematosus (SLE) disease activity index (DAI) was in the TAC group relative to the CYC group at 12 months (P = 0.031, P = 0.003, respectively). The eGFR improved in the TAC group from 59.90 ± 23.64 mL/min/1.73m(2) at baseline to 93.75 ± 28.52 mL/min/1.73m(2) after 12 months, P = 0.001. In the CYC group, two patients developed end-stage renal disease (ESRD), three patients experienced serious pneumonia, and one patient died. Our preliminary study showed TAC is a safe and effective treatment for LN with severe renal disease, and with less-severe adverse events compared with CYC followed Aza therapy. Further larger sample studies are needed to confirm our conclusion.