Cyclophosphamide (CYC) has proven beneficial in preserving renal function in patients with lupus with diffuse proliferative glomerulonephritis (DPGN). However, the optimal route of CYC administration is unknown because direct comparative studies are unavailable. In this open study, we compared the renal outcome of two historical cohorts of patients with diffuse proliferative lupus nephritis (World Health Organization classes IVa and IVb) treated with either intravenous (IV) pulse CYC (group A; n = 22) or sequential oral CYC followed by azathioprine (AZA; group B; n = 21) and followed up prospectively. Both groups of patients had similar clinical, biochemical, and renal parameters at baseline. At 24 months posttreatment, significant improvements in proteinuria, creatinine clearance, serum albumin level, and lupus serological results were evident in both groups. Compared with patients in group A, patients in group B had more complete or partial remission (90% versus 73%) and less risk for treatment failure (5% versus 14%), renal flares (5% versus 14%), and doubling of creatinine levels (5% versus 9%), but the difference was not statistically significant. However, patients treated with oral immunosuppression had an insignificant increase in rates of herpes zoster infection (19% versus 9%) and menstrual disturbance (50% versus 29%). We conclude that sequential oral immunosuppression with CYC and AZA tended to have better efficacy than IV pulse CYC in the treatment of lupus DPGN but was associated with more toxicities. Additional randomized trials involving a larger cohort of patients with a longer period of observation are necessary.