In the present phase II study, we tested the efficacy of a single course of rituximab (375 mg/m2 on days 1, 8, 15 and 22) as treatment for relapsed myeloma. The rationale for this study was the identification of a population of clonotypic CD20+ B cells that are believed to be precursors of malignant plasma cells. In addition, CD20 was expressed on 10% and 50% of bone marrow plasma cells in two of the ten patients enrolled. Following rituximab treatment, none of the patients achieved an objective response. Two patients had stable disease at month 6, the predefined end of the study, while, at that time, two patients were classified as having progressive disease. One patient opted to withdraw from the study at month 3, at which time he had stable disease. The other five patients had to be withdrawn early from the post-treatment observation because of need of salvage therapy for progressive disease. WHO grade <or=2 toxicity was seen in four patients. Peripheral B cells significantly decreased at 3 months, while no significant change of bone marrow myeloma cells was noted at that time. Mean paraprotein levels increased slightly during follow-up but IgM levels dropped in all patients, indicating an effective targeting of normal, short-lived plasma cells. Taken together, rituximab treatment yielded significant reductions in circulating B cells and serum IgM levels but had no beneficial clinical effect.