BACKGROUND Thalidomide is an oral agent with significant activity in one-third of patients with refractory myeloma. However, long-term continuous administration of thalidomide can be associated with significant side effects such as deep-vein thrombosis and peripheral neuropathy. Furthermore, it is not clear whether continuous administration of thalidomide is necessary for its antimyeloma effect. We performed a phase II study with a combination that was based on the intermittent administration of thalidomide. METHODS A total of 53 patients with previously treated myeloma received cyclophosphamide 150 mg/m(2) p.o. every 12 h before meals on days 1-5, thalidomide 400 mg p.o. in the evening on days 1-5 and 14-18 and dexamethasone 20 mg/m(2) in the morning after breakfast on days 1-5 and 14-18 (CTD). The CTD combination was repeated every 28 days for three courses. Subsequently, responding patients were scheduled to receive maintenance treatment with monthly courses of CTD administered only for the first five days of each month. RESULTS On an intention-to-treat basis, 32 patients (60%) achieved a partial response with a median time to response of 1.5 months. Among the 43 thalidomide-naïve patients, 67% responded. Toxicities were mild or moderate and the cumulative incidence of deep-vein thrombosis and peripheral neuropathy was 4 and 2%, respectively. The median time to progression for responding patients was 12 months and the median overall survival for all patients was 17.5 months. CONCLUSIONS The oral, outpatient pulsed CTD regimen is associated with significant activity in patients with previously treated multiple myeloma. The incidence of deep-vein thrombosis and peripheral neuropathy appears to be lower than expected when thalidomide is being administered on a continuous basis.