In 1990 the Italian Non-Hodgkin's Lymphoma Cooperative Study Group (NHLSG) started a multicenter study on the role of interferon (IFN) in multiple myeloma (MM). The schedule of treatment was based on the assumption that melphalan plus prednisone (MP) would be better for good-prognosis patients, whereas poor-prognosis patients would benefit from polychemotherapy. Accordingly, IFN was included randomly for the induction treatment of good-prognosis patients and randomly as maintenance of the response achieved in both groups. Up to now 78 patients of the 124 enrolled have completed the induction treatment and are evaluable for response and response duration. The overall response rate was 59%. Sixty-two percent of good-prognosis patients obtained objective response, 9/14 (64%) with MP and 9/15 (60%) with MP+IFN. Up to now, with a median follow-up of 9 months from the evaluation of response, no difference has been recorded between the maintenance and no maintenance groups on relapse rate, neither in good- nor in poor-prognosis patients.