Objective A newly designed combination chemotherapy for multiple myeloma, MMCP [ranimustine (MCNU), melphalan (MPH), cyclophosphamide (CPM) and prednisolone (PSL)], was analyzed and compared with the results of our previous randomized trial of VMCP [vincristine, MPH, CPM and PSL] and MMPP [MCNU, MPH, procarbazine and PSL]. Methods MCNU (33.3 mg/m2, div) on day 1 and MPH (4 mg/m2, po), CPM (66.7 mg/m2, po) and PSL (30 mg/m2, po) from day 1 to 4, were administered. Each cycle was repeated every 3 weeks. Patients or materials From January 1991 until August 1995, 104 patients with multiple myeloma diagnosed at 10 hospitals of Nagoya Cooperative Study Group were enrolled.Results Of the 87 evaluable patients, partial response rate for MMCP was 65.5% and was significantly higher than that ofVMCP (13/47=27.7%, p<0.0001) and that of MMPP (21/47=44.7 % , p=0.0196). A plateau attainment was observed in 49.4%. The percentage of the patients who attained plateau was significantly increased in the MMCP arm than in the VMCP arm (19.1%, p=0.0017) but was not in comparison with that of MMPP arm (42.6%, p=0.6790). Patients treated with MMCP survived significantly longer than those treated with VMCP or MMPP (p=0.0009 by generalized Wilcoxon test, p=0.0023 by log-rank test) with median survival for MMCP being 31.6 months, for VMCP 22.5 months, and for MMPP 22.9 months. No significant differences were observed with respect to adverse effects among the three regimens.Conclusion The newly designed MMCP is a candidate as an induction chemotherapy for multiple myeloma. (Internal Medicine 41: 290-294, 2002)