BACKGROUND CHOP is a four-drug, first-generation combination chemotherapy regimen that has cured approximately 30% of all patients with advanced stages of intermediate- or high-grade non-Hodgkin's lymphoma in national cooperative group trials. Initial single-institution studies of third-generation regimens such as m-BACOD, ProMACE-CytaBOM, and MACOP-B suggested that 55%-60% of these patients might be cured. METHODS In order to make a valid comparison between these regimens, the Southwest Oncology Group and the Eastern Cooperative Oncology Group initiated a randomized phase III comparison of CHOP vs. m-BACOD vs. ProMACE-CytaBOM vs. MACOP-B. Endpoints of the study were response rate, time to treatment failure, overall survival, and severe or life-threatening toxicity. Received dose intensity calculations and subset analyses were also performed. RESULTS 1138 patients were registered on this clinical trial. Each treatment arm contained between 218 and 233 eligible patients. Known prognostic factors were equally distributed. The initial results of this study were recently published. There were no significant differences in either the partial or complete response rates between treatment arms. Now after a median follow-up of 49 months and a maximum follow-up of 84 months, there is still no difference in time to treatment failure (p = 0.40) or overall survival (p = 0.68). No subset of patients was found to have significantly improved survival with the third-generation regimens. The received dose intensity data were comparable to previously published data for these regimens. Fatal toxicity was 1% for CHOP, 3% for ProMACE-CytaBOM, 5% for m-BACOD, and 6% for MACOP-B. CONCLUSIONS Based on similar failure-free and overall survival with lower cost and lower severe toxicity, CHOP remains the standard chemotherapy for patients with advanced-stage, intermediate- or high-grade non-Hodgkin's lymphoma. New treatment strategies need to be developed to improve the prognosis of these patients.