A multicenter phase III randomized study comparing the efficacies of two adjuvant polychemotherapeutic regimens in 145 patients with stage II node-positive breast cancer: the standard chemotherapy combination, CMF (cyclophosphamide, methotrexate, 5-fluorouracil), and an experimental protocol, CNF (cyclophosphamide, mitoxantrone [Novantrone], 5-fluorouracil) in which mitoxantrone replaced methotrexate. The finding of a significant advantage ( p= 0.04) in the disease-free survival for those receiving mitoxantrone (mean survival 4.4 years for CNF versus 2.7 years for CMF) led the authors to break the data down in subpopulations to determine exactly which groups of women responded more favorably to CNF than CMF. An advantage in disease-free survival was found, most notable in four subgroups: Sephardic women, women less than 45 years of age, premenopausal women, and women with 4 to 10 positive axillary lymph nodes. Although the small numbers of women in each of these subgroups rule out drawing definitive conclusions, the trend merits further study to confirm these observations.