The sequential doxorubicin --> CMF (CMF=cyclophosphamide, methotrexate, fluorouracil) regimen has never been compared to CMF in a randomised trial. The role of adding goserelin and tamoxifen after chemotherapy is unclear. In all, 466 premenopausal node-positive patients were randomised to: (a) CMF x 6 cycles (CMF); (b) doxorubicin x 4 cycles followed by CMF x 6 cycles (A --> CMF); (c) CMF x 6 cycles followed by goserelin plus tamoxifen x 2 years (CMF --> GT); and (d) doxorubicin x 4 cycles followed by CMF x 6 cycles followed by goserelin plus tamoxifen x 2 years (A --> CMF --> GT). The study used a 2 x 2 factorial experimental design to assess: (1) the effect of the chemotherapy regimens (CMF vs A --> CMF or arms a+c vs b+d) and (2) the effect of adding GT after chemotherapy (arms a+b vs c+d). At a median follow-up of 72 months, A --> CMF as compared to CMF significantly improved disease-free survival (DFS) with a multivariate hazard ratio (HR)=0.740 (95% confidence interval (CI): 0.556-0.986; P=0.040) and produced a nonsignificant improvement of overall survival (OS) (HR=0.764; 95% CI: 0.489-1.193). The addition of GT after chemotherapy significantly improved DFS (HR=0.74; 95% CI: 0.555-0.987; P=0.040), with a nonsignificant improvement of OS (HR=0.84; 95% CI: 0.54-1.32). A --> CMF is superior to CMF. Adding GT after chemotherapy is beneficial for premenopausal node-positive patients.