Aromatase inhibitors are superior to tamoxifen as adjuvant therapy in postmenopausal patients with hormone-responsive breast cancer. We report the follow-up efficacy results from the N-SAS BC 03 trial (UMIN CTRID: C000000056) where anastrozole was compared with tamoxifen as adjuvant therapy in postmenopausal Japanese patients with hormone-responsive early breast cancer. The full analysis set contained 696 patients (anastrozole arm, n = 345; tamoxifen arm, n = 351). The log-rank test was used to compare the two groups in terms of disease-free survival (DFS) and relapse-free survival (RFS); Kaplan-Meier estimates were calculated. The treatment effects were estimated by Cox's proportional hazards model. To examine time-varying effect of hazard ratios, we estimated time-varying hazard ratios at time t [HR(t)] using data from time t up to 12 months. After a median follow-up of 98.5 months, hazard ratios (95% CIs) were 0.90 (0.65-1.24; log-rank p = 0.526) for DFS and 0.83 (0.56-1.23; log-rank p = 0.344) for RFS. Hazard ratios (95% CIs) for DFS and RFS up to 36 months were 0.69 (0.40-1.17) and 0.54 (0.27-1.06) and those after 36 months were 1.06 (0.70-1.59) and 1.05 (0.64-1.73), respectively. Time-varying hazard ratios for both DFS and RFS showed that hazard ratios were initially in favor of anastrozole and approached 1.0 at around 36 months. Superior efficacy of anastrozole to tamoxifen suggested by the initial analysis was not confirmed in the present analysis after a long-term follow-up period. Advantage of anastrozole was the greatest immediately after switching from tamoxifen and then decreased thereafter.