In Reply We appreciate Armbruster et al and Yu et al for their interest and comments regarding our A-PLUS study. 1 Armbruster et al point out that there is an alternative way to illustrate the benefit of the bevacizumab, cisplatin, and etoposide (BEEP) regimen in the control of patients with breast cancer brain metastasis (BCBM) by using restricted mean survival time (RMST). This is reassuring. When we applied RMST with the time period truncated to 12 months to the masked central review survival outcome, we observed significant benefit of the BEEP in terms of brain-specific progression-free survival (8.70 vs 7.29 months, P = .03), demonstrating the advantage of RMST in circumstances where the proportional hazard assumptions may not be valid.As Yu et al advocated, we agree that the presence and status of extracranial disease is a major factor in determining the survival of patients with BCBM. In the study population, 1 the distribution of extracentral nervous system (CNS), including liver (62% vs 50%), lung (62% vs 53%), and bones (63% vs 53%), were not significantly different between the 2 study arms. In the control arm, all patients received systemic treatment immediately after whole-brain radiotherapy (WBRT). For both arms, after systemic treatments (eTable 5 in Supplement 1 1 ), there were no significant differences in extra-CNS objective response rate (ORR) between the 2 arms (Table 2), 1 and thus the influence of extra-CNS tumor control on BCBM should be minimal. Yu et al were also concerned about the differences in the median time to relapse (15.7 vs 6.8 months) in the treatment arms. However, if we separate the patient groups into de novo stage IV disease (44.6% vs 44.7%), time from primary to metastatic interval less than 3 years (27.0% vs 26.3%), and time from primary to metastatic interval greater than 3 years (28.4% vs 29.0%), we then can see that there were no significant differences between the treatment arms. We also believe that bevacizumab-induced vascular normalization could enhance the effect of WBRT in controlling micrometastatic tumors in the brain 2 -as demonstrated by both the significantly longer CNS-specific PFS and higher 8-month CNS-specific PFS rate in our study, 1 which could not be assessed solely by CNS-specific ORR that was based on measurable lesions.As both Yu et al and Armbruster et al mentioned, the cognitive function effect should be addressed after WBRT, not to mention that in our study, 1 WBRT was carried out without hippoc ampal avoidance or memantine concomitantly. Nevertheless, the A-PLUS study was designed in 2013, at a time when the randomized clinical trial of memantine for WBRT was just published. 3 Memantine was not indicated nor reimbursed for protection of cognitive function during WBRT in Taiwan. The evolution of clinical practices since then highlights the importance of incorporating cognitive preservation strategies in the treatment of BCBM. Although long-term cognition assessment was not implemented in our protocol, all patients in both arms required stable neur...