A phase II trial in metastatic breast cancer (MBC) (NO16853) failed to show noninferiority (progression-free survival, PFS) of capecitabine 825 mg/m(2) plus docetaxel 75 mg/m(2) to the registered capecitabine dose of 1,250 mg/m(2) plus docetaxel 75 mg/m(2). We developed a modeling framework based on NO16853 and the pivotal phase III MBC study, SO14999, to characterize the link between capecitabine dose, tumor growth, PFS, and survival to simulate response to a range of capecitabine doses and determine a minimum capecitabine dose noninferior to 1,250 mg/m(2). Simulation showed NO16853 had little power to demonstrate noninferiority (69%). The power reached 80% with a 1,000 mg/m(2) starting dose and an increased number of PFS events. A starting dose of 1,000 mg/m(2) could be established as noninferior in terms of efficacy to the registered dose in the second-line MBC setting, with a potentially improved safety, in line with medical practice.CPT: Pharmacometrics & Systems Pharmacology (2012) 1, e19; doi:10.1038/psp.2012.20; advance online publication 26 December 2012.