Surrogacy of tumor response and progression-free survival for overall survival in metastatic breast cancer resistant to both anthracyclines and taxanes

In breast cancer, the validity of surrogate endpoints for overall survival (OS) is a matter of controversy. In order to generate a hypothesis, we evaluated whether tumor response or progression-free survival (PFS) could be valid surrogates for OS in patients with metastatic breast cancer. Data from 30 patients were available from a phase II study of trastuzumab and capecitabine in human epidermal growth factor receptor 2-overexpressing metastatic breast cancer resistant to both anthracyclines and taxanes. The proportional hazards (PH) model was applied to evaluate the relationship between OS and tumor response or PFS. In addition, to explore prognostic factors influencing OS or post-progression survival, the PH model with a stepwise regression procedure was applied. The relationship between tumor response and PFS was highly significant (P = 0.0036); however, there was no significant relationship between tumor response and OS or between PFS and OS. In the multivariate analysis, the sum of the longest diameter of target lesions (P = 0.0011), neutrophil count (P = 0.0033), and creatinine (P = 0.0085) were statistically significantly associated with OS. We generated a hypothesis that neither PFS nor tumor response were valid as surrogate endpoints for OS, at least in the phase II trial for metastatic breast cancer resistant to both anthracyclines and taxanes. We also found that the sum of the longest diameter of target lesions, neutrophil count, and creatinine were prognostic factors for OS.