Tailored endpoints: A proposal for design of future clinical trials in metastatic breast cancer (MBC).

Abstract

e13058 Background: Widespread access to active agents against MBC has produced significant improvements in survival, even though few phase III trials are powered to detect overall survival (OS) effects. However, regulatory agencies hold OS as gold standard for approval of new drugs. OS can be portioned into the sum of progression-free survival (PFS) and survival post-progression (SPP). Recent data suggest that OS is a reasonable primary endpoint only when median SPP is short. On the other hand, patients enrolled in first-line trials could have a long life-expectancy. Aim of this study was to evaluate PFS, OS and SPP in a consecutive series of patients with MBC in order to obtain information as a basis for design of future clinical trials. Methods: Four hundred patients with MBC diagnosed from January 2004 to February 2011 at University Hospital of Udine, Italy, were included in the study. We examined the role of potential of disease and patients’ characteristics in influencing the different measures of outcome. Results: Median OS was 33.71 months (mo), in line with the best literature. Median PFS at first-line (PFS1) was 9.33 mo, with linear decrease at second (5.06), third (3.58), and fourth (3.19) line, respectively. Median SPP after first-line (SPP1) was 18.69 mo. Interestingly, differences in outcome were noticed according to immunophenotype. The best prognosis was observed in luminal A disease (OS= 46.72 mo, PFS1= 15.61 mo, SPP1= 25.43 mo). In luminal B disease, median OS was 27.66 mo, PFS1 8.94 mo, and SPP1 13.21 mo, respectively. In patients with HER2-positive disease, mainly treated with anti-HER2 therapy, outcome approached that of luminal A disease (OS= 41.10 mo, PFS= 9.89 mo, SPP1 18.69 mo). Patients with triple negative disease (TNBC) experienced the worst prognosis (OS= 8.54 mo, PFS1= 4.04 mo, SPP1= 2.83 mo). Conclusions: The study demonstrated different results in the measures of outcome according to the line of treatment and specific disease characteristics. As a consequence, endpoints of clinical trials should be tailored taking into consideration prognostic/predictive factors as well as the line of treatment.