Background: Capecitabine (X) monotherapy is an effective and well-tolerated treatment for locally advanced or metastatic breast cancer (LABC/MBC), showing consistently high activity in both the 1st-line setting andpretreated disease. The registered dose is 1250 mg/m2 bid, d1-14 q21d, but there is considerable variation in the administration of X (dose, line, monotherapy vs combination) in everyday practice. Patients and methods: ELIXIR is a French, multicenter, prospective, observational cohort study assessing the use of X in routine oncology practice in patients (pts) with LABC/MBC. The observation period is 24 months.

Results: Between Dec 05 and Jan 08, 668 pts were enrolled, of whom 655 were evaluable. The majority of pts (540; 82%) received X as monotherapy (1st line: n=201; 2nd line: n=205; ≥3rd line: n=134). The remaining 115 pts received X combined with chemotherapy (1st line: n=60; 2nd line: n=36; ≥3rd line: n=19), typically vinorelbine (n=66) or a taxane (n=42). Overall, 57 pts (9%) received X in combination with trastuzumab. Among pts receiving X monotherapy, only 21 % began therapy at the full registered dose, 60% of pts started at 75% of the registered dose and in 19% of pts, therapy was initiated at ≥50% of the registered dose. At the 12-month analysis, median progression-free survival (PFS) in the overall population was 8.0 months (95% CI 7.1-8.8), 61% of pts were still alive, and median overall survival had not been reached. Analysis of safety data among pts receiving X monotherapy revealed no major differences in tolerability according to treatment line, except for less grade 3 hand-foot syndrome (HFS) and grade 3/4 diarrhea in pts treated in later lines (Table).

Conclusion: ELIXIR is one of the largest prospective cohort studies of X in LABC/MBC. Tolerability in the 1st-line setting compares favorably with that of X 1000 or 1250 mg/m2 bid observed in the exclusively 1st-line GBG39 (MoniCa) and PELICAN trials, presumably reflecting the selection of a lower dose in most pts treated in ELIXIR. Despite the tailored dose, median PFS of 8.0 months is consistent with the median time to progression of 7.1-7.3 months reported in GBG39 and PELICAN. These data suggest that the efficacy of X seen in clinical trials can be reproduced in the real-life setting, despite widespread use of lower doses to improve tolerability. The wide array of variables in routine clinical practice makes interpretation of the data complex. Slight qualitative differences in the safety profile between treatment lines may be attributable to differences in starting dose and/or disease symptoms and require further investigation. Analysis of efficacy and safety results according to treatment line and starting dose is ongoing and 24-month results will be presented.

Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P6-11-04.