Maintenance Treatment With Capecitabine Versus Observation in Breast Cancer Patients

Patients will be stratified as per investigational site, previous adjuvant chemotherapy (anthracyclines versus anthracyclines plus taxanes), and number of affected axillary lymph nodes (0, 1-3, \>= 4). Node negative patients must present a tumour size \> 2 cm to be eligible. At least 6 lymph nodes must be analysed to confirm the number of affected nodes. Patients will be randomised to receive: 8 courses of capecitabine 1000 mg/m2 by mouth, twice a day (p.o. bid) for 14 days, followed by a 7 day rest versus observation. Tissue samples must be analysed by a central laboratory, to confirm estrogen receptor (ER), progesterone receptor (PgR), human epidermal growth factor receptor-2 (HER2), cytokeratins (CK) 5/6 and epidermal growth factor receptor (EGFR) status. The following data were obtained from the database of the "El Alamo" project. One thousand six hundred and twenty-seven (1,627) in total were considered during the years 1990 to 1997. The population is formed of patients with operable breast cancer, with surgery, positive nodes, and negative hormone receptors, or negative nodes, negative hormone receptors and T2-3 tumors. For these patient groups, estimated 5-year disease-free survival is 64.72%. Assuming an exponential distribution, the aim is to detect an increase of 64.72% to 73.7% in 5 years Disease Free survival rate corresponds to a Hazard Ratio of 0.701 and a risk reduction of about 30%, with a power of 80% using a two-tailed log-rank test at 0.05 and whereas 4 years of recruitment period and 3 years of follow-up period. We would need 255 events, 834 patients without considering any dropouts. Considering a drop-out rate of 5% post-randomization, the final sample size will be 876 patients, 438 per treatment arm. The sample size calculation was performed by the program package "EAST" version 5.2.