Predictive factors for efficacy of capecitabine in heavily pretreated patients with metastatic breast cancer

The purpose of the present study is to evaluate what clinical factors affect the efficacy, time to treatment failure (TTF), and overall survival (OS) of oral capecitabine monotherapy in heavily pretreated patients with metastatic breast cancer (MBC). A total of 102 consecutive patients with MBC who had been administered capecitabine monotherapy between June 2003 and August 2004 were retrospectively reviewed. Capecitabine (828 mg/m(2)) was given twice daily for 3 weeks followed by a 1-week rest period; this was repeated every 4 weeks. We evaluated the potential clinical factors for TTF and OS, using univariate analysis (log-rank test) and the multivariate Cox regression model. Median follow-up was 16.9 months. A total of 100 patients (98%) had been pretreated with either anthracyclines or taxanes, and 81 patients (79%) with both anthracyclines and taxanes. Response rate was 17% and clinical benefit rate was 41%. Median TTF and OS were 4.9 and 24.3 months, respectively. Multivariate analysis demonstrated that no liver metastasis (P = 0.015), good performance status (P = 0.033), longer disease-free interval (P = 0.036), and hormone receptor-positive tumor (P = 0.038) were significant for TTF. No liver metastasis (P = 0.00012), objective response to capecitabine (P = 0.00084), and good performance status (P = 0.0011) were significant for OS. Capecitabine monotherapy is effective over the long term for heavily pretreated patients with MBC who have no liver metastasis, good performance status, longer disease-free interval, or hormone receptor-positive tumor. Patients who have no liver metastasis, who respond to capecitabine, or who have good performance status are expected to survive even longer.