Efficacy of Exemestane in Korean Patients with Metastatic Breast Cancer after Failure of Nonsteroidal Aromatase Inhibitors / 한국유방암학회지

PURPOSE: Exemestane has shown good efficacy and tolerability in postmenopausal women with hormone receptor-positive metastatic breast cancer. However, clinical outcomes in Korean patients have not yet been reported. METHODS: Data on 112 postmenopausal women with metastatic breast cancer were obtained retrospectively. Clinicopathological characteristics and treatment history were extracted from medical records. All patients received 25 mg exemestane daily until objective disease progression. Progression-free survival (PFS) was the primary endpoint, and secondary endpoints were overall survival (OS), objective response rate (ORR), and clinical benefit rate (CBR=complete response+partial response+stable disease for 6 months). RESULTS: The median age of the subjects was 55 years (range, 28-76 years). Exemestane treatment resulted in a median PFS of 5.7 months (95% confidence interval [CI], 4.4-7.0 months) and median OS of 21.9 months (95% CI, 13.6-30.3 months). ORR was 6.4% and CBR was 46.4% for the 110 patients with evaluable lesions. Symptomatic visceral disease was independently associated with shorter PFS (hazard ratio, 3.611; 95% CI, 1.904-6.848; p<0.001), compared with bone-dominant disease in a multivariate analysis of PFS after adjusting for age, hormone receptor, human epidermal growth factor receptor 2, Ki-67 status, dominant metastasis site, and sensitivity to nonsteroidal aromatase inhibitor (AI) treatment. Sensitivity to previous nonsteroidal AI treatment was not associated with PFS, suggesting no cross-resistance between exemestane and nonsteroidal AIs. CONCLUSION: Exemestane was effective in postmenopausal Korean women with hormone receptor-positive metastatic breast cancer who failed previous nonsteroidal AI treatment.

An improved cluster labeling method for support vector clustering.

The support vector clustering (SVC) algorithm is a recently emerged unsupervised learning method inspired by support vector machines. One key step involved in the SVC algorithm is the cluster assignment of each data point. A new cluster labeling method for SVC is developed based on some invariant topological properties of a trained kernel radius function. Benchmark results show that the proposed method outperforms previously reported labeling techniques.