Anthracycline could be essential for triple-negative breast cancer: A randomised phase II study by the Kanagawa Breast Oncology Group (KBOG) 1101.

It is important to determine whether anthracycline-containing regimens or taxane-containing regimens are more effective in individual patients. The present study compared the efficacy of six cycles of docetaxel and cyclophosphamide (TC6) with that of three cycles of 5-fluorouracil, epirubicin and cyclophosphamide followed by docetaxel (FEC-D) in Japanese patients with hormone receptor (HR)-negative breast cancer (BC) to identify subtypes requiring anthracycline treatment. The study included 103 patients with operable HR-negative BC. Of these patients 53 received FEC-D and 50 received TC6. The primary endpoint was pathological complete response (pCR). The secondary endpoints were safety, breast-conserving surgery, disease-free survival (DFS) and overall survival (OS). The predictive factors for each regimen were evaluated. Of the 103 patients, 97 completed the study (FEC-D, 50 patients; TC6, 47 patients). The pCR rate was higher with FEC-D (36%) than with TC6 (25.5%); however, the difference was not significant (P = 0.265). TC6 was safer than FEC-D, as the adverse events with docetaxel in the FEC-D regimen were similar to those with the TC6 regimen. Among patients with basal BC, the pCR rate was significantly higher with FEC-D (42.9%) than with TC6 (13.6%; P = 0.033). Among patients with triple-negative breast cancer (TNBC), the DFS and OS were significantly better with FEC-D than with TC6 (P = 0.016 and P = 0.034, respectively). TC6 was not as effective as FEC-D for treating HR-negative BC, as TC6 was not sufficient to treat TNBC, particularly the basal subtype. Our findings suggest that anthracyclines are better treatment options than taxanes for basal BC. • The pCR rate was higher with FEC-D than with TC6 (P = 0.265). • For basal BC, the pCR rate was higher with FEC-D than with TC6 (P = 0.033). • For TNBC, DFS and OS were better with FEC-D than with TC6 (P = 0.016 and P = 0.034). [ABSTRACT FROM AUTHOR]