Background: The National Comprehensive Cancer Network (NCCN) guidelines for treatment of breast cancer recommend several adjuvant chemotherapy regimens, including sequential or concurrent anthracycline-cyclophosphamide and taxane-based regimens (sequential AC-T or concurrent ACT, respectively); anthracycline alone; cyclophosphamide, methotrexate, and 5-fluorouracil (CMF); and docetaxel and cyclophosphamide (TC). Sequential AC-T is the most commonly prescribed standard regimen. Two types of regimens without anthracycline, TC and platinum-containing regimens, have been speculated to have similar efficacy to that of ACT-based regimens. Platinum-containing regimens have also demonstrated high efficacy in breast cancer with tolerable adverse effects in several previous studies. However, because of the limitations of conventional meta-analysis, we do not know how these regimens compare with one another in terms of survival or adverse events. Thus, the best adjuvant chemotherapy regimen for breast cancer is not known. Bayesian network meta-analysis allows comparison of treatments for which there has been no head-to-head comparison by using indirect comparisons. Methods: We searched the MEDLINE, EMBASE, and Cochrane Library databases for articles published before January 2014; the American Society of Clinical Oncology annual meeting abstracts for 1983-2013; and the American Association for Cancer Research annual meeting abstracts for 1916-2013. We included only randomized controlled trials of adjuvant treatments for breast cancer that compared 2 or more of the following: observation alone; sequential AC-T; concurrent ACT; anthracycline alone; CMF; TC; and platinum-containing regimens. We compared regimens by using a network meta-analysis approach and random-effects models. Network meta-analysis allows derivation of hazard ratios (HRs) for death for each regimen and comparison of these HRs with each other even when there are no direct comparisons between 2 regimens. Results: We identified 24 trials with a total of 28,853 patients. Network meta-analysis showed that OS with TC and platinum-containing regimens were similar to OS with sequential AC-T (TC: HR=0.94; 95% CI, 0.58-1.52; platinum: HR=0.90; 95% CI, 0.69-1.19). Patients receiving CMF or anthracycline alone had significantly worse OS than those with sequential AC-T (CMF: hazard ratio [HR]=1.62; 95% CI, 1.31-2.01; anthracycline: HR=1.23; 95% CI, 1.07-1.43). For overall adverse events, the mean number of adverse events per patient was higher for platinum-containing regimens (2.1) and concurrent ACT (1.22) than for sequential AC-T (1.17). The mean number of hematological adverse events was higher for concurrent ACT (0.67) and TC (0.63), than for sequential AC-T (0.48). The mean number of nonhematological adverse events was higher for platinum-containing regimens (1.76) than for sequential AT (0.7). Conclusion: Our findings suggest that sequential AC-T should continue to be considered the optimal treatment for early-stage breast cancer because of the equivalent survival benefit of concurrent ACT, TC and platinum-containing regiments but superior adverse-event outcomes.

Citation Format: Takeo Fujii, Fanny Le Du, Lianchun Xiao, Takahiro Kogawa, Vicente Valero, Yu Shen, Naoto T Ueno. Identification of optimal adjuvant chemotherapy regimen for early-stage breast cancer: A systematic review and bayesian network meta-analysis [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr P3-09-05.