Endocrine Responsiveness and Tailoring Adjuvant Therapy for Postmenopausal Lymph Node-Negative Breast Cancer: A Randomized Trial

Background: The role of adjuvant chemotherapy in postmenopausal patients with lymph node-negative breast cancer is controversial. After demonstrating the efficacy of chemotherapy combined with tamoxifen for postmenopausal patients with lymph node-positive disease, the International Breast Cancer Study Group launched a randomized trial (Trial IX) to evaluate the role of adjuvant chemotherapy preceding treatment with tamoxifen for patients with lymph node-negative disease. Methods: After stratification by estrogen receptor (ER) status, patients were randomly assigned to receive three 28-day courses of “classical” adjuvant CMF chemotherapy (cyclophosphamide at 100 mg/m2 on days 1–14, orally; methotrexate at 40 mg/m2 on days 1 and 8, intravenously; and 5-fluorouracil at 600 mg/m2 on days 1 and 8, intravenously) followed by tamoxifen (20 mg/day, orally for 57 months) (CMF→tamoxifen) or to receive tamoxifen alone (20 mg/day, orally for 60 months). We enrolled 1669 eligible patients, 382 (23%) with ER-negative tumors, 1217 (73%) with ER-positive tumors, and 70 (4%) with unknown ER status. The median follow-up was 71 months. All statistical tests were two-sided. Results: The added benefit of CMF followed by tamoxifen over tamoxifen alone was statistically significantly dependent on ER status (tests for interaction: P = .01 for disease-free survival [DFS] and P = .07 for overall survival [OS]). For patients with ER-negative tumors, the addition of CMF statistically significantly improved DFS (5-year DFS = 84% for CMF→tamoxifen versus 69% for tamoxifen alone; difference = 15%; 95% confidence interval [CI] = 6% to 24%; risk ratio [RR] = 0.52; 95% CI = 0.34 to 0.79; P = .003) and OS (5-year OS = 89% for CMF→tamoxifen versus 81% for tamoxifen alone; difference = 8%; 95% CI = 0% to 16%; RR = 0.51; 95% CI = 0.30 to 0.87; P = .01). By contrast, for patients with ER-positive tumors, addition of CMF provided no benefit in terms of DFS (5-year DFS = 84% for CMF→tamoxifen versus 85% for tamoxifen alone; difference = –1; 95% CI = –6% to 4%; RR = 0.99; 95% CI = 0.75 to 1.30; P = .92) or OS (5-year OS = 95% for CMF→tamoxifen versus 93% for tamoxifen alone; difference = 2%; 95% CI = –1% to 5%; RR = 0.95; 95% CI = 0.64 to 1.40; P = .80). Conclusions: Postmenopausal patients with lymph node-negative breast cancer benefited substantially from adjuvant chemotherapy if their cancer was ER-negative (i.e., endocrine-nonresponsive). In contrast, if their cancer was ER-positive (i.e., end