The potential risk of neoadjuvant chemotherapy in breast cancer patients--results from a prospective randomized trial of the Austrian Breast and Colorectal Cancer Study Group (ABCSG-07).

PURPOSE: To evaluate the impact that pre- and postoperatively administered chemotherapy with cyclophosphamide, methotrexate and fluorouracil (CMF) and postoperative chemotherapy vs. postoperative chemotherapy alone have on long-term prognosis.

PATIENTS AND METHODS: The ABCSG conducted a nationwide randomized phase III trial in high-risk endocrine non-responsive breast cancer patients comparing pre- and postoperative chemotherapy containing CMF as preoperative treatment vs. postoperative chemotherapy alone between 1991 and 1999. From 1996 the ABCSG-07 protocol was amended to also allow randomization of high-risk endocrine-responsive patients. Of 423 eligible patients with high-risk primary breast cancer, 203 patients were randomly assigned to preoperatively receive three cycles of CMF (cyclophosphamide, methotrexate, fluorouracil; 600/40/600 mg/m(2)) intravenously on day 1 and 8, while 195 patients received postoperative chemotherapy alone. In both groups, three cycles of CMF were given initially, and another three cycles of CMF were administered in node-negative patients, whereas node-positive patients received three cycles of EC (epirubicin, cyclophosphamide; 70/600 mg/m(2)).

RESULTS: Overall response rate to preoperative chemotherapy with three cycles of CMF was 56.2%; complete pathological response was achieved in 12 patients (5.9%). Recurrence-free survival was significantly better in patients receiving chemotherapy postoperatively (HR 0.7, 0.515-0.955; P = 0.024). No survival difference was observed between the two therapy groups (HR 0.800, 0.563-1.136; P = 0.213).

DISCUSSION: Preoperative chemotherapy with CMF has to be considered as insufficient in high-risk breast cancer patients. Delayed surgery and anthracycline-based chemotherapy result in shorter recurrence-free survival but not overall survival.