ADEQUATE LOCOREGIONAL TREATMENT FOR EARLY BREAST-CANCER MAY PREVENT SECONDARY DISSEMINATION

Purpose: To analyze different events that determine event-free surviva l (EFS) in a randomized trial on adjuvant radiotherapy in early breast cancer patients with more than 15 years of follow-up evaluation. Pati ents and Methods: The trial included 960 patients with a unilateral, o perable breast cancer, Surgery consisted of a modified radical mastect omy. The trial compared three arms, as follows: preoperative radiother apy, postoperative radiotherapy, and no adjuvant treatment, Events wer e analyzed by a competing-risk approach. A proportional hazards multip le regression model was used to analyze the effects of radiotherapy on the risk of distant metastasis. Similar analyses were performed separ ately for node-negative [N(-)] and node-positive [N(+)] patients in th e two groups that did not include preoperative radiotherapy.Results: R adiotherapy produced a fivefold decrease of the risk of local recurren ce (P < .0001), In N(+) patients, postoperative radiotherapy decreased the risk of distant dissemination (relative risk, 0.63). When local r ecurrence was introduced in the model as a time-dependent covariate, t his factor wets predictive of distant dissemination (P < .0001) and nu llified the effect of postoperative radiotherapy. This finding suggest s that the decrease of distant metastases was related to the preventio n of local recurrence. A similar effect was found in models that used overall survival as an end point. Conclusion: This study shows that po stmastectomy radiotherapy in N(+) breast cancer patients may decrease the distant metastasis rate by preventing local recurrences and thus a voiding secondary dissemination. (C) 1995 by American Society of Clini cal Oncology.