Local F ailure I s R esponsible f or t he D ecrease i n S urvival for P atients W ith B reast C ancer T reated W ith C onservative Surgery a nd P ostoperative R adiotherapy

Purpose: The aim of the present study was to evaluate the role of local failure (LF) in the survival of patients treated with lumpectomy and postoperative radiotherapy and to investigate whether LF is not only a marker for distant metastasis (DM) but also a cause. Methods: Charts of patients treated with breast conservative surgery between 1969 and 1991 were reviewed retrospectively. There were 2,030 patients available for analysis. The median duration of follow-up was 6 years. A Cox regression multivariate analysis was performed using LF as a time-dependent covariate. Results: Local control (LC) was 87% at 10 years. Local failure led to poorer survival at 10 years than local control (55% v 75%, P F .00). In a Cox model, local failure was a powerful predictor of mortality. The relative risk associated with LF was 3.6 for mortality and 5.1 for DM (P F .00). In patients with LF, the rate of DM peaked at 5 to 6 years, whereas it peaked at 2 years for patients with LC. The mean time between surgery and DM was 1,050 days for patients without LF and 1,650 days for patients with LF (P F .00). Conclusion: Our results show that local failure is associated with an increase in mortality. The difference in the time distribution of distant metastasis for LF and LC could imply distinct mechanisms of dissemination. Local failure should be considered not only as a marker of occult circulating distant metastases but also as a source for new distant metastases and subsequent mortality. J Clin Oncol 17:101-109. r 1999 by American Society of Clinical Oncology. T IS OF UTMOST importance to understand the significance of local failure (LF) in breast cancer treated with local excision and radiotherapy, even though it occurs in only 13% 1 of patients at 10 years. Considering that two thirds of the 182,000 women with breast cancer in the United States 1 are treated with a breast-conservative approach, this translates into 15,600 women with LF every year in this country. This number is greater than the total number of invasive carcinomas of the cervix or Hodgkin’s disease. If local failure is responsible for any increase in distant metastasis (DM) and subsequent mortality, its cause and consequences should be carefully analyzed. The impact of local failure on ultimate outcome in breast cancer treated with conservative surgery and postoperative radiotherapy remains controversial. For many investigators, local failure after conservative treatment has no detrimental effect on the survival of patients because breast cancer is a systemic disease that is already disseminated even before diagnosis. 2-5 Some risk factors seem to predict both local and distant recurrences, whereas others suggest an increased risk of local recurrence and seem to have little effect on the risk of metastatic disease. Overall, the relation between LF and the risk of systemic metastasis is poorly understood. 6 Some authors 7-9 have demonstrated that local failure is associated with more chances of distant metastasis, and Whelan et al 10 have shown that LF is associated with an increased risk of distant relapse and mortality. Fisher et al 7 also showed that patients with local failure have more distant metastases than patients without local failure, but they considered LF to be a marker for distant metastasis,