Search for a command to run...
Six hundred nineteen patients with metastatic breast cancer, treated with a combination of 5-fluorouracil, Adriamycin, and cyclophosphamide, or close variations of this program, with or without immunotherapy were analyzed retrospectively to identify those host, tumor, or treatment characteristics that might be of prognostic importance in predicting response to chemotherapy and survival from onset of the 5-fluorouracil-Adriamycin-cyclophosphamide treatments. Primary tumor characteristics such as size of primary, number of axillary nodes involved, stage at diagnosis, and type of surgery used for primary treatment were not found to be of prognostic significance. Host characteristics such as age, menstrual status, or family history of breast cancer were similarly unrelated to outcome. Non-Caucasian patients had a lower response rate and somewhat shorter survival than did Caucasians. Pretreatment weight loss, poor performance status, and abnormal biochemical and hematological values were of adverse prognostic significance. An estimate of total extent of disease based on criteria for rating extent of involvement at 12 potential sites was a much more important prognostic factor related to response and survival than actual sites of involvement or the traditional "dominant site" classification. There was a trend, however, for patients with bone involvement to have a longer survival than did patients with metastases to other organ sites. Shorter survival times were observed among patients exposed to extensive prior radiotherapy and those who failed to respond to prior hormonal treatment. The prognostic variables identified in this paper should be used for the design and comparison of clinical trials in the future.