Therapy of small breast cancer: a prospective study on 1036 patients with special emphasis on prognostic factors.

In 1983, The German Breast Cancer Study Group, sponsored by the Federal Ministry of Research and Technology, started a prospective multicenter trial on the treatment of early breast cancer pT1 pN0 M0. Treatment consisted of initial tumorectomy with microscopically free margins and lower axillary dissection. After conformation of a pT1 pN0-stage, additional treatment was either mastectomy or adjuvant radiotherapy (50 Gy in 25 fractions to the entire breast plus 12 Gy electron boost). In medially located tumors, the parasternal and supraclavicular area was also irradiated with 50 Gy. A randomization between both treatment modalities was initially planned but was not feasible and abandoned. Nearly all patients were treated according to their own choice. From November 1983 through December 1989, 1119 patients were recruited. Eighty-three were excluded from the protocol. Out of the remaining 1036 patients, 733 (71%) underwent breast preservation and 303 (29%) mastectomy. A detailed pathohistological examination of all tumorectomy specimens was performed in a pathologic reference center. Oncogen overexpression was evaluated by immunohistological detection of the transmembrane protein p-185 (corresponding to c-erb-B2) in 425 cases. After a median follow-up of 48 months, the frequency of local recurrences (4.7%), regional recurrences (1%), and distant metastases (5.4%) was the same in the breast preservation group and the mastectomy group. The 3-year disease-free survival was 90% after breast preservation and 88% after mastectomy (p = 0.21). In the breast preserving group, 24 patients with microscopically involved margins had a poorer disease-free survival than the study group (75% vs 90% after 3 years). The width of the margins had no impact on prognosis. Other prognostic factors in an univariate and multivariate analysis were tumor size and tumor grade. Age, menopausal status, hormone receptor status, histological tumor type, and treatment (mastectomy vs breast preservation) were not significant. P-185-expression was dependent on tumor grade and was the strongest prognostic factor in an univariate and multivariate analysis (p less than 0.001). The results emphasize the central role of tumor grade for prognosis and suggest the independent prognostic significance of the c-erb-B2 oncogen (corresponding to p-185) in pN0-patients.