RESUMO
PURPOSE: The aim of the present analysis is to determine the long-term results in terms of breast relapse and specific survival in patients treated with conserving surgery and adjuvant treatment for early breast cancer. METHODS: From January 1980 to December 2001, 3834 patients with pT1-T2 breast cancer were treated consecutively at the University of Florence. The median age of the patient population was 55 years (range 30-80). All patients were followed for a median of 7.4 years (range 0.6 year to 22.5 years). The crude probability of survival (or local recurrence) was estimated by using Kaplan-Meier method, and survival (or local recurrence) comparisons were carried out using Cox proportional hazard regression models. RESULTS: The Cox regression model by stepwise selection showed some parameters, such as chemotherapy (HR 1.53; CI 1.19-1.95), pT status (HR 1.62, CI 1.31-2.01), positive axillary lymph nodes (HR 1.92, CI 1.66-2.22), and local recurrence (HR 4.58; CI 3.66-5.73), as independent prognostic factors for breast cancer death. Moreover, we found lower rate survival among patients treated before 1991 in comparison to women treated after 1991 (p=0.0001) probably due to inadequate treatment. For local disease free survival, age at presentation (HR 0.47; CI 0.35-0.63), use of tamoxifen (HR 0.42; CI 0.25-0.71), surgical margins (HR 2.00; CI 1.21-3.30), and chemotherapy (HR 0.53; CI 0.31-0.91) emerged by multivariate analyses as significant breast relapse predictors. CONCLUSION: In our experience breast conserving surgery followed by adjuvant radiotherapy treatment gives high rates of local control in women with early breast cancer. The use of routinely adjuvant chemotherapy and hormone therapy lowered the local recurrence and probably the modification of therapeutic approach in the last decades also improved the specific survival.