Tamoxifen versus surgery plus Tamoxifen as primary treatment for elderly patients with breast cancer: combined data from the \u201cGRETA\u201d and \u201cCRC\u201d trials

Success of tamoxifen in controlling advanced breast cancer and prolonging life in postmenopausal women led to the supposition that tamoxifen alone might control breast cancer in elderly women and obviate the need for surgery. Similar trials were independently conducted in Italy and the UK randomizing elderly patients (70 years and older) with operable breast cancer to tamoxifen alone or tamoxifen plus breast surgery. The italian GRETA trial recruited 466 patients and the CRC trial in the UK accrued 447. These trials were merged in a single database and results were analyzed. Combined median follow-up is 5 years and median age is 76. The majority of tumors were T2 in both trials. Although there were more T1 tumors in the GRETA trial (42% versus 24% in CRC) mastectomy was more common in the Italian trial (84% versus 30%). Tamoxifen schedules were different: the GRETA used a one-day 160 mg loading dose on the non-surgical arm and subsequent 20 mg daily applied to both arms. The CRC trial used a daily dose of 40 mg on both arms. Overall survival did not differ significantly according to treatment. In a proportional hazards model, neither treatment nor group are significant at the 0.05 level. The hazard ratio for Surgery is 0.86 (95% CI 0.71-1.03). The risk of death attributed to breast cancer, however, is significantly reduced by Surgery with hazard ratio 0.70 (95% CI 0.51-0.95). In multivariate analysis, age and tumor size are strongly predictive of survival. Age determines survival in later years (over 74). In those under 75 initial Surgery determines survival, with hazard ratio 0.66 (95% CI 0.48-0.93). In this subset overall mortality is reduced by one third. The overall survival of the whole population is not affected by Surgery. Only tumor size is strongly predictive of death attributed to breast cancer