A high incidence of anti-thyroid antibodies (TAb) has been found in patients with breast cancer (BC). The aim of this study was to evaluate the prognostic value of TAb in a group of 47 women submitted to mastectomy for high malignancy degree BC. All patients were evaluated for thyroid disorders after breast surgery and before any anti-tumoral adjuvant therapy. Five yr after BC diagnosis 31/47 (65.9%) patients were alive (survivors group: SG) and 16/47 (34.1%) were dead (deaths group: DG). The overall prevalence of TAb was 15/47 (31.9%): 14/31 (45.1%) in SG and 1/16 (6.2%) in DG (p=0.008). Five-yr mortality was 15/32 (46.9%) in TAb- and 1/15 (6.7%) in TAb+ patients (p=0.01). Eight out of 47 (17.0%) patients had Hashimoto's thyroiditis and 7 of them (87.5%) were in SG. Estrogen receptor (ER) was measured in 43/47 (91.5%) BC specimens. ER was detected in 19/30 (63.0%) patients in SG and 3/13 (23.1%) in DG (p=0.01). Five-yr mortality was 10/21 (47.6%) in ER- and 3/22 (13.6%) in ER+ patients (p=0.008). Absence of ER expression [odds ratio (OR) 6.54; p=0.006] and absence of TAb (OR 9.37; p=0.03) were related to a higher mortality rate. TAb were detected in 8/21 (38.1%) ER- and in 7/22 (31.8%) ER+ patients; no relation was found between ER expression and TAb positivity (p=ns). Patients with ER+ and TAb+ have a better prognosis and the absence of a significant relationship between these two parameters suggests an independent prognostic role in high malignancy degree BC women.