Stewart J F, King R J, Winter P J, Tong D, Hayward J L, Rubens R D
Eur J Cancer Clin Oncol. 1982 Dec;18(12):1315-20. doi: 10.1016/0277-5379(82)90135-3.
One hundred and twenty-four patients with stage III breast cancer who had oestrogen receptor analyses (ER) performed on primary tumour were studied. For operable disease (T3a, N0.1, M0), patients with ER-negative (ER less than 5 fmol/mg cytosol protein) tumours had a shorter duration of symptoms, but no other differences were observed at presentation. In patients with inoperable (T3b,4, N0.1, M0 or any T, N2.3, M0) tumours there were no differences in the clinical characteristics that rendered the primary tumour inoperable between those with ER+ or ER- tumours. The median disease-free interval (DFI) in patients with operable tumours and uninvolved axillary nodes was significantly longer in those with ER+ tumours (P less than 0.05). Median survival was significantly longer in patients with ER+ tumours than in those with ER- tumours (50 vs 27.5 months, P less than 0.05), and when survival was analysed according to various methods of initial treatment, it was significantly prolonged in patients with operable (T3a, N0.1, M0) ER+ tumours compared to those with operable ER- tumours. In patients with inoperable tumours no effect of ER status on prognosis was demonstrated. These results suggest that oestrogen receptor content may be a prognostic variable in patients with operable stage III breast cancer, but further studies of patients with similar methods of initial treatment are needed to confirm this. Patients with operable stage III tumours should be considered separately from those with locally advanced inoperable tumours in view of the significant differences in their survival prospects.