Roland Hähnel1, Katrina Spilsbury. 1. Centre for Molecular Immunology and Instrumentation, University of Western Australia, Sir Charles Gairdner Hospital, Nedlands, Western Australia, Australia. hahnel@cyllene.uwa.edu.au
Abstract
BACKGROUND: The oestrogen receptor status of a breast tumour predicts the response to hormonal treatment and is an important prognostic marker; women with oestrogen receptor positive tumours having a better short-term survival outcome. METHODS: Kaplan-Meier estimates and Cox proportional hazard model were used to estimate the association between oestrogen receptor levels and long-term breast cancer-specific survival outcomes in 5735 women diagnosed with breast carcinoma from 1970 to 1997 in Western Australia. Further analysis was performed on a subset of women for whom biochemical and tumour characteristics were also available. RESULTS: Five-year breast cancer-specific survival estimates for women with oestrogen receptor positive tumours was 0.85 (95% CI 0.84-86) compared to 0.72 (95% CI 0.70-74) for women with oestrogen receptor negative tumours. The relative survival advantage of an oestrogen positive tumour over oestrogen negative tumours disappeared by the fourth year (0.8, 95% CI 0.6-1.0). Conditional upon surviving 5 years, long-term breast cancer-specific survival was better for women with oestrogen receptor and progesterone receptor negative tumours compared to other women (log rank test P-value <0.05). CONCLUSION: Despite an earlier survival advantage for women diagnosed with oestrogen receptor positive tumours, after 5 years of survival, women with oestrogen receptor negative and progesterone receptor negative tumours had better long-term survival outcomes from breast cancer compared to other women.
BACKGROUND: The oestrogen receptor status of a breast tumour predicts the response to hormonal treatment and is an important prognostic marker; women with oestrogen receptor positive tumours having a better short-term survival outcome. METHODS: Kaplan-Meier estimates and Cox proportional hazard model were used to estimate the association between oestrogen receptor levels and long-term breast cancer-specific survival outcomes in 5735 women diagnosed with breast carcinoma from 1970 to 1997 in Western Australia. Further analysis was performed on a subset of women for whom biochemical and tumour characteristics were also available. RESULTS: Five-year breast cancer-specific survival estimates for women with oestrogen receptor positive tumours was 0.85 (95% CI 0.84-86) compared to 0.72 (95% CI 0.70-74) for women with oestrogen receptor negative tumours. The relative survival advantage of an oestrogen positive tumour over oestrogen negative tumours disappeared by the fourth year (0.8, 95% CI 0.6-1.0). Conditional upon surviving 5 years, long-term breast cancer-specific survival was better for women with oestrogen receptor and progesterone receptor negative tumours compared to other women (log rank test P-value <0.05). CONCLUSION: Despite an earlier survival advantage for women diagnosed with oestrogen receptor positive tumours, after 5 years of survival, women with oestrogen receptor negative and progesterone receptor negative tumours had better long-term survival outcomes from breast cancer compared to other women.