Pre-operative oestradiol levels - relation to survival in breast cancer

Aims: There are clinical observations that operation during the luteal phas e of the menstrual cycle (with high oestradiol levels) may positively influ ence prognosis in breast cancer. However, few studies have information on p lasma levels of hormones pre-operatively. Methods: We studied 774 women treated for breast cancer where plasma levels of oestradiol had been measured 1-2 days pre-operatively. Date and cause o f death were ascertained from the files of the Swedish Cancer Register and 5434 person-years were observed, The endpoint was death with breast cancer as the underlying cause (n=41 and n=158 in the pre- and post-menopausal gro up, respectively). Results: In life-table analyses, only pre-menopausal patients with oestradi ol 500 pmol/l and above had a tendency (not statistically significant) for better survival. Multivariate Cox models with oestradiol modelled in contin uous form yielded relative hazards (RH) close to unity in all women and in strata according to menopausal status. Conclusions: When oestradiol was analysed in categorized form, only women w ith the highest levels had a tendency for improved prognosis (RH around 0.7 ; not statistically significant). Moreover, this pattern was not apparent f or pre-menopausal women. Our findings contradict the notion that the pre-op erative oestradiol level is independently associated with breast cancer pro gnosis. (C) 2001 Harcourt Publishers Ltd.