Experts have devised a test to show which breast cancer patients will benefit from hormonal treatments.
Some breast cancers are fuelled by oestrogen
Certain types of breast cancer depend on oestrogen to fuel their growth.
But doctors currently can only get a rough indication of who would benefit from anti-hormonal treatments such as tamoxifen or anastrozole.
Now Cancer Research UK scientists have devised a more accurate test which will indicate whether women should have hormonal or other treatments.
Both anastrozole and tamoxifen work by interfering with the female hormone oestrogen.
Tamoxifen prevents its growth-promoting action of oestrogen on breast cells by blocking a molecule called the oestrogen receptor (ER).
It is estimated that 30% of women who receive tamoxifen do not benefit from it and would fare better if they were given other treatments.
Anastrozole, which is effective for post-menopausal women, shuts down the production of oestrogen.
Doctors currently measure how much ER women have in their tumours.
But even if women do have high levels of the molevule, many do not respond to the treatment because the ER is not properly active and not important for the growth of their cancers.
ER affects cell growth by switching on several genes.
Researchers tested ER activity by using a special test gene, which produces a bright yellow colour when it is turned on.
By measuring the intensity of the yellow colour, they were able to measure the switching-on activity of ER.
Most women with high levels of ER also had high ER activity, but in about one in five cases ER activity was low, even though levels of the molecule were high.
Researchers based at Imperial College London and the Kennedy Institute of Rheumatology found that assessing the activity of ER, rather than just measuring how much is in a woman's tumour, is a much better way of predicting whether treatment would work.
In addition to being used to better predict who is likely to respond well to particular anti-hormonal drugs, the test could also be used to monitor patients over the course of hormone treatment to tell doctors if a tumour becomes resistant to drugs.
Women usually take tamoxifen for around five years.
The researchers found ER activity dropped in women whose cancers developed resistance to tamoxifen during their treatment.
They say the test could show when it was time to switch to other drugs.
Professor Charles Coombes, who led the research, said: "Tamoxifen and anastrozole effectively knock out the influence of oestrogen on breast cancer growth.
"There's no point in doing that in patients where oestrogen receptor doesn't work properly, since in these cases the tumour will have learned to live without the hormone.
"Our new test will, for the first time, allow doctors to assess directly the importance of the oestrogen receptor molecule for the growth of a particular tumour, which means they'll know much more accurately whether or not tamoxifen or other anti-hormonal drugs are likely to work."
Professor Robert Souhami, director of clinical research at Cancer Research UK, said: "Tamoxifen has proved an incredibly useful drug, but it doesn't benefit all women.
"This test could give us much more reliable information about whether or not tamoxifen or other anti-hormonal drugs should be used, or whether chemotherapy should be used instead."
The research is published in the International Journal of Cancer.