Doctors have long argued over whether a screening programme - such as those for cervical and breast cancers - is feasible or justifiable.

But research published in The Lancet medical journal suggests there may be clear benefits in such a scheme.

Its authors are calling for a larger scale trial to establish exactly how great the benefits would be.

Simple tests

Dr John Toy is director of clinical programmes at the Cancer Research Campaign.

The earlier cancer is detected, the sooner it can be removed

He said: "Ovarian cancer is an awful disease because it presents late, so a lot of women die because the cancer is well advanced.

"If you can detect cancers earlier, it may present a better outcome for women."

Earlier studies have suggested that early stage ovarian cancer can be detected before symptoms arrive.

However, not enough research has been done to establish if the tests currently in use can provide an early enough diagnosis.

Cancer tests

They suggest cancer can be exposed by a blood test for the antibody CA 125 or by ultrasound scans.

These are the tests Dr Ian Jacobs, a consultant gynaecological oncologist at St Bartholomew's Hospital in London, used in his research.

It looked at post-menopausal women aged 45 or over.

They were split into two groups - 10,997 were offered screening, 10,958 were not.

Those in the screening group were given annual screens for three years - 29 possible cases of cancer were identified, six of which were confirmed.

Follow-up period

After the screening period, Dr Jacobs and colleagues followed the women's progress for another seven years.

In that time, 10 more cases of ovarian cancer were found in the screening group, and 20 in the control group.

Overall, nine women in the screening group died of ovarian cancer, compared with 18 in the other group.

Survival times of women who developed the cancer were longer in the screening group.

Women in the screening group who developed cancer lived for an average of 72.9 months after diagnosis.

Those in the other group survived an average of 41.8 months.

The study's authors said: "Screening for ovarian cancer would be a major undertaking with disadvantages and limitations as well as benefits."

They called for further research to assess whether screening could cut the death rate among the general population.

Dr Toy agreed that this study made a future large scale trial feasible.