His test picks up traces of the cancer in four out of five women who've yet to develop symptoms - twice as many as previous screening methods.

Ian Jacobs started this work when, as a cancer surgeon in the UK, he saw many women with advanced ovarian cancer die. He's been determined to find a way to detect early traces of the disease ever since.

The test he has helped design picks up changes in the concentration of a protein in the blood, which is found in almost all women diagnosed with this type of cancer. An algorithm then determines a women's risk of developing the disease, the fourth most common cause of cancer deaths in women.

The algorithm can pick up 86 percent of women who have ovarian cancer before they have symptoms.

"The sensitivity is very, very high - much higher than people thought would be possible," said Professor Jacobs, who last year left the UK to lead one of Australia's largest universities, the University of NSW.

But before the test adopted as a population screening program, the team will have to wait until later this year to find out whether the test can pick up signs of ovarian cancer early enough to save women's lives.

"That's the primary point of the study, but we won't know that until June," said Professor Jacobs.

"I think it's unlikely that there will be no benefit," he said. "I'll be really upset and disappointed if there isn't, but it's hard to predict just how big a benefit there is."

The results of a randomised control trial of more than 202,000 postmenopausal women in the UK that compared the new statistical screening program with other screening methods are published in this week's Journal of Clinical Oncology.

A proportion of women were randomly assigned to be screened using the new algorithm, while two other similar-sized groups were screened using a vaginal ultrasound or not screened at all.

"It's one of the largest randomised trials ever performed in the history of medicine," Professor Jacobs said.

While scientists first discovered the protein CA125 could be used as an ovarian cancer biomarker in the 1980s, clinicians would typically classify a women as abnormal if they had high level of the protein. But some women with high levels don't have the cancer, while others with levels below the threshold ended up having the cancer.

The new screening method detected twice as many women with early signs of ovarian cancer than the fixed cut-off.

"What's normal for one woman may not be so for another. It is the change in levels of this protein that's important," Professor Jacobs said.

Each year, 100,000 women around the world die from ovarian cancer. One problem diagnosing the disease is that most women experience non-specific symptoms, such as bloating and bowel disturbances, which often means the cancer isn't diagnosed until an advanced stage when there's only a 30 per cent to 40 per cent chance a patient will survive five years after treatment. Women diagnosed early have a 90 percent chance of surviving past five years.

If the June results show the test does save women's lives, Professor Jacobs said it would likely be adopted as an annual screening program for postmenopausal women, when the risk of developing ovarian cancer increases.

"The great thing about CA125 is it's a very simple blood test. It's cheap and easy to do in the lab," Professor Jacobs said.

Dr Katherine Nielsen from Ovarian Cancer Australia said the well-designed study was important because it had identified women at risk of the disease before they had developed it.

But she said the real question to be answered was whether this approach would demonstrate a survival benefit in these women, the results of which will be released later this year.