ASCO 2011 – Jury still out on ovarian cancer screening

Screening must be accurate and reliable – we’re not there yet for ovarian cancer

Thanks to research and improved treatments, survival from ovarian cancer has almost doubled over the last 30 years. Yet more than 4,000 women still die from the disease each year in the UK.

Clearly, more research is needed to help reduce this toll.

One of the main issues is that ovarian cancer is often detected at a late stage, when treatment is less likely to be successful. This is because the symptoms are quite mild and easy to overlook, so treatment is often started late – something we’ve previously written about on the blog.

There have been a number of efforts across the globe to develop an effective screening strategy for ovarian cancer but, sadly, no trial has been able to demonstrate improvement in survival so far.

Screening for any type of cancer must be accurate and reliable – it needs to accurately detect the disease and it must not give too many so-called ‘false positive’ results in people who do not have cancer.

Researchers continue to look at different ways to screen for ovarian cancer, and there were a few sessions at ASCO this year talking about two tests in particular that are being investigated for their potential to detect ovarian cancer early – testing for a protein called CA-125, and transvaginal ultrasound. As the conference organisers tweeted:

http://twitter.com/#!/ASCO/statuses/77087481354391552

Screening tests for ovarian cancer

CA-125 is a protein that’s produced by some ovarian cancers and can be detected with a simple blood test.

Researchers have been looking into whether doctors can measure the levels of this protein and determine whether high levels could be a sign of ovarian cancer. There are complications around this method, as high levels of CA-125 don’t always mean that a woman has ovarian cancer and, equally, not all women with ovarian cancer have high CA-125 levels.

Transvaginal ultrasound is a test that scans the ovaries from inside the vagina, giving a clearer image than scanning the abdomen.

At ASCO this year, experts from around the world discussed two trials looking at the use of a combination of CA-125 and transvaginal ultrasound– PLCO and UKCTOCS.

New results published from US study

PLCO is an American trial that screened women over 10 years – looking for high levels of CA-125 for the six years alongside an annual ultrasound examination (for four years). Participants were screened at 10 centres across the United States between November 1993 and July 2001 and were followed for up to 13 years after the screening period was over.

New results from the PLCO trial were reported at ASCO, showing that this method of screening did not result in fewer deaths from the disease compared with usual care. Furthermore, false-positive results from the two screening methods often led to unnecessary surgeries and some women had serious complications, meaning that women who were screened were actually worse off.

Indeed, 15 per cent of women who received a false positive diagnosis and were then followed up were found to have had a major complication (such as infection or cardiovascular complications) in their subsequent unnecessary treatment.

These findings are important for informing future efforts to detect the disease early. Plus this trial does not rule out the future use of CA-125 and ultrasound – if used differently than in PLCO, they could still be effective.

UKCTOCS – the UK’s ovarian screening study

This is why results from the UK ovarian screening trial – UKCTOCS – are eagerly awaited. Funded by Cancer Research UK, in collaboration with the Medical Research Council and the National Institute for Health Research, UKCTOCS is also looking at the potential of CA-125 combined with ultrasound screening, but it is looking at these tests in a different way to the PLCO trial.

The UKCTOCS researchers are using a calculation called the ‘risk of ovarian cancer algorithm’. Rather than just looking at the results of a single CA-125 test, they look for changes in CA-125 over time to try to predict the woman’s risk of having the disease.

The results from the algorithm are used to determine what should happen next. Women at low risk return to the normal study cycle of annual testing; women at high risk have further investigations to find out whether they do have ovarian cancer.

There are other key differences in the trial design as well. In the PLCO trial, the women who were screened were offered a CA-125 test every year for six years, as well as an annual ultrasound exam every year for four years. Participants were followed up in 2010 to see whether they developed ovarian cancer or had died from the disease. This meant that women who were recruited at the beginning of the trial in 1993 had a longer follow-up period than those recruited later on.

In her discussion about PLCO at ASCO, Professor Usha Menon, trial coordinator and one of the principal investigators of UKCTOCS, said that it was unlikely that screening would have had an impact in the women who were left to be followed up for longer. This ‘dilutes’ the effect of screening across the trial, as the last time these women were screened could potentially have been several years ago and cancers may have developed in the intervening period.

Changing trial design for more accurate results

The UKCTOCS trial recognised this limitation part way through and changed its design so that all the women who were screened continued their screening right up until the end of the trial. This means that the women who are screened will have between 7 and 11 screens depending on when they were recruited. All women will be followed up 3 years later in the hope that the screening effect won’t be diluted.

Dr Menon said that the results of the PLCO trial and others demonstrate that we haven’t yet found a strategy for screening for ovarian cancer that will work. She also said that “we don’t know if UKCTOCS will define a strategy that works”, but whatever the results are, they will give us insights into this challenging problem.

Next, she raised the importance of having a comprehensive follow-up strategy to accompany screening programmes. This needs to be well-organised and work properly if screening is to make an impact on deaths from cancer.

Dr Menon also talked about some of the other issues around ovarian cancer screening. Most deaths come from Type II ovarian cancer – these are high-grade tumours and have almost always spread beyond the ovaries when the disease is spotted. Around eight out of ten of these cancers come back after treatment.

This poses the question as to whether treatment is really making an impact if recurrence is so high – will detecting ovarian cancer earlier really make a difference to survival if, ultimately, it will come back again?

Still seeking answers

At the moment, there is no solid evidence for ovarian cancer screening being effective in the general population, and more research is needed to find a screening strategy that will work. We’ll have to wait for the results from UKCTOCS for further insights into the potential use of CA-125 and ultrasound.

Dr James Brenton, an ovarian cancer doctor and scientist based at our Cambridge Research Institute, commented on the American PLCO study on behalf of Cancer Research UK:

“This important research suggests that having a yearly ultrasound test along with a blood test which provides a snapshot of the levels of a protein associated with ovarian cancer – the serum CA125 test – is not going to cure more women of ovarian cancer. Ovarian cancer is very hard to detect at an early stage before it has spread.

Ongoing research funded by Cancer Research UK is testing whether smaller rises in CA125 over time can be a better predictor of early ovarian cancer and is working with international groups to identify common genes that might slightly increase the risk of ovarian cancer.

Combining genetic tests with ultrasound and serum markers will provide a more accurate way of detecting those women who have a greater chance of developing ovarian cancer, and curing them.”

As the current national screening programmes for breast, bowel and cervical cancers highlight, screening (when implemented effectively) can have a tremendous impact on cancer.

Although the PLCO trial will not be remembered as a major success of this year’s ASCO conference, it is nevertheless invaluable research. And it will help to inform ongoing work to identify a successful method to screen for ovarian cancer.

Thanks to our supporters, we’re proud to be contributing to these continued efforts through the UKCTOCS trial. By continuing to explore exactly what kinds of tests work, and how best to use them, we will determinedly make our way towards an effective screening programme for ovarian cancer.

Nina Callard