The Gynaecological Cancer Research Centre (GCRC) at University...

The Gynaecological Cancer Research Centre (GCRC) at University College London co-ordinates GCaPPS, UKCTOCS, UK FOCSS and UKOPS.

We have received a number of enquiries about the OvPlex? test following articles which appeared in the Daily Mail on 5 November 2008 and 25 August 2009. The OvPlex? test combines CA125 and four other chemicals in the blood produced by ovarian cancer cells (biomarkers) in order to detect ovarian cancer. These articles implied that the OvPlex? test is superior to the CA125 test in detecting ovarian cancer and that the CA125 test is not good at detecting ovarian cancer in its earlier stages. In addition, they implied that OvPlex? could be used as a test to screen for ovarian cancer.

The following points may be helpful to those taking part in GCaPPS, UKCTOCS, UK FOCSS and UKOPS:

1. According to information the GCRC has received from the company marketing OvPlex?, the test is designed for use in women who have symptoms rather than women without symptoms. This means that OvPlex? is not currently considered suitable for use as a screening test.
2.

In order to establish whether a test is good at diagnosing cancer in women with symptoms, it is first necessary to see if the test can tell the difference between women without cancer (controls) and women with cancer (cases).
1. OvPlex? has been tested using blood samples from 97 controls and 82 cases (women whose ovarian cancers were detected as a result of them being investigated for symptoms which made them go to their doctor).
2. The ability of the test to correctly identify women with ovarian cancer was then checked in a further 115 controls and 68 cases (women with ovarian cancer and symptoms, of whom 39 had early stage ovarian cancer).
3. Because there is not yet enough evidence to confirm or refute any potential advantage OvPlex? may have over CA125 in helping to diagnose ovarian cancer in women with symptoms, the company which markets OvPlex? is in the process of checking the accuracy of the test in a much larger study involving over 1,000 women worldwide.
3. OvPlex? has not been tested in samples taken months or years before women had a diagnosis of ovarian cancer. There is therefore no evidence at present that the test can detect ovarian cancer before it causes symptoms.
4.

According to the first publication on the performance of OvPlex?, the test was marginally better than CA125 alone at distinguishing cases (women with cancer who have symptoms) from controls (women without cancer or symptoms), see (a) and (b) below. It is important to note that the ideal control group would have been women with symptoms and benign pelvic tumours, as this is the group of women in whom ruling out ovarian cancer is extremely important. Because this control group was not used in the only OvPlex? study reported so far, we do not currently know if the advantage of the OvPlex? test over the CA125 test in diagnosing ovarian cancer would have been seen in a group of women with symptoms and tumours.
1. The ability of the OvPlex? test to pick up cancer (sensitivity) was 94.1% compared with 92.6% for the CA125 test.
2. The accuracy of the OvPlex? test in classifying healthy women as having a normal result (specificity) was 91.3% compared with 89.6% for the CA125 test.
5. The screening strategy in UKCTOCS and UK FOCSS uses CA125 in a different way to one-off tests such as OvPlex?, which are used to diagnose ovarian cancer in women with symptoms. In both trials, a complex computer program looks for subtle changes in a woman?s level of CA125 over time. When tested in post-menopausal women, this Risk of Ovarian Cancer Algorithm (ROCA) is more sensitive and specific than the CA125 test used alone.
6. Our team has a major programme of research to develop new biomarkers and tests, which may improve on or add to the CA 125 test?s use as a potential screening tool. This includes evaluating and reporting on potential diagnostic tests like OvPlex?. There is a constant stream of new biomarkers which seem to have potential, but despite major efforts over the last 25 years the CA 125 test remains the best and most thoroughly assessed screening tool to date. We hope that this will change in the near future, with new technologies and sophisticated laboratory approaches. In the longer term we envisage a test panel which includes CA 125 and 5 or 6 other biomarkers with proven ability to detect ovarian cancer in women before symptoms develop.