A new method of detecting those at risk of cervical cancer at an earlier stage could benefit thousands of women said The Daily Telegraph. The newspaper reports that if it is “successfully developed, the new DNA-based test could save more lives”.
The reports are based on a study which highlights the importance of testing for the particular strains of human papilloma virus that are linked to cervical cancer. The results appear to show that there are benefits for screening when the DNA test is used in combination with the normal cervical smear test.
However, other recent studies suggest that the prevalence of this virus in the community can also be reduced by vaccination, before it gets a chance to produce cancerous changes. The authors advise waiting for further studies and a cost-effectiveness analysis before deciding how relevant this test will be in their country.
Where did the story come from?
Dr Bulkmans from Department of Pathology at the VU University Medical Centre in Amsterdam, and other experts in epidemiology and gynaecology from elsewhere in Holland, carried out this research. The study was funded by a Dutch organisation for Health Research and Development. It was published in the peer-reviewed medical journal: The Lancet. Some authors had lectured or participated in pharmaceutical company research.
What kind of scientific study was this?
This was a randomised controlled trial of two screening strategies to investigate whether the combination of a genetic test for the human papilloma virus (HPV DNA test) and a usual smear test would be better at identifying those women at high risk for cervical cancer – usually indicated by an abnormal cervical smear – than a smear test alone.
Over 18,000 women in Holland, aged between 29 and 56 years, who were already participating in a regular cervical screening programme, were randomly allocated to two groups. One group was offered a conventional smear test and the other was offered a smear test combined with an HPV DNA test; certain strains of HPV are known to cause cancer of the cervix and these can be identified by their DNA. After 5 years, a second round of regular screening took place, and both groups underwent the combined smear and HPV DNA testing.
The researchers looked at the results of both the smear tests and the combined HPV DNA and smear tests, and referred any women who were at high risk of cervical cancer (following normal screening procedures for the smear test and according to defined criteria for the combined test) for further investigation. Abnormal cells in the cervix were then identified and a biopsy was taken to see if they were potentially pre-cancerous lesions (called cervical intraepithelial neoplasia grade 3 – CIN3+).
What were the results of the study?
Over 8,500 women were allocated to each group and were followed for more than six and a half years. At the first round of testing, after further investigation in any high-risk women, more lesions were found in the group that had the normal smear test combined with HPV DNA testing. This confirms the results from previous studies that show that DNA testing is more sensitive than normal smear test at detecting the CIN3+ lesions.
The number of lesions found in the HPV DNA group was 70% higher. Sixty-eight women were found to have lesions compared with 40 women in the smear test-only group.
In the second round of testing, where all the women had HPV DNA tests, fewer lesions were detected in the HPV DNA testing group, since the previous abnormalities had been picked up early, a decrease of 55%.
Reassuringly, about the same number of CIN3+ lesions were detected over both screening rounds in the two groups (94 and 92) suggesting that the advantage gained by this new test is earlier detection and this, by inference, could possibly extend the time interval between cervical screening tests.
What interpretations did the researchers draw from these results?
The time between regular cervical screening in Holland is currently five years and the researchers suggested that this could be extended by one year.
What does the NHS Knowledge Service make of this study?
There are several positive features to this well conducted and reported trial, and the authors discuss features that suggest the results can be applied to a Dutch population:
- The rate of HPV prevalence in the community (4.5%) for this study, was low compared with other community studies.
- The detection of 4.7 lesions for every 1,000 women screened in the control group is similar to the detection rate claimed by the Dutch national screening programme. The authors claim that the results are therefore representative of the results obtained by the Dutch national screening programme, operating outside of trial conditions.
- Previous studies quoted by the authors have shown an increased sensitivity of 23-43% for HPV DNA testing above normal smear tests, but at the expense of a 5-8% loss in specificity. This means that although it is a good test for picking up people who have HPV, it is not as good at correctly excluding those people without the virus.
In a screening context, this small loss of specificity may not be important to the women tested, as only a few would receive unnecessary testing; however the cost of providing DNA testing will also need to be considered by those who fund the service. The delicate balance between benefits to the individual and the overall costs to the health service of these improved tests will need further evaluation. In addition, the application of this test to other countries, such as the UK, where the virus may be more or less common, will also need testing.
Sir Muir Gray adds...
Women do not like the cervical smear test, both the test itself and the anxiety caused by a false-positive test result. Anything that can be done to reduce the number of tests needed, while maintaining the preventive effect of regular testing, will be very welcome
This additional test might be able to reduce the burden of testing for individual women without reducing the benefits of the screening programme; it will be examined closely to assess its relevance to the UK.
Analysis by Bazian
Edited by NHS Website
Links to the headlines
The Daily Telegraph, 4 October 2007
Links to the science
Lancet 2007 Oct 4; [Epub ahead of print]
Cochrane Database Syst Rev 2002, Issue 3