Skip to main content

New smear test improves cervical cancer screening

Thursday 15 December 2011

Cervical cancer smear tests “could be improved by adding a further test looking for signs of a virus which causes it”, the BBC has reported.

This story is based on a large Dutch trial that looked at whether existing smear tests could be improved if genetic (DNA) tests were also performed to look for certain types of human papilloma virus (HPV). HPV is known to increase the risk of cervical cancer.

The research compared one group of women given a standard smear test with a group given both the smear test and the HPV DNA test. The researchers assessed how many women in each group were detected as having pre-cancerous cells and how advanced they were. They found that the DNA plus smear test identified more cases of lower grade abnormalities. These are less likely to develop into more serious abnormalities and into cervical cancer if treated. Additionally, when all the women were screened five years later, the group that received HPV DNA testing had fewer high grade abnormalities.

The results indicate that adding DNA testing for HPV could improve the detection of cervical cell abnormalities at an earlier grade when they are more easily treated. Further research will be needed, however, to determine if the test is appropriate or necessary for all women participating in a screening programme, or if specific subgroups would benefit more.

Where did the story come from?

The study was carried out by researchers from Cancer Research UK and from VU University and the Centre for Gynaecological Oncology in the Netherlands.

The study was published in the peer-reviewed medical journal Lancet Oncology.

The BBC reported on this study appropriately and thoroughly.

What kind of research was this?

This was a randomised trial that compared two screening methods for detecting potential cervical cancer. Cervical cancer screening programmes are designed to detect cell abnormalities before they develop into cancerous cells. These abnormalities, or ‘pre-cancerous cells’, are referred to medically as cervical intraepithelial neoplasia (CIN) or carcinoma in situ. This means that there are abnormalities in some of the cells of the cervix, but these abnormal cells have not spread beyond the surface layer of the cervix. This is not cancer, but may eventually progress to cancer if untreated. CIN is classified as grade 1 (mild changes) to 3 (severe changes), with lower grades carrying lower risk.

The study compared the standard examination of cells after a smear test (cytology) to using a DNA-based test for HPV alongside the standard cytology tests. If the addition of DNA testing is proven to lead to the detection of more low-grade abnormalities than the smear test alone, it could be a useful addition to the screening programme and have a meaningful impact on the number of cervical cancer cases prevented across the country.

What did the research involve?

As part of the Netherlands cervical cancer screening programme the researchers had access to 44,938 women aged 29 to 56 years, who they randomly divided into two groups: 22,420 women were assigned to the control group (cytology testing) and 22,518 to the intervention group (cytology plus HPV DNA testing). The women were given screening again five years later, with all participants receiving the HPV DNA test alongside their cytology.

The test results were classified as normal; grade 1, grade 2 or grade 3 CIN; or invasive cancer. The researchers collected data on the total number of cell abnormalities detected, as well as the grade of the abnormality, and compared these two outcomes across the groups, at both the first and second screenings.

What were the basic results?

The researchers initially compared the results of the first round of screening, in which the control group received just cytology and the intervention group received both cytology and a DNA test for HPV. The researchers found that:

  • The number of tests showing normal results was similar between the two groups.
  • The number of grade 1 abnormalities detected was similar between the two groups.
  • An additional 0.16% had grade 2 abnormalities detected than in the control group (96 vs. 65, risk difference 0.16%, p=0.014).
  • The number of grade 3 abnormalities detected was similar between the two groups.
  • The number of cancers detected was similar between the two groups.
  • In the intervention group 27% more grade 2 or worse abnormalities were detected compared with the control group (267 vs. 215, risk difference 0.27%, p=0.015).

The researchers then looked at what happened during the second round of screening, when both the control and intervention groups received cytology and the DNA test for HPV:

  • The number of normal tests was similar between the two groups.
  • The number of grade 1 abnormalities detected was similar between the two groups.
  • The number of grade 2 abnormalities detected was similar between the two groups.
  • The number of grade 3 abnormalities detected was similar between the two groups.
  • There were fewer cancers detected in the intervention group (4 vs. 14, risk difference 0.29%, p=0.031).
  • In the intervention group 0.17% fewer grade 3 or worse abnormalities were detected compared with the control group (88 vs. 122, risk difference -0.17%, p=0.023).

The researchers found that within the intervention group there was an association between detection of a strain of HPV called HPV16 during the first screen and the chances of detecting an abnormality of grade 3 or worse during the second screen. Previous research has shown that HVP16 is the strain of HPV that most often causes cervical cancer.

How did the researchers interpret the results?

The researchers say that their results indicate that adding HPV DNA testing to a cervical cancer screening programme can improve the detection of lower grade cell abnormalities at an initial screening. Detecting such abnormalities at grade 2 can lead to effective treatment and a reduced risk of them developing into grade 3 or higher abnormalities. They also say that detection of HPV16-related grade 3 abnormalities is expected to have an effect on long-term cervical cancer mortality.


This was a large study that compared two methods of screening for cervical cancer. It compared the standard method of examining cells after smear tests to a programme combining smear tests and a DNA test to detect HPV. The screening techniques used were similar to those of the NHS cervical screening programme and the participants comparable with those who would normally undergo NHS cervical screening. As such, it is likely that these results can be generalised to a UK population.

The current cervical smear screening programme is able to detect these early pre-cancerous changes (with abnormal screening results later confirmed by biopsy), but this research analysis indicates that adding HPV DNA testing to the current screening programme could be effective in increasing the number of these early abnormalities that are detected. Being able to detect more of these abnormalities would be important in the prevention of cervical cancer, as evidence shows that pre-cancerous abnormalities are treatable and further development into more risky abnormalities or cancers can be reduced.

Although the technique has shown that it may increase the rate of abnormalities detected, further follow-up will be needed to determine if the addition of HPV DNA testing would actually have an effect on the number of cervical cancer diagnoses and deaths in the long-term. Additionally, the use of such a test may be more appropriate for certain subgroups than others. In particular, a woman’s age is likely to have an effect on the risks and benefits associated with adding an HPV DNA test to the current method. The researchers suggest that, for certain age groups, additional screening may also lead to the problem of over-diagnosis. This is where tests detect abnormalities that would otherwise have regressed and progressed no further.

The balance of risks and benefits is a key factor that needs to be taken into account when considering any screening test. Additional research may now reveal which populations would benefit the most from such testing, and whether the information provided by testing can be used to improve cancer diagnoses and survival rates.

Analysis by Bazian
Edited by NHS Website

Links to the headlines

Test 'improves cancer screening'

BBC News, 15 December 2011

Links to the science

Rijkaart D, Berkhof J, Rozendaal L, et al.

Human papillomavirus testing for the detection of high grade cervical intraepithelial neoplasia and cancer: final results of the POBASCAM randomised controlled trial

Lancet Oncology. December 15 2011. Early online publication.