These long-term studies provide up to 30 years follow-up following HPV infection on the risks of cervical cancer and pre-cancer. HPV testing is now replacing cervical cytology in the national cervical cancer screening programme, and our primary aim is to provide recommendations on screening intervals and triage policy.
Over 15 academic publications in peer-reviewed journals have stemmed from our work on cervical screening.
Human papillomavirus (HPV) infection is known to cause cervical cancer, but it is a relatively common infection, especially in young women, which usually clears without any symptoms or long-lasting effects. There are different strains (known as genotypes) of HPV, and some are more likely to cause pre-cancer or cancer than others.
The latest scientific evidence (including results from the ARTISTIC trial) shows that screening for HPV infection as a first test (known as “primary HPV screening”) is better than screening for abnormal cytology with a “smear” test. This method is currently being tested in six pilot areas in England, and there are plans to roll it out nationally by December 2019. It’s not as simple as swapping one test for another, however, and there are still decisions to be made on how best to go about it.
The disadvantage of primary HPV testing is that more women, particularly young women, will test positive for HPV including many with normal cytology. It is not sensible to refer a large number of women for further tests when their infection is likely to clear on its own and so a second test has to be used to identify the women who are at highest risk (called a triage test). A balance must be achieved so that most of the women with abnormal cells are identified but unnecessary referral and anxiety for women are minimised. At the moment women who are positive for HPV will be referred only if they also have some abnormal cells (from a cytology test).
The NHS currently uses an HPV test which can separately identify HPV 16 and HPV 18. Women with HPV16 in particular, have a much higher risk of pre-cancer and cancer and so it makes sense to use this information in the screening programme. It is possible to analyse HPV samples to identify the actual types (genotypes) of HPV present. The 13 high-risk types can only be individually identified using specialist laboratory tests within a research study.
HPV infections that come and go are harmless but infections that persist put women at higher risk of developing disease in the future. Young women in particular may acquire an infection which clears and then acquire another infection at a later date. This means that they could test positive for HPV at two time points which might appear that they have a long-lasting infection when in fact they have two transient infections. This happens more often than you might think because about 30% of women aged 25-29 and 15% of women aged 30-39 are positive for HPV at any point in time. Genotyping can help establish whether the same HPV type is present at the two time points (persistent), or whether there are two separate infections (“type-swap”)
ARTISTIC: A Randomised Trial In Screening To Improve Cytology
Between July 2001 and October 2003, women aged 20-64 years attending for routine cervical screening (NHS programme) in Greater Manchester were invited to participate in the trial. In total, 24,510 participants had a cervical sample collected for liquid based cytology and HPV testing, and were randomly allocated in a ratio of 3:1 to have the HPV result revealed and acted upon, or concealed and further management based on cytology alone. Management of women with abnormal cytology was identical on both arms and in line with the national guidance for the English Cervical Screening Programme (NHS CSP 2004). Women in the revealed arm with negative cytology who tested HPV positive were invited for repeat HPV testing at 12 months and if still positive could choose between immediate colposcopy or a repeat HPV test at 24 months and colposcopy if still positive.
The cohort was followed beyond the original protocol for a third screening round. During this study extension, the ethics committee required that HPV results were linked to anonymised data, so women on both randomised arms were managed on the basis of cytology according to national guidelines. HPV results in round 3 were ignored until March 2008, when the Manchester Cytology Centre became one of six Sentinel Sites for HPV triage in England. The aim of the Sentinel Sites project was to evaluate the roll-out of HPV triage of low grade cytological abnormalities. Therefore, women with borderline and mild dyskaryosis cytology who tested HPV positive were referred to colposcopy, whilst those who were negative were returned to routine recall.
The Manchester Study
Between 1987 and 1993, in collaboration with over 100 general practitioners and screening clinics in the Greater Manchester area who used the Christie Hospital cytology laboratory (now the Manchester Cytology Centre sited at Manchester Royal Infirmary), 78,062 cervical cell samples were collected from 61,564 women attending for routine screening. There was no age restriction. Participating practices and clinics covered a wide area in and around the city of Manchester, and offered screening either in the context of well-woman clinics or in association with family planning services.
The study was approved by the local ethics committee. Verbal informed consent obtained when the smear was taken was deemed appropriate, as the clinical significance of HPV infection was not then known. HPV assays were performed after recruitment had ended, and no HPV results were reported either to the cytology laboratory or to the women.
At recruitment a cervical smear was obtained from each woman, usually using an Ayre spatula. A second scrape was then taken using the same spatula and the tip, together with adherent cervical cells, was broken off into a sterile container containing 10ml of storage buffer (0.9% PBS pH 7.2 + 0.1% SDS). Spatula samples were sent together with the corresponding smears and pathology request forms to the Christie Hospital Cytology Department. Smears that were accompanied by a spatula were flagged on receipt on the laboratory database. Spatulas were vortexed, and the cell suspensions transferred to freezer tubes. All samples are stored at -25°C.
Samples taken before Jan 1989 were centrifuged and only the pellet was stored. This procedure (in the pre-PCR era) entailed some loss of DNA and the possibility of contamination. The proposed study will therefore be restricted to the 49,549 women recruited 1989-93.
HPV testing was carried out between 1990 and 1996 on a random sample of 7,278 of the women in the cohort, and on selected cases diagnosed with CIN3. PCR analysis was by HPV L1 consensus PCR amplification using MY09/MY11 primers. A 286bp human b-globin fragment was amplified as an internal PCR control. HPV positive samples were dot blotted onto membranes and hybridised with a series of biotinylated type-specific HPV probes. The methods, and results on the relationship between HPV infection and development of high-grade cervical neoplasia over the following 5 years, are described in Peto et al (2004).
Long-term follow-up of ARTISTIC and Manchester Study cervical screening cohort
Women were followed through local cytology laboratories for cytology and histology to 2009 for the ARTISTIC study and to 1998 for the Manchester Study.
We are now undertaking complete cytological follow-up by linkage to NHS cervical screening call-recall records for both cohorts to include women who have moved from the Manchester area. The cohorts are also being followed-up for CIN3 and cancer incidence and mortality by linkage to the NHS Central Register.
Only the highest degree of pre-cancer, CIN3, is registered nationally (along with cancers) but women with CIN2 and CIN3 are usually treated the same. This means that it is useful to count both CIN2 and CIN3 when analysing the data since for practical purposes they will be treated the same by a doctor. We now propose to extend the ARTISTIC follow-up to 2021 by obtaining histology results from local laboratories from 2009 to 2021. These laboratories still process samples for around 90% of the women in the study. This means that we will get diagnoses of CIN2 for this period of time, as well as CIN3. There are roughly equal numbers of each, so we shall double the number of diagnoses which will make our calculations more precise.
In addition to collecting additional data on CIN2+ diagnoses from the ARTISTIC cohort, we are also proposing to collect further samples from these women and to test them again for HPV. About 10,000 cervical tests among ARTISTIC women will be done over the 5 year period between 2017 and 2021 in the Manchester Cytology Centre which covers the Greater Manchester area. These samples will be tested for HPV in the specialist Virology Laboratory, the positive samples will be genotyped, and the residual samples will be stored for future research on new tests.
We also plan to link the approximately 1700 women who participated in both the ARTISTIC and Manchester studies and to relate HPV results taken at these 2 time points around 10 years apart.
Objectives of long-term follow-up of both study cohorts
To obtain the data required to evaluate the long-term benefits of alternative screening strategies using primary HPV testing:
- The long-term protection of a negative HPV test and hence the safe screening interval at different ages
- Optimal ages at start and stopping screening
- The role of HPV typing and test sensitivity
- Triage of HPV positive women: what follow-up tests should be done in women who test positive for HPV?
Nested case-control study of incident cases of CIN3 and cervical cancer
This case-control study, nested within the Manchester Cohort, will compare HPV status (in 1989-1993) and cervical screening histories in women who subsequently developed CIN3 or cervical cancer against a set of 320 age-matched control women. The expected number of cervical cancers in the cohort is 130 based on English national rates, but the rate will be somewhat lower in this screened cohort. We estimate that there will be about 80 incident cervical cancers, including about 25 who were aged over 40 years at entry to the study.
We are following these cohorts passively, there is no contact with the women and therefore their screening and care is not affected by this study. All data files: original study data, cervical screening records, histology data, cancer and death registrations are linked using NHS numbers. Every woman has a unique study number which is used in our study database. Along with study number, the data stored are: date of birth, date and result of smear tests, HPV test results, cervical histology results (biopsy results), cancer registrations, death registrations with causes of death. No names or addresses are stored in this master data file. It is kept on secure password-protected computers at LSHTM and can only be accessed by the study statistician. NHS numbers, names and addresses were sent securely to NHS Digital for linkage to cancer registration and mortality data. Our processing activities have been approved by an ethics committee, we have Section 251 approval from the Confidentiality Advisory Group and further permission from ONS and NHS Digital to use cancer and death registration data. The statistical analyses of the data produces tabular or graphical results where no member of the studies could be identified.
If you participated in either the Manchester Study (1988-93) or ARTISTIC Trial (2001-04) and wish to find out more, or wish to opt out of your data being used for this purpose, please e-mail Clare Gilham.