Primary cervical screening with high risk human papillomavirus testing: observational study
BMJ 2019; 364 doi: https://doi.org/10.1136/bmj.l240 (Published 06 February 2019) Cite this as: BMJ 2019;364:l240All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Primary HPV testing is replacing cytology in England because both CIN3 and cancer rates are so much lower in HPV negative women than in women with normal cytology, and this implies that the screening interval can be longer. The results of the English cervical screening pilot published by Rebolj et al(1) confirm this and also illustrate the importance of differences in triage as well as in the primary test. These data cannot address the two outstanding questions: how long the recall interval should be for HPV negative women, and how HPV positive women should be managed. In the absence of direct evidence these must be addressed by modelling, and further data from this very large study will be crucial.
A unique aspect of this very large study is the large number of cancers, and the times when they were diagnosed should be examined. The authors concluded that HPV testing increased detection of both CIN3 and cancer, but the data suggest that cytology led to delayed rather than missed diagnosis. The protocols are not compared over the 4 years since baseline, but the sum of the CIN3+ rates for women aged 24-49 at entry and women aged 24-46 when they were routinely retested 3 years later suggests little increase in overall detection (1.8% for HPV and 1.7% for LBC). This supports the conclusion from the ARTISTIC trial, that HPV testing in women with normal cytology led to earlier diagnosis mainly because cytology was repeated one or two years later instead of at the next routine test 3 years later(2). The main clinical benefit is therefore earlier diagnosis of cancer. Detecting CIN3 a year or two earlier will prevent a very small risk of stage 1A cancer but detecting cancer a year or two earlier may well save a life. A Kaplan-Meier comparison of cumulative cancer diagnoses over 4 years would be the most useful analysis. Unfortunately this would be incomplete due to the authors’ inability to link these women to their national screening and cancer registration records. This was presumably due to unnecessary “ethical” restrictions, as linkage by NHS number is technically trivial.
A recent study in the US reported that a higher proportion of negative cytology results were upgraded if the positive HPV result was known by the reviewing cytopathologist(3). This bias can be seen in table 1, which shows that the immediate referral rate is lower in the primary cytology arm (3.8%) than in the primary HPV testing arm (4.2%). For women who test positive for both HPV and cytology it should not matter which test was done first, and the small number of HPV negative women with high-grade cytology were referred only in the primary cytology arm. The higher referral rate following primary HPV testing is presumably due to cytopathologists looking harder for abnormalities when they know someone is HPV positive, which must improve the sensitivity of HPV testing with cytology triage.
CIN2+ PPVs are reported under several scenarios including at baseline in HPV positive women with LSIL cytology (17% for primary cytology, 19% for primary HPV), after 12 months with persistent HPV 16/18 but negative cytology (13%) and after 24 months with persistent infection with other HR HPV types (10%, or 6% with negative cytology). It would be interesting to see PPVs for CIN3 and cancer in these groups to assess whether immediate referral to colposcopy is warranted in women with LSIL cytology, as this constitutes a major clinical burden. In a previous report on this pilot study(4), only 2/58 cervical cancers in the primary cytology arm presented with borderline or mild cytology. We look forward to further data from this study which will continue to provide valuable evidence to further improve the national screening programme.
Clare Gilham and Julian Peto
London School of Hygiene & Tropical Medicine, UK
1. Rebolj, M., et al., Primary cervical screening with high risk human papillomavirus testing: observational study. BMJ, 2019. 364: p. l240.
2. Kitchener, H.C., et al., HPV testing in combination with liquid-based cytology in primary cervical screening (ARTISTIC): a randomised controlled trial. Lancet Oncol, 2009. 10(7): p. 672-82.
3. Doxtader, E.E., et al., Knowledge of the HPV status biases cytotechnologists' interpretation of Pap tests originally diagnosed as negative for intraepithelial lesion or malignancy. Cancer Cytopathol, 2017. 125(1): p. 60-69.
4. Moss, S. and A. Gibney. HPV Primary Screening Pilots: Evaluation report to the National Screening Committee. 2015 [cited; Available from: https://legacyscreening.phe.org.uk/policydb_download.php?doc=560.
Competing interests: No competing interests
Discovery science and the impact agenda: lessons from HPV testing
Primary HPV testing will be rolled out in the cervical screening programme across the whole of England this year and the recent study by Rebolj et al confirms the benefits of this approach in increasing the sensitivity of detecting cervical intraepithelial neoplasia 3 and cervical cancer (1). This test for high risk HPV types will replace cytology as the first test in cervical screening.
The HPV tests approved for this HPV triage are all based on DNA amplification, an approach that we first demonstrated and published in the BMJ in 1989 (2). Looking back, we had no idea that this method would be of routine clinical value, we were just trying to use the then recently reported polymerase chain reaction (PCR) to develop more sensitive assays to study the natural history of HPV infection in a young population. We realised, however, that the routine testing of thousands of samples required a rigorous and quality assured laboratory environment and was therefore not appropriate in our experimental setting (3). Our solution was to work with a clinical virology laboratory in a hospital to validate and apply the test for our own epidemiological study (4). While other research laboratories continued to develop PCR testing for HPV in cervical cytology in epidemiological studies,(5) it became obvious to the commercial sector that this was a very lucrative opportunity. The rest is history but there are some salutary lessons here for funding agencies and governments determined to identify and measure the impact of basic research.
Firstly, like many colleagues in the biomedical field, although our study was the first to adapt and apply a new technology, we are rarely referenced for doing so. This is understandable as, unlike the various biotechnology companies, we didn’t have the expertise or funding to develop and commercialise the test. But is does mean that the impact of our original research goes unrecognised.
Secondly, the current obsession with research that has an obvious and immediate impact is clearly not applicable in this case where it took over 20 years of additional work and investment, mainly by the commercial sector, to hone the technology and develop its use for routine clinical testing (6). I’m only thankful that in the halcyon days of the mid-1980s the Cancer Research Campaign, as was, was prepared to fund our prospective collection of samples and to provide us with the freedom to experiment with and adapt novel technologies. It is sad that the liberty to explore new avenues and to take such risks is no longer available to young, creative scientists.
References
1. Rebolj M, Rimmer J, Denton K et al. Primary cervical screening with high risk human papillomavirus testing: observational study. Br Med J. 2019: 364:1240.
2. Young LS, Bevan IS, Johnson MA et al. The polymerase chain reaction: A new epidemiological tool for the investigation of cervical HPV infection. Br Med J. 1989; 298: 14-18.
3. Young LS, Tierney RJ, Ellis JRM et al. PCR for the detection of human papillomavirus infection: A mixed blessing. Ann Med. 1992; 24: 215 219.
4. Woodman CBJ, Collins S, Winter H et al. Natural history of cervical human papillomavirus infection in young women: A longitudinal cohort study. Lancet. 2001; 357: 1831-1836.
5. Coutlee F, Mayrand M-H, Provencher D et al. The future of HPV testing in clinical laboratories and applied virology research. Clin Diagn Virol. 1997; 8:123-141.
6. Cox JT. History of the use of HPV testing in cervical screening and in the management of abnormal cervical screening results. J Clin Virol. 2009; S1:S3-S12.
Competing interests: No competing interests