An article published Online First by The Lancet reveals that vaginal self-sampling for human papillomavirus (HPV) at home could unveil many cases of cervical cancer or precancerous lesions that could otherwise be left undiagnosed in developing countries. Self-testing could also prove to be beneficial in resource-deprived parts of high-income countries.

For detecting cervical intraepithelial neoplasia (CIN), the precursor to cervical cancer, grade 2 or worse, vaginal HPV tests are comparable or slightly more sensitive than cytology, also known as smear tests in clinic-based settings. Their effectiveness in home-settings is so far unknown.

Professor Attila T Lorincz at the Centre for Cancer Prevention, Wolfson Institute of Preventive Medicine, Barts and The London School of Medicine, Queen Mary University of London, UK, and Professor Eduardo Lazcano-Ponce, Centre of Investigation in Population Health, National Institute of Public Health, Morelos, Mexico, and his team across Mexico decided to examine the relative sensitivity and positive predictive value for HPV screening of vaginal samples self-collected at home compared with clinic-based cervical cytology.

For their randomized controlled trial the researchers recruited 25,610 Mexican women aged 25-65 years of low socioeconomic status from 540 medically underserved predominantly rural communities in Morelos, Guerrero and the state of Mexico. Eight community nurses who did not know their patients previously were assigned to make home visits by using a daily allocation list that contained the participants’ names and addresses.

Those who showed positive results in either test were referred to colposcopy, a visual inspection of the cervix using a magnifying instrument, which may include taking a biopsy sample. Researchers determined the primary endpoint as CIN 2 or worse detected by colposcopy.

Of the 12,330 women who were randomly allocated to HPV screening, 9,202 women adhered to the protocol, whilst from 12,731 women allocated to receive cervical cytology, 11,054 adhered to protocol.

9.8% of women had HPV, with 0.38% of women having an abnormal cytology rate. The researchers found that HPV tests revealed 117 women with CIN 2 or worse per 10,000 compared with 34 women with CIN 2 or worse per 10 000 identified by cytology, demonstrating a 3·4 times greater relative sensitivity of HPV compared with cytology. In addition they also found that HPV tests detected 4·2 times more invasive cancers (30 per 10,000) compared with cytology (7 per 10 000), however, the positive predictive value of HPV testing for CIN 2 or worse was 12·2%. This means that only 12% of home based HPV test cases referred for colposcopy actually proved positive for CIN 2 or worse by histology compared with 90% for cytology.

The researchers comment:

“The HPV test had a relatively low but acceptable positive predictive value, and our findings indicate the need for additional study of potentially suitable methods for triage. Vaginal self-sampling for HPV testing reduces the need for cytology clinics and permits an increase in screening coverage, especially in marginalized areas, in conjunction with more sensitive detection of cervical cancers and precursor lesions.

The challenge facing such testing as a strategy for cervical cancer prevention in low-income countries is identification of the most effective triage for HPV DNA positive women because the test substantially increases the number of colposcopy referrals, associated costs, and risks of over-treatment. With our results we aim to inform policy formulation and program implementation in deprived rural areas of Mexico, but these findings could also be relevant elsewhere, including in developed countries with low participation rates and localized resource-poor settings.”

Dr Nubia Muñoz at the Colombian Cancer Institute in Bogota, Colombia, and Dr Rolando Herrero at the Prevention and Implementation Group, International Agency for Research on Cancer in Lyon, France, say in a linked comment that despite the fact that these findings largely support those of earlier studies, it is the first community-based randomized trial that evaluates the use of HPV testing of home-based self-collected vaginal specimen.

They also comment that HPV-positive women need to be triaged, meaning they need to be sorted into groups based on their need by using either visual inspection with acetic acid (VIA), high-quality cytology, repeat HPV testing, or other biomarkers. Each method has its advantages and limitations. They refer to a US study, which demonstrated that primary HPV screening of home-based self-collected vaginal specimens with subsequent clinic-based cytological triage was accurate and cost effective.

They summarize by saying that:

“Mexico is, to our knowledge, the first country in the world to establish primary testing with HPV and subsequent cytological triage as the national policy, having already established a large network of high technology laboratories and done several million HPV tests…The experience in Mexico shows what can be achieved when scientific judgment guides public health policy.”

Written by: Petra Rattue