Skip to main content
Premium Trial:

Request an Annual Quote

Dutch Study Highlights Accuracy of Self-Sampling for HPV Screening


NEW YORK (GenomeWeb) – Molecular testing for human papillomavirus using patient-collected cervico-vaginal samples has shown similar levels of accuracy to testing done on clinician-collected samples and could boost participation in overall HPV screening, according to a research team that just completed a large study in the Netherlands.

Published this month in Lancet Oncology, the study evaluated the accuracy of a GP5+/6+ PCR enzyme immunoassay from Labo Biomedical Products in Rijswijk, Netherlands, in identifying up to 14 high-risk HPV types in women undergoing HPV screening. 

It is among the first large, randomized trials conducted as part of an organized screening program that tested the accuracy of using a self-sampling device in combination with a clinically validated, PCR-based HPV assay, according to J. (Hans) Berkhof, a study author and professor at Amsterdam UMC.

The study was completed by the VU University Medical Centre in Amsterdam, the Erasmus University Medical Centre in Rotterdam, and the Radboud University Medical Centre in Nijmegen in collaboration with screening organizations Midden-West, Zuid-West, and Oost.

Detecting cancer lesions in women who periodically attend clinics can be more difficult than detecting them in underscreened women who have rarely or never been screened, or have refused to attend clinics following an invitation from a screening organization, Berkhof said.

Underscreened women tend to have more advanced lesions that are easier to detect, he said, and added that most previous studies have focused on the underscreened. Consequently, researchers and clinicians have found difficulty in determining how well self-sampling in combination with PCR testing would work in women who attend routine screening.

The Dutch study examined 7,643 women in a self-sampling group and 6,282 in a clinician-collected sampling group. The primary endpoints of the study were detection of cervical intraepithelial neoplasia of grade 2 and worse or grade 3 and worse (CIN2+ or CIN3+).

Of the self-collecting patients, 569, or 7.4 percent, tested positive for HPV, while in the physician-collected group, 451, or 7.2 percent tested positive. The CIN2+ sensitivity and specificity of HPV testing did not differ between the two groups, while CIN3+ relative sensitivity was 0.99 with essentially equivalent specificity.

The data support the idea that underscreened women and those who have been attending screening in clinics could instead collect their own samples at home as part of a primary screening program, Berkhof said.

The Netherlands has a national screening program backed by the government whereby eligible women are invited to attend clinics for free HPV screening every five years. In the study, a woman was classified as underscreened if she was 34 years or older; had a previous screening invitation; and it had been seven years or more since her previous cervical cytology exam.

In the Netherlands, about 7 percent of all women who are invited to do HPV screening currently opt for self-collection of samples, Berkhof said.

Many find HPV self-sampling to be more convenient, less embarrassing, less uncomfortable, and less painful than sampling done by a clinician, but they are concerned about test accuracy and that has limited the uptake of self-collection tests, Berkhof noted.

"We were initially concerned that HPV PCR testing in combination with self-collection could be inferior to HPV PCR testing in combination with clinician collection in a routine screening," Berkhof said.

The study findings could be particularly important where the prevalence of cervical cancer is highest in low- and medium-resource settings in both undeveloped and developed countries. 

Self-sampling permits a variety of advantages in extending and speeding screening programs and is especially useful in low-resource settings in which the number of clinicians trained to preform procedures using a speculum is limited, Mark Schiffman, a senior investigator at the National Cancer Institute, said in an interview. 

"The [Netherlands-based] study adds to the growing and compelling evidence that self-sampling following by a sensitive HPV test is equivalent to clinician sampling with HPV testing for cervical screening," said Schiffman, who last year testified about molecular HPV testing and self-collection of samples at a US Food and Drug Administration workshop.

Adoption of self-collected samples for cervical cancer screening by HPV testing must overcome regulatory hurdles and gain clinician acceptance. In the US, for example, the FDA has not approved self-sampling for HPV testing, and only women who test positive need a second test to guide treatment, such as cytology, because of the sensitivity of HPV testing in clinics. "It is possible that molecular tests to clarify which HPV-positive women need treatment might one day be validated on self-sampled specimens as well," Schiffman said. "Perhaps, this [study] and the overall Dutch experience could motivate a commercial application for this indication."

Each year in the US alone, there are about 13,000 new cases of cervical cancer and more than 4,000 women die from the disease, according to the American Cancer Society. At least 93 percent of these cases could be prevented with a combination of screening and vaccination, the US Centers for Disease Control and Prevention has noted.

Berkhof said he believes that the Netherlands study and its results could have implications for screening women for HPV on a global basis. Each country needs to make its own decision about how to best implement HPV screening and testing, he said. "I think in many countries, it is easier to set up self-sampling than clinician sampling," he added.

In countries that already offer screening in clinics, women could be given the option of doing self-collection or attending clinics, he said.  

"Self-collection has potential to become a first screening test in many different countries," Berkhof said. "It creates a high participation rate, which is difficult to achieve. We have cleared the first hurdle, but now we need to see how we can implement this kind of screening."