Persistent infection with high-risk strains of human papillomavirus can cause cervical cancer, and uninfected individuals have a low risk of developing the disease. As such, validating assays purporting to detect HPV infection is seen as an essential step in bringing such tests to the public.
A group in the Netherlands recently showed that a PCR-based assay for 15 high-risk HPV types, called HPV-Risk, has comparable sensitivity and specificty to a previously validated assay. Their study, published this month in The Journal of Clinical Microbiology, also demonstrated that the assay — currently commercially available from Dutch company Self-screen — produced similar results when run on samples collected by a clinician versus those collected by women themselves.
Most PCR-based assays target the E1 or L1 open reading frames of HPV, but the study authors suggested integration of the virus into the host genome often takes place between these regions, potentially disrupting PCR and leading to a false-negative test result. According to the study, HPV-Risk is a novel real-time PCR assay because it targets the E7 region instead.
Self-screen is a five-year-old biotech spinout of the VU University Medical Center in Amsterdam, according to the company's website. Its HPV-Risk test was developed to target a 150-bp fragment of the E7 region of 15 high-risk HPV types. It provides individual read-outs of HPV16 and HPV18 (the two most carcinogenic strains), and reports 13 other strains as a pooled result.
According to the study's authors, one of whom is a member of the scientific staff of Self-screen, the test is also novel because it includes the HPV67 strain not measured in other tests. In addition, according to Self-screen's website the assay is compatible with a variety of nucleic acid extraction procedures — such as those offered by Qiagen, Roche, Macheri-Nagel, Biomérieux — and can be run on Life Technologies' ABI 7500 Fast or Life Tech ViiA7 real-time PCR systems; Bio-Rad's CFX96; or equivalent platforms.
In the JMD study, the researchers first measured the assay's sensitivity and specificity on DNA from plasmids containing complete genomes of a number of HPV genotypes, serially diluted in a background of human placental DNA to mimic cervical specimens. They found the test was comparable to a previously clinically validated assay, L1-based GP5+/6+-PCR, in detecting cervical intraepithelial neoplasia of grade 2 or worse (CIN2+). HPV-Risk showed a clinical sensitivity of 97.1 percent and specificity of 94.3 percent, which the authors deemed non-inferior to the reference assay.
The work was then expanded to a collection of 1,444 clinician-obtained cervical samples. These samples included 70 that were histologically confirmed CIN2+, 20 of which had had normal cytology but tested HPV-positive with the reference assay. Here again, they showed the new test to be non-inferior. Importantly, they also demonstrated high intra-lab reproducibility and inter-lab agreement with the new test.
Finally, the authors took an additional step of determining whether HPV-Risk could also reliably detect HPV in self-collected samples. They compared clinician-collected samples to those that women collected themselves using either lavage- or brush-based collection devices. The self-collected cervico-vaginal specimens showed high agreement with the clinician-taken cervical scrapings from the same subjects, for both self-sampling methods, using the HPV-Risk assay.
Self-sampling is a developing practice that has yet to be widely accepted. A meta-analysis published this week in The Lancet Oncology attempted to determine whether self-collected samples were as accurate as clinician-collected for HPV testing, using data from 36 studies on a total of 154,556 women.
The selected studies were conducted between 1990 and 2013, and included many types of collection devices (swabs, spatulas, brushes, lavage, and tampon-like devices). Overall, self-samples were then tested using more than a dozen different HPV assays, including Qiagen's HC2, Gen-Probe's Aptima, as well as tests from Abbott and Roche. The authors concluded that, for signal-based assays, sampling by a clinician should be recommended. HPV testing on a self-sample could prove an additional strategy to reach women not participating in regular screening programs, they said, and "some PCR-based HPV tests could be considered for routine screening after careful piloting assessing feasibility, logistics, population compliance, and costs."
However, in a commentary accompanying the Lancet Oncology paper, other researchers argued that this should not dissuade development of self-collection methods. They pointed out that, in the meta-analysis, "when PCR-based HPV tests were used, the relative sensitivity was similar to clinician-collected samples in all cases, and the relative specificity was similar in six of seven of the PCR-based tests used, suggesting that use of a test with a higher analytic sensitivity is needed to ensure similar accuracy between clinician-collected and self-collected samples."
The authors also reinforced the unique benefits of self-collection, asserting "screening with either method is successful only when most of the eligible women participate. Self-sampling is a promising new screening alternative for women who are reluctant or unable to attend a healthcare facility for routine cervical cancer screening. Such strategies are already being broadly assessed … as alternatives in low-income and middle-income countries that have failed to successfully introduce Pap testing ... [and], in resource-poor scenarios with substantial cultural diversity, self-collected sampling has been shown to be well accepted."
As a cervical cancer screen, HPV testing recommendations are somewhat variable among experts. The US Centers for Disease Control and Prevention has combined the recommendations of the American Cancer Society, US Preventative Services Task Force, and American College of Obstetrician and Gynecologists into a chart to guide physicians ─ in general, for women ages 30 to 65 (or, about 72 million US women, according to 2012 US census bureau data), they suggest cytology via Pap test every three years, with HPV co-testing every 5 years. Adopting this recommendation could amount to millions of HPV tests per year, with self-testing potentially expanding the market by recruiting women who might normally avoid the doctor's office. A recent Cochrane review also suggested that HPV self-tests could be used as follow-up, and be a more efficient triage than repeated Pap tests for women who have minor cervical lesions.
HPV-Risk is currently commercially available from Self-screen in Europe, but not in the US. According to the company's website, it has also developed a test called PreCursor-M, a real-time PCR assay to help distinguish women with cervical pre-cancer by detecting hypermethylation of three disease-causing genes.