PALM SPRINGS, CALIF. — Women who do not participate in routine screening programs may prefer self-testing for human papillomavirus with a tampon or cotton swab over testing administered by a physician, Dr. Robert P. Edwards reported in a poster session at the annual meeting of the Society of Gynecologic Oncologists.
“This is a feasible approach to use for women not currently participating in screening,” Dr. Edwards, the study's lead author, said later in an interview. “The concept of the study was to try to evaluate women who are not currently participating in screening, to see if there is some way we can identify women who might need extra resources to get them into a physician's office.
“For instance, elderly black women over the age of 65 have about six to seven times the incidence of cervical cancer compared with the remainder of the population. It's largely because they're not participating in any screening programs.”
For the study, 171 women aged 18 years and older with an abnormal referral Pap smear were recruited on referral to the colposcopy clinic at the University of Louisville (Ky.). Women who had a prior hysterectomy were excluded from the study, as were those who were pregnant, were taking corticosteroids, were HIV positive, or were taking immunosuppressants.
Study participants performed HPV-DNA self-sampling by using a vaginal swab first and then a vaginal tampon. Then they filled out a questionnaire that asked them about the acceptability of each method compared with previous screening experiences.
A physician then performed a Pap smear and colposcopy and researchers collected data on the sensitivity, specificity, predictive power, and overall test efficiency of each of the three collection methods. The colposcopy results were used as the preferred method, said Dr. Edwards, professor of obstetrics and gynecology at the University of Pittsburgh.
“We were looking to see how well the tests would hold up against physician-directed HPV testing in a population where we knew there would be disease,” he said.
Colposcopy biopsy confirmed that 16% of patients had cervical intraepithelial neoplasia (CIN) 2 and 3; 24% had CIN 1; 1% had cervical cancer, 26% had a normal biopsy, 20% had inflammation, and the rest did not have any cervical biopsies done due to a negative colposcopy.
The researchers found that the tampon collection method was 66% sensitive, 62% specific, and had a positive predictive value of 60% and an overall test efficiency rate of 64%.
The swab collection method was 55% sensitive, 63% specific, and had a positive predictive value of 62% and an overall test efficiency of 63%.
Meanwhile, the physician-collected method was 90% sensitive, 43% specific, had a positive predictive value of 58%, and an overall test efficiency of 65%.
Dr. Edwards says he was surprised that the self-sampling methods were generally more specific than the physician-collected method.
“I would have expected the physician test that's taken directly from the cervix would perform better in that regard,” he said. “That may just be an aberrancy of this particular study.”
He said he was also surprised that 51% of patients preferred self-sampling with a swab while 46% preferred self-sampling with a tampon. “We picked the tampon because we figured women would feel more comfortable inserting that into the vagina,” he said. “But in actuality the swab was just as acceptable.”
The tampon did produce a 10-fold higher cell yield compared with the swab.
Most women (70%) indicated they would prefer self-collection for their next screening.
The study was funded by the James Graham Brown Cancer Center at the University of Louisville.