oa Southern African Journal of Gynaecological Oncology - Human papillomavirus DNA testing on self-collected vaginal tampon samples as a cervical cancer screening test in a Gauteng population : original research
Background : There is a need to simplify cervical cancer screening to reach more women. Tampon-collected specimens can be tested using molecular methods, but this type of self-screening has not been properly evaluated as a screening method in South Africa before. The objective of this study was to evaluate human papillomavirus (HPV) DNA testing of self-collected tampons as a screening method in an urban and peri-urban population in Gauteng by comparing the results with the current standard of conventional cytology. In addition, HPV prevalence, type, distribution and incidence of cytological abnormalities in this population are described.
Method : Seven hundred and twenty women attending public healthcare facilities in and around Tshwane, Gauteng province, were invited to participate. The women collected a tampon sample for molecular testing, and were then screened by healthcare workers collecting a conventional cervical cytology smear. HPV testing was undertaken using the Linear Array® HPV Genotyping Test (Roche Molecular Systems).
Results : Data for analysis were available for 631 women. Three hundred and fifty-four (58%) were positive for high-risk HPV, while (15.4%) had an abnormal cytology result. Women aged 30-39 years had the highest prevalence of both high-risk HPV (75%) and abnormal cytology (22%). Infection with multiple types was common. Higher-risk viruses were not over-represented in, and no dramatic decrease in HPV prevalence was observed in, older women. Cytological abnormalities were detected in only 3.74% of women who tested negative for high-risk HPV, but were found in 24.2% of high-risk HPV positive women.
Conclusion : HPV testing on self-collected tampon samples was feasible, highly sensitive and demonstrated a high negative predictive value for current cytological abnormalities in this population.
Article metrics loading...