Human Papillomavirus testing using self-collected cervicovaginal samples vs doctor/nurse-collected samples for cervical cancer screening in women attending gynaecology clinics at a tertiary hospital in Pretoria, South Africa
PhD summary
Background. Cervical cancer is a significant public health problem in South Africa. In 2017, theSouth African National Department of Health (NDoH) Cervical Cancer Prevention and ControlPolicy was revised. The policy recommended a scaled-up approach to HPV-DNA testing andthe investigation of self-sampling.
Aims: The research investigated the performance ofapplicator tampon self-collected samples for human papillomavirus (HPV) deoxyribonucleicacid (DNA) and hr-HPV E6/E7 messenger ribonucleic acid (mRNA) testing, using doctor/nursecollected samples as the gold/reference standard, the experiences, and preferences of selfsampling; and investigated the prevalence of high-risk (hr) HPV and hr-HPV E6/E7 mRNA, andfactors associated with hr-HPV infection.
Methods: The research design was an analytic crosssectional approach. A total of 527 participants were recruited and enrolled in the GyneOncology and TOP (termination of pregnancy) clinics at a tertiary hospital, Dr George MukhariAcademic Hospital (DGMAH). Participants collected self-samples using an applicator tampon,with immediate preservation, into the saline buffer solution provided in the SelfCerv kt andthen transferred to a ThinPrep PreservCyt solution. A healthcare worker later collected asecond sample using a Cervex Brush® Comb and rinsed the brush into ThinPrep PreservCytsolution. Demographic and clinical characteristics were collected using a self-administeredquestionnaire. The questionnaire also included a section on the experiences and preferencesof self-sampling using the SelfCerv applicator tampon. All samples were tested for hr-HPVusing the Abbott HR HPV assay. Samples positive for the 14 hr-HPV types were tested for hrHPV E6/E7 mRNA using the Aptima HR HPV assay.
Results: In this research, there was nostatistical difference in hr-HPV DNA detection rate between the two collection methods(48.0%; 95%CI: 43.8-52.4 vs 47.6%; 95%CI: 43.3-52.0, p-value=0.90). However, for hr-HPVE6/E7 mRNA, there was a statistically significant difference between the two methods, 78.6%(95%CI: 73.4–83.2) vs 58.6% (95%CI: 52.6–64.4), p<0.0001) (Chapter 3). The agreement levelwas good (87.1%; κ: 0.74) between the two methods for hr-HPV DNA testing. However, a fairagreement for hr-HPV E6/E7 mRNA testing (70.2%; κ: 0.34) was observed. Self-sampling usingthe SelfCerv for hr-HPV testing was found to be comfortable by the majority (90.5%), painless(86.7%), and highly preferable by many as an alternative method for future screening (87.1%)(Chapter 4). About half (48.1%; 95%CI: 43.8-52.4) of the study population were infected withhr-HPV, and the prevalence of hr-HPV E6/E7 mRNA was substantially higher (84.2%; (95%CI:79.1-88.5) in hr-HPV DNA-positive samples. Being 40 to 49 years was significantly associatedwith an increased risk (POR: 1.38; 95%CI: 1.20–1.60) of hr-HPV E6/E7 mRNA expression. Basedon multivariate binomial regression analysis, unemployment (POR: 1.50; 95%CI: 1.23–1.83)and being married (POR: 0.61; 95%CI: 0.47–0.81) were identified as statistically significant(p<0.0001) predictive and protective factors, respectively, for hr-HPV infection (Chapter 5).
Conclusions: Overall, this research shows that the utility of SelfCerv for self-sampling mightbe a promising modality for clinic-based settings. This will encourage several women visitingclinics to participate in cervical cancer screening. However, this device might not be suitablefor populations that are harder to reach (rural locations); hence, alternative strategies shouldbe assessed. The high prevalence of hr-HPV E6/E7 mRNA indicates active HPV infection and,thus, a greater risk of developing cervical disease. It is essential to investigate the use of hrHPV E6/E7 mRNA as a primary screening test in South Africa. These findings support theimplementation of self-sampling as an alternative to healthcare worker sampling to reducecervical cancer incidence and mortality. The method offers women privacy during screening,which could increase cervical cancer screening coverage in the country. Furthermore, data onassociated factors could be necessary for strengthening screening programs.