Cervical cancer remains one of South Africa’s biggest women’s health problems. Although there is a national screening policy, screening using cervical cytology is mainly done on an opportunistic basis reaching only a 13% coverage. Lack of accuracy and reproducibility of cytology screening combined with the low access to health care and the high incidence of HIV-positivity further reduces the effectiveness of cervical screening. Therefore new, objective diagnostic/screening methods should be evaluated. This study aims to evaluate the value of testing for several biomarkers, including hrHPV testing and DNA methylation marker testing, alone or in combination, as an alternative diagnostic/screening method to detect cervical cancer in a high risk South African population.
466 women have been enrolled in this cross-sectional study: 73 were referred for treatment because of HSIL, 37 were referred for clinical staging because of a newly diagnosed cervical carcinoma, and 356 were referred for cervical cancer screening because of their high risk due to HIV-positivity. Liquid-based cytology (LBC) samples were collected at the start of their visit (before treatment, staging or screening, respectively) for testing by cytology, hrHPV presence and DNA methylation of CADM1, MAL and miR124-2 genes. A colposcopy-directed biopsy was taken in all women who came for screening. Performance of cytology, hrHPV and methylation marker testing for detection of CIN3+ was compared.
Presently, some data of 405 women have been collected. In the HIV+ screening population (n=356), primary hrHPV testing showed a high sensitivity for CIN3+ (90.3% (95% CI 79.9% - 100%), at a specificity of 60.5% (95% CI 54.5% - 66.4%). Cytology revealed lower sensitivity (74.2% (95% CI 58.8% - 89.6%), but higher specificity (78.7% (95% CI 73.7% – 83.7%). Methylation analyses at different threshold settings are currently being performed. Preliminary data indicate that all women with cervical cancer score methylation positive. Further data and combined analyses of the various markers will be presented.
In this high risk population, sensitivity of primary hrHPV testing for the detection of CIN3+ is high, but at the cost of a low specificity. Cytology had lower sensitivity but higher specificity for CIN3+ testing. Preliminary results show that full molecular testing for CIN3+ in a low/middle income country is feasible.