Utility of extended HPV genotyping as primary cervical screen in an unscreened population with high HIV co-infection rate

Abstract

OBJECTIVE : Screening with primary human papillomavirus (HPV) testing has been evaluated in highly prescreened populations with lower HPV and HIV prevalence than what is the case in South Africa. High prevalence of HPV and underlying precancer in women living with HIV (WLWH) affect the clinical performance of screening tests significantly. This study investigates the utility and performance of an extended genotyping HPV test in detection of precancer in a population with a high coinfection rate with HIV. METHODS : A total of 1,001 women aged 25 to 65 years with no cervical cancer screening in the preceding 5 years were tested with cytology and primary extended genotyping HPV testing. The cohort of 1,001 women included 430 WLWH (43.0%) and 564 HIV-negative (56.3%) women. RESULTS : Abnormal cytology (atypical squamous cells of undetermined significance or higher) was significantly higher in WLWH (37.2% vs 15.9%) and high-grade squamous intraepithelial lesion or above (23.5% vs 5.2%). The WLWH also tested positive more often for any HPV type (44.3% vs 19.6%; p < .0001) The specificity for cervical intraepithelial neoplasia 2+ at 91.2% of a combination of HPV types, 16/18/45 (very high risk) and 31/33/58/52 (moderate risk), performed better than cytology or any HPV-positive result to predict cervical intraepithelial neoplasia 3+ on histology. The additional genotype information supports direct referral to treatment or colposcopy in a larger proportion of the screen-positive population. CONCLUSIONS : The potential contribution of extended genotyping is demonstrated. The ideal choice of sensitivity and specificity ultimately depends on the health budget. More information will allow a screening algorithm, guiding management according to risk.