HPV DNA testing and treatment strategies for cervical pre-cancer among women living with HIV in Papua New Guinea: a prospective clinical cohort study

BACKGROUND: The World Health Organization recommends human papillomavirus (HPV) testing and same-day treatment for cervical screening among women in the general population in low- and middle-income countries (LMIC). Among women living with HIV, additional triage is recommended due to lower specificity and predictive value of HPV testing alone in this population. We evaluated the performance of HPV DNA testing alone and in combination with alternative triage strategies for the early detection and treatment of cervical pre-cancer among women living with HIV in Papua New Guinea (PNG). METHODS: We did a prospective clinical cohort study among women living with HIV aged 30-59 years attending an HIV treatment and care service. Eligible women self-collected a vaginal specimen for point-of-care HPV DNA testing on the GeneXpert platform. All women testing HPV-positive, and a random sample of HPV-negative women, had enhanced visualisation of the cervix with digital colposcopy (EVA) and provided a clinician-collected cervical specimen for off-site liquid-based cytology. The primary outcome was the clinical performance (sensitivity, specificity, positive and negative predictive value [PPV, NPV]) of HPV testing alone and in combination with alternative triage strategies (EVA, HPV type restriction, modified assay cut-off values) for detection of high-grade squamous intraepithelial lesion (HSIL) or worse. RESULTS: Between May 3, 2024 and December 12, 2024, we recruited 404 women, of whom 168 (41.6%) tested positive for one or more of 14 oncogenic HPV types. Estimated sensitivity and NPV for HSIL or worse using manufacturer-defined cycle threshold cutoffs was >99%, specificity 63.3% (95%CI: 58.3%-68.3%), and PPV 18.6% (12.6%-24.6%). Restricting screen-positive to one or more of the 8 HPV types of greatest oncogenic potential had comparable sensitivity and NPV (>99%), with specificity 70.2% (65.5%-74.9%) and PPV 22.6% (15.6%-29.6%). HPV 8-type restriction with more stringent assay cut-offs (cycle threshold <33.99) had sensitivity 86.5% (75.3%-97.7%), specificity 80.1% (75.9%-84.3%), PPV 30.7% (21.7%-39.7%), and NPV 98.4% (96.9%-99.9%). EVA provided no additional benefits to HPV testing alone. CONCLUSIONS: Applying HPV 8-type restriction with or without modified manufacturer assay cut-off values to the WHO pre-qualified Xpert HPV test improved clinical performance for detection of HSIL or worse among women living with HIV in PNG.