Background: High-risk human papillomavirus (hrHPV) testing has higher sensitivity but lower specificity than cytology for cervical (pre)-cancerous lesions. Therefore, triage of hrHPV-positive women is needed in cervical cancer screening.
Methods: A cohort of 1,100 hrHPV-positive women, from a population-based screening trial (POBASCAM: n = 44,938; 29-61 years), was used to evaluate 10 triage strategies, involving testing at baseline and six months with combinations of cytology, HPV16/18 genotyping, and/or repeat hrHPV testing. Clinical endpoint was cervical intraepithelial neoplasia grade 3 or worse (CIN3(+)) detected within four years; results were adjusted for women not attending repeat testing. A triage strategy was considered acceptable, when the probability of no CIN3(+) after negative triage (negative predictive value, NPV) was at least 98%, and the CIN3(+) risk after positive triage (positive predictive value, PPV) was at least 20%.
Results: Triage at baseline with cytology only yielded an NPV of 94.3% [95% confidence interval (CI), 92.0-96.0] and a PPV of 39.7% (95% CI, 34.0-45.6). An increase in NPV, against a modest decrease in PPV, was obtained by triaging women with negative baseline cytology by repeat cytology (NPV 98.5% and PPV 34.0%) or by baseline HPV16/18 genotyping (NPV 98.8% and PPV 28.5%). The inclusion of both HPV16/18 genotyping at baseline and repeat cytology testing provided a high NPV (99.6%) and a moderately high PPV (25.6%).
Conclusions: Triaging hrHPV-positive women by cytology at baseline and after 6 to 12 months, possibly in combination with baseline HPV16/18 genotyping, seems acceptable for cervical cancer screening.
Impact: Implementable triage strategies are provided for primary hrHPV screening in an organized setting.