Abstract
BACKGROUND: This study aimed to evaluate the performance of two different principles of HPV testing in primary cervical cancer screening and ASC-US triage in rural areas. METHODS: 3,328 and 3,913 women were enrolled in Shanxi, China in 2017 and 2018, respectively, and screened using liquid-based cytology and different HPV tests with a 4-year follow-up. Different screening methods commonly used in clinical practice were evaluated. RESULTS: In the HPV PCR test cohort, the prevalence of HPV infection was 14.90%. A total of 38 cases of CIN2+ were identified at baseline, 2 of which were in the HPV-negative cohort and the rest in the HPV-positive cohort (2 = 186.85, p < 0.001). Fifty-three cases of CIN2+ were accumulated over 4 years. The HPV infection rate in the HPV DNA chip test cohort was 21.10%. A total of 26 CIN2+ cases were identified at baseline, all in the HPV-positive population (2 = 92.96, p < 0.001). 54 CIN2+ cases were cumulative over 4 years. At 4-year follow-up, HPV-negative results were significantly more protective against cervical intraepithelial neoplasia grade 2 or worse (CIN2+) than normal cytologic results at baseline. HPV screening was more sensitive and specific than cytologic screening (using ASC-US as the threshold) and performed better on the HPV DNA microarray test. In addition, compared with HPV 16/18 testing, sensitivity increases and specificity decreases when using HPV testing for cytologic ASC-US triage, regardless of which HPV test is used. CONCLUSION: In the rural areas where we implemented the study, HPV tests performed well for screening than LBC and HPV DNA chip testing performed better than HPV PCR testing in the screening cohort. Optimal screening was achieved technically when used in combination with LBC for ASC-US population triage, without thinking the feasibility for resource availability.