Human papillomavirus testing and cervical cytology in primary screening for cervical cancer among women in rural China: comparison of sensitivity, specificity, and frequency of referral

Abstract

The causal relationship between persistent high-risk human papillomavirus infection and cervical cancer is widely accepted. HR-HPV DNA testing, alone or in combination with Pap smear testing, may have a role in primary screening. The screening results (VIA, VILI, Pap, and HR-HPV DNA) of 9,057 women in rural China were analyzed to determine the screening performance for the detection of CIN3+. All screening strategies had comparable AUCs (0.9). Cotesting strategies had the overall highest sensitivity for CIN3+ (99.4%), followed by HR-HPV DNA testing alone (96.3%), Pap alone (80.2%), and reflex testing (75.4%). Reflex testing had the highest specificity (96.7%), followed by Pap alone (93.3%), HR-HPV DNA testing alone (85.5%), and both cotesting strategies (LSIL: 84.8%, HSIL: 84.8%). Of the single-test strategies, HR-HPV DNA testing had a higher sensitivity (96.3% vs. 80.2%) compared with Pap testing. The specificity of the Pap test was higher (93.3% vs. 85.5%) and it had a lower percent referred for colposcopy (7.8% vs. 15.8%) than HR-HPV DNA testing. HR-HPV DNA testing with a 10.0 cutoff point (relative light units/cutoff ratio) had a sensitivity (85.2%) and specificity (90.6%) estimate comparable to Pap testing. A single-test primary screening strategy with adequate performance would permit less frequent screening and be most appropriate. Of the primary screening strategies investigated in this setting in China, the performance of HR-HPV DNA testing with an increased cutoff-point might best meet these criteria.