The performance of human papillomavirus DNA detection with type 16/18 genotyping by hybrid capture in primary test of cervical cancer screening: a cross-sectional study in 10,669 Chinese women
X Zhao,Q Wu,X Wang,Y Fu,X Zhang,X Tian,B Cheng,B Lu,X Yu,S Lan,W Lu,D Ma,X Cheng,X Xie
DOI: https://doi.org/10.1016/j.cmi.2018.02.027
Abstract:Objectives: We aimed to assess the performance of DH3 human papillomavirus (HPV) assay, a newly developed hybrid capture technique that detects 14 high-risk HPVs with type 16/18 genotyping, as a primary test in cervical cancer screening. Methods: In total 11,356 Chinese women aged 21-65 years participated in a cervical cancer screening programme using cytology (Thinprep, Hologic) and HPV testing (Cobas 4800 Test, Roche). Residual samples were used to detect HPV by DH3 HPV. Results: In total 10,669 women with valid results were included in the study. Of those, 135 were diagnosed as CIN2+, and 83 were diagnosed as CIN3+; 1056 women (9.9%) were DH3 HPV-positive and 255 (2.4%) of those were 16/18-positive, while 990 (9.3%) women were Cobas HPV-positive and 243 (2.3%) of those were 16/18-positive. DH3 HPV was non-inferior to Cobas HPV in identifying CIN1- and CIN2+ using predetermined thresholds (both p < 0.001). The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of DH3 HPV were 93.3% (95% confidence interval [CI] = 87.7-96.9), 91.2% (95%CI = 90.6-91.7), 12.0% (95%CI = 10.1-14.1) and 99.9% (95%CI = 99.8-100), respectively, similar to those of Cobas HPV (91.1%, 95%CI = 85.0-5.3; 91.8%, 95%CI = 91.2-92.3; 12.5%, 95%CI = 10.5-14.7; and 99.9%, 95%CI = 99.8-99.9, respectively), in identifying CIN2+ (all p > 0.05). When DH3 HPV and Cobas HPV were respectively used as primary testing in screening strategy, the performance of two strategies were similar in identifying CIN2+. The results were similar in identifying CIN3+. Conclusion: Our data suggest that DH3 HPV performs similarly to Cobas HPV in identifying high-grade CIN in cervical cancer screening.