Comparison of human papillomavirus detection and typing by hybrid capture 2, linear array, DNA chip, and cycle sequencing in cervical swab samples

Abstract

Although the Hybrid Capture II (HC II) assay can detect 13 high-risk human papillomavirus (HPVs), it does not yield any genotype-specific information. We evaluated the performance of 4 HPV DNA tests, namely, HC II, Linear Array (LA), DNA chip, and cycle sequencing for their capacity to detect the presence of high-risk HPV DNA and HPV-associated cervical lesions. Seventy-six women who were referred to the colposcopy clinic for abnormal cytology were enrolled. The women were examined using liquid-based cytology, colposcopy-directed biopsy, and HPV DNA tests. After DNA extraction from a single sample, HPV DNA tests were performed by all 4 methods on the same specimen. The LA test has higher HPV-positive rates than HC II for cervical intraepithelial neoplasia I (83.3% vs 61.1%; P < 0.01) and for cervical intraepithelial neoplasia II and more severe lesions (100.0% vs 80.0%; P < 0.01). The concordance between the DNA chip and LA tests was 89.5%, confirming substantial agreement (kappa coefficient = 0.73), and the concordance between HC II and the DNA chip was 80.3%, also showing substantial agreement (kappa coefficient = 0.738). The concordance for 15 high-risk HPV genotypes between LA and sequencing was 82.5% with a kappa value of 0.536. Furthermore, the LA test was more sensitive in the detection of high-grade cervical lesions than HC II (100% vs 92.3%, P < 0.01). The LA test showed superior sensitivity in the detection of clinically relevant HPV infections and has proven to be an accurate tool for identifying individual HPV types, especially in cases of multiple HPV infections.