Concordant testing results between various human papillomavirus assays in primary cervical cancer screening: systematic review

• Published in: Clinical microbiology and infection Vol. 24; no. 1; pp. 29 - 36
• Main Authors: de Thurah, L., Bonde, J., Lam, J.U.H., Rebolj, M.
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 01.01.2018
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Assays; Cervical cancer; Concordance; Female; Human papillomavirus; Humans; Mass Screening - methods; Middle Aged; Molecular Diagnostic Techniques - methods; Papillomaviridae - genetics; Papillomavirus Infections - diagnosis; Papillomavirus Infections - epidemiology; Reagent Kits, Diagnostic - statistics & numerical data; Screening; Uterine Cervical Dysplasia - diagnosis; Uterine Cervical Dysplasia - epidemiology; Uterine Cervical Dysplasia - virology; Uterine Cervical Neoplasms - diagnosis; Uterine Cervical Neoplasms - epidemiology; Uterine Cervical Neoplasms - virology; Young Adult; Screening; Human papillomavirus; Concordance; Cervical cancer; Assays
• ISSN: 1198-743X; 1469-0691
• DOI: 10.1016/j.cmi.2017.05.020

Human papillomavirus (HPV) assays are increasingly used for primary cervical screening and HPV-vaccination-effect monitoring. We undertook a systematic literature review to determine the concordance in positive test results (i.e. detection of HPV infections) between Hybrid Capture 2 (HC2) and other assays. We searched PubMed, Embase and Scopus for studies of primary screening with HC2 and one or more assays, with cross-tabulated testing results for the assays. Two authors applied inclusion criteria and three authors extracted data from included studies. For each inter-assay comparison, we calculated the concordance by comparing the number of concordant samples with the number of samples that tested positive on at least one assay. Sixteen studies fulfilled inclusion criteria, comparing nine assays to HC2, and including 392 to 9451 patients each. The calculated concordance varied between 48% and 69% for HC2 and APTIMA, Cobas, Abbott RealTime, Cervista, GP5+/6+, CLART, BD HPV test, Amplicor and Linear Array, i.e. 31%–52% of all positive tests on any pair of compared assays were discordant. Although modest variation in the degree of concordance with HC2 was suggested for particular assays, the numbers of studies per assay were generally low. No pronounced systematic patterns were observed by study (e.g. liquid medium) or population characteristics. The ten commercially available assays do not detect the same HPV infections. Even in the most favourable case, the two assays provided discordant test results in 31% of all detected infections.