RESUMO
OBJECTIVES: To determine a management strategy for women testing negative with cervical cytology and positive for high-risk human papillomavirus (HR-HPV). METHODS: Using the data from the large population-based Shenzhen Cervical Cancer Screening Trials II and III (SHENCCAST II/III), we compared the risk for cervical intraepithelial neoplasia grade 3 or cancer (CIN 3+) in women with negative cytology but testing positive for HR-HPV DNA using Cervista HPV HR or matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF), followed by secondary screening with type-specific Cervista HPV-16/18 or MALDI-TOF. The study aim was to seek the most sensitive and specific triage assay for referral for colposcopy. RESULTS: A total of 8,556 women had complete data. The proportion of women with negative cytology and positive HR-HPV by Cervista HR-HPV (5.30%, 453/8,556) was slightly lower than that of women with negative cytology and HR-HPV-positive tests by MALDI-TOF (5.82%, 499/8,556, p = .015). The proportion of women having negative cervical cytology and a positive HR-HPV by Cervista HR who have HPV-16 and/or -18 by Cervista HPV-16/18 (11.8%, 53/448) was less than that of women with a negative cervical cytology and positive HR-HPV by MALDI-TOF who have HPV-16 and/or -18 by MALDI-TOF (19.4%, 97/499, p = .001). The proportion of CIN 3+ within negative cervical cytology and positive HR-HPV that were HPV-16 and/or -18 for the Cervista 16/18 assay (61.5%, 8/13) was similar to that for the MALDI-TOF 16/18 assay (66.7%, 10/15, p = 0.8). CONCLUSIONS: In the cytology-negative HR-HPV-positive population, Cervista 16/18 as the HPV detection method would refer 11.8% of women for colposcopy and diagnose 61.5% of the CIN 3+, while MALDI-TOF16/18 would refer 19.4% and diagnose 66.7% of the CIN 3+. Cervista HPV-16/18 seems to be the superior triage test. However, in resource-limited settings, an assay that includes 16/18 genotyping in the primary result (rather than a second test) may be more cost efficient.
