Abstract
OBJECTIVE: Human papillomavirus (HPV)-associated cancers are important public health problems in HIV-infected people. Assays based on HPV virus-like particles (VLP) and pseudoviruses (PsV) are commonly used to examine HPV antibody responses in HIV-infected people, but neutralization assays with native HPV have not been utilized and a comparison of these three assays is lacking. We evaluated the agreement of assays using VLP, native HPV and PsV in detecting HPV16 and 18 antibodies in HIV-infected women. METHODS: The VLP-based ELISA (VLP-ELISA) was used to detect antibody responses to HPV16 and 18 and cottontail rabbit papillomavirus (CRPV) VLP antigens. Neutralization assays with native HPV (NA-HPV) and with PsV (NA-PsV) were conducted to examine HPV16 or 18 neutralizing antibodies. Intra class correlation coefficients (ICC) and kappa coefficients were used to assess the agreements of seropositivity between the assays. RESULTS: The seroprevalence detected by the VLP-ELISA, NA-HPV and NA-PsV in 94 HIV-infected women was 35%, 51% and 27% for HPV16 and 14%, 44% and 21% for HPV18. Cross-reactivity between HPV16 and HPV18 was 0.35, 0.04 and 0.33 (kappa coefficients) for the VLP-ELISA, NA-HPV and NA-PsV. The agreements of seropositivity between the three assays were low. Six women who were HPV16 DNA positive were seropositive by the NA-HPV but only two were HPV16 seropositive by the VLP-ELISA or NA-PsV. One HPV18 DNA positive woman was seropositive by all three assays. Repeated tests indicated excellent reproducibility of the NA-HPV. CONCLUSION: HPV serology results vary across different assays. The NA-HPV appears to be a sensitive and reliable approach in detecting natural HPV antibodies in HIV-infected women. The NA-HPV can be applied in both HPV natural history studies and vaccine studies in HIV-infected people.