Research Of Hiv-1 Screening Strategy And Estimation Of Incidence Among Msm Population In China

BackgroundIn recent years, HIV surveillance system has been constantly improved and played an important role in identification of HIV infections. However, a great number of HIV infection persons were not discovered and window period infections are difficult to diagnose. To screen out HIV infection person, especially window period infection as early as possible is one of the urgent problems to be solved in HIV control and prevention. Identification of window period infections is important to investigate emerging patterns of HIV transmission prospectively and reduce the spread of drug-resistant strains because acute infections are at high risk for secondary spread of HIV infection. Thus, considering current HIV epidemic situation, it is important to develop HIV detection strategy of early infection among high-risk population in China.Recently, HIV prevalence data show the increasing burden of HIV among Men who have sex with men (MSM) in China. It is critical to develop effective control and prevention strategy and better understand the HIV epidemics in MSM population. Cohort study was gold standard for HIV incidence surveillance, though was affected by follow-up rate and selection bias. Therefore, it is important to develop laboratory method using cross-sectional sample for HIV incidence estimation. Among various HIV detection methods for recent infection, BED-CEIA method as a reproducibility and stability assay, has been widely used to estimate the HIV-1 incidence. However, the BED method might be influenced by CD4+ T cell count, antiretroviral treatment use and advanced clinical disease. Therefore, we also focus on other preferable laboratory methods for detection of HIV recent infection. Objectives1. To establish HIV screening strategy for HIV early infection among MSM population in China;2. To develop new laboratory method for detection of HIV-1 recent infection.Methods1. From April,2008 to September,2009,11023 participants were enrolled from MSM population in Heilongjiang, Liaoning, Neimenggu, Ningxia and Qinghai province in China. From November 2008 to September 2009,2235 subjects were from a cohort survey in MSM population conducted in You-an Hospital.2. After standard testing with current HIV screening strategy, HIV antibody-negative specimens were screened using two strategies, including pooling NAAT and 4th generation ELISA. The estimated prevalence, sensitivity, specificity, false positive rate, false negative rate, positive predictive value and negative predictive value of the three HIV screening strategies were calculated.3. After standard testing with current HIV screening strategy, HIV positive specimens were screened with BED IgG capture enzyme immunoassay (BED-CEIA) and analyse the western blot band pattern for estimating the mean window period of seroconversion of p31 antigen and calculating the HIV-1 incidence based on two methods, respectively. HIV negative and indeterminate specimens were pooled and screened for HIV using NAAT. HIV window period infections were followed up. The mean duration of HIV RNA prior to seroconversion and was calculated using the follow-up specimens. The HIV-1 indicences were estimated by BED-CEIA assay, pooling NAAT and p31 antigen menthod seperately.Results1.282 of 11 023 samples were HIV antibody-positive,25 were indeterminate.28 were RNA positive.10 716 samples were screened by pooling NAAT and 4th generation ELISA, respectively.28 were antibody-negative but RNA positive on the pooling NAAT test; 100 were antibody-positive on the 4th generation ELISA test, and 40 were WB indeterminate and 60 were WB negative. Of the 40 WB indeterminate samples,15 were HIV RNA positive; of the 60 WB negative,0 was HIV RNA positive. The viral load of 28 window period infections ranges from 3.84～6.84Log10Copies/mL. Among 28 window period infections, the viral load of 10 samples were 106Copies/mL (35.71%),9 samples were higher than 105 Copies/mL (32.14%),7 samples were 104 Copies/mL (25%).2. Current HIV screening test missed 28 window period infections, and 4th generation ELISA missed 13 window period infection. HIV acute infection prevalence was 2.54‰. the HIV positive rate estimated by '3rd generation ELISA+WB','3rd generation ELISA+4th generation ELISA+WB','3rd generation ELISA+WB+ pooling NAAT'were 2.56%(95%CI=2.26%-2.86%),2.70%(95%CI 2.39%-3.01%),2.82%(95%CI=2.50%-3.13%) and 2.82%(95%CI=2.50%-3.13%), respectively. The differences were not significant (P=0.8261). The sensitivity of three strategies was 90.97%,95.81% and 100%, respectively. The specificity was 100%.3.92 of 2235 subjects were HIV infected,82 were antibody positive; 10 were antibody negative but RNA positive. All of the 20 acute infections were informed,2 were lost to follow-up,18 were offered partner counselling and referral services, and enrolled in medical care. And the 18 patients were EIA and WB positive. HIV viral load and CD4 T cell count of acute infections was measured. The mean duration of HIV RNA prior to seroconversion was 28.9 days among MSM population in China. The mean window period of seroconversion of p31 antigen was estimated 38.9 days by results of western blot pattern for follow-up specimens. The estimated HIV-1 incidence was 5.36%(95% confidence interval [CI]=3.95%-6.78%),5.55%(95%CI=2.11%-8.99%) and 5.88% (95%CI=2.80%-8.96%) based on BED-CEIA assay, pooling NAAT and p31 antigen method, respectively.4. The HIV-1 incidence estimated by pooling NAAT method among MSM population in Heilongjiang, Liaoning and Qinghai province were 2.92%(95%CI=1.01%-4.83%), 3.97%(95%CI=1.82%-6.13%) and 4.73%(95%CI=0.00%-10.08%), respectively; and the BED-CEIA estimated HIV-1 incidence were 2.36%(95%CI=1.63%-3.08%), 2.86%(95%CI=2.17%-3.55%) and 4.35%(95%CI= 0.09%-8.62%), respectively. The HIV-1 incidence base on pooling NAAT method using new estimatedμvalue (duration of RNA positive prior to seroconversion) were 2.81%(95%CI 0.97%-4.65%) 3.82%(95%CI= 1.75%-5.91%) 4.55%(95%CI= 0%-9.70%), and the differences were significant (p=0.0186)ConclusionsHIV-1 detection strategy of early detetion among MSM population:1. After standard testing with ELISA and Western blot confirmation, HIV antibody-negative samples were screened by pooling NAAT or 4th generation ELISA to effectively screen out window period infections, which could be adopted as HIV-1 detection strategy of early detetion among MSM population;2.'3rd generation ELISA+WB+pooling NAAT'was more effective than 4th generation ELISA for screening out window period infections;3. Combination of 4th generation ELISA with WB could partly replace pooling NAAT as a sensitive screening test in HIV negative population to evaluate HIV epidemic situation.HIV-1 incidence surveillance among MSM population:1. Preliminary research indicated p31 antigen could be useful in HIV incidence estimation, though further study needed for putting the method into practice;2. Preliminary research showed the HIV incidece estimates obtained from BED-CEIA assay and pooling NAAT are similar. The influenced factors of fomula for estimating HIV-1 incidence were size of positive pool and mean window period of RNA positive prior to seroconversion.