Evaluation And Applied Research Of Bed Assay In China

BackgroundOnly HIV prevallence but not incidence can be calculatd from routine HIV test. BED IgG capture enzyme immunoassay was now widely used in China and played an important role in the estimation of the HIV-1 incidence. According to the experiences and recommendations from previous studys, the window period and 4 adjustment coeffients should be needed to get a better estimation of HIV-1 incidence in China. Therefore, comprehensive study and estimation of BED assay was needed urgently. Furthermore, the assay might be influenced by low CD4+T cell count and ART through the level changes of the HIV viral load. Therefore, the further study on the effect of viral load to BED assay should be undertaken.Objectives1. To obtain the window period and 4 new correction coefficients of incidence calculation for BED assay in China and study the different impact factors like CD4+ T cell count, ART and VL on the BED results;2. To study the feasibility of taking use of DBS specimens in BED-CEIA assay and estimating the HIV-1 incidence with DBS in China;3. To track dynamic changes in HIV-1 prevalence and incidence among focal populations in HIV hard-hit areas. Methods 1.603 seroconverter specimens were obtained from two cohort study in Dehong and Youan Hospital. All of samples were tested by BED assay and viral test. And the individual data, included CD4+T cell count and ART, were also collected. ROC curve and scatter diagram was used to calculate the window period and 4 adjustment factors of BED assay in China. The effect of CD4+T cell count, ART and viral load to the BED results was also analysed.2.10,313 participants from 27 VCT sites were enrolled in this study. Both plasma and DBS specimens were collected.349 eligible pairs of HIV-1 positive specimens confirmed by WB were available for BED-CEIA and the test results were compared. Negative control, low positive control, high positive control and calibrator in DBS were used to evaluate the stability and repeatability of the detection results of DBS specimens.3. Consecutive cross-sectional surveys were conducted in focal populations. Specimens seropositive for HIV were tested with the BED IgG capture enzyme immunoassay to identify recent seroconversions using normalized optical density of 0.8 and adjustments.7175 specimens were tested from 2000 to 2009 in HIV surveillance sites in Yunnan, Xinjiang, Sichuan and Chongqing, et al. And totally 3759 specimens were test in Dehong among IDU, discordant couples, female sex workers, pregnant women and pre-marriage couples. In Liangshan Prefecture,4197 specimens were got from general population.Results1. The window period of BED assay in China should be 168 days according to the ROC curve results. The 4 new correction coefficients of incidence calculation in China were calculated and a-0.8098,0=0.7571, y=0.9315 and s-0.0685, respectively. BED incidence estimates are very sensitive to changes in the proportion of long-term false-recent cases. The BED false rate is highest when the CD4<50 and ART. Patients with a log io viral load of 4 or less than 4 were more likely to test false recent than those with higher viral loads.2. Of the total HIV positives,294 and 53 participants were tested as long-term and recent HIV infections by both DBS and plasma specimens. Two participants got different results from DBS and plasma specimens, but their final OD-n values were close to the cutoff value of BED-CEIA assay. The concordance rate between two types of specimens is 99.43% and the R2 linear of the stability and repeatability results of DBS specimens reached 0.9551 and 0.95, respectively.3. From 2004 to 2008, estimated annual HIV incidence among injecting drug users (IDUs) decreased significantly [from 15.0%(95% CI=11.4%-18.5%) in 2004 to 4.3% (95% CI=2.4%-6.2%) in 2008; trend test P<0.0001]. The incidence among other focal populations, such as HIV discordant couples (varying from 5.5% to 4.7%), female sex workers (varying from 1.4% to 1.3%), pregnant women (0.1%), and pre-marital couples (0.2 to 0.1%) remained stable. Overall, the proportion of recent HIV-1 infections was higher among females than males (P<0.0001).4. The HIV-1 incidence among general population were>1.00 in Butuo(1.27) and Meigu(1.11), and were 0.50-1.00 in Yuexi(0.66) and Zhaojue(0.60). And the HIV-1 incidence in Puge(0.47), Xide(0.20), Mianning(0.18), Ganluo(0.18) and Jinyang(0.12) were 0.10-0.50, and were<0.10 in Dechang(0.06), Huili(0.05), Huidong, Ningnan, Leibo, Yanyuan, Muli and Xichang.5. The rank of HIV incidence of different areas in IDU population is Xinjiang, Yunnan, Sichuan, Chongqing, Guangxi, Guangdong and Guizhou. The HIV incidence trends decreased or remained stable in these areas except Chongqing. And the HIV incidence from STD clinic in Yunnan decreased gradually, while the HIV incidence among MSM increased rapidly.Conclusions1.4 new coefficients should be adopted in BED tests (plasma or DBS) to reflect the real trends in HIV-1 incidence in China. The BED assay was suitable to be applied in China but relevant influence factors, such as ART and low CD4+T count cases should be excluded.2. The HIV/AIDS control and prevention methods were effective that the HIV-1 incidence among IDU had decreased or remained stable. Limited data from STD clinic indicated that the trends of HIV-1 incidence decreased. While the HIV-1 incidence among MSM increased rapidly and intensive prevention methods are needed.3. The HIV epidemic in Dehong continued to expand during a five-year period but at a slowing rate among IDUs, and HIV incidence remains high among IDUs and discordant couples. Intensive prevention measures should target sub-groups at highest risk to further slow the epidemic and control the migration of HIV to other areas of China, and multivariate analysis is needed to explore which measures are more effective for different populations.4. Further HIV control and prevention methods are urgently needed to the local particular subgroups. Continuous HIV incidence surveillance should be conducted for the estimation of future local HIV epidemic situation.