HIV/AIDS is a global pandemic with 34 million individuals living with HIV infection worldwide. The objectives of this project are to define the unique epidemiological, clinical, virologic, and immunologic features of HIV infection in developing countries, to determine the viral kinetics associated with sexual transmission, and to characterize the molecular strains of HIV internationally for infectiousness and progression of disease. We previously reported the successful results of a randomized clinical trial of circumcision to prevent HIV acquisition. We have also shown that male circumcision also decreases high-risk human papillomavirus (HR-HPV) prevalence. The reduced HR-HPV prevalence by male circumcision is due to decreased incidence and increased clearance. In addition, male circumcision reduced HR-HPV prevalence and the incidence of multiple HR-HPV infections in HIV-positive men. Among female partners of circumcised men in a randomized controlled trial, we found that circumcision also reduced HR-HPV prevalence by 28%, reduced HR-HPV incidence and increased HR-HPV clearance in female partners. The association between HPV infection and HIV seroconversion risk is unclear and the genital cellular immunology has not previously been evaluated. Thus, we conducted a case-control analysis and found that HPV clearance was associated with subsequent HIV seroconversion. HPV infection was associated with increased epidermal dendritic cell density that potentially mediated HIV seroconversion. Accurate methods to estimate HIV incidence are needed to monitor the leading edge of the epidemic, identify groups at high risk of infection, and evaluate the effectiveness of prevention interventions. We previously determined that the BED capture immunoassay (BED-CEIA), the current assay that is used to estimate the number of new infections in the US was associated with multiple sources of misclassification including: viral suppression, AIDS, and race. To overcome these problems, we developed a multi-assay algorithm (MAA) for estimating cross-sectional HIV incidence in a clade B setting that includes the BED-CEIA, an antibody avidity assay, HIV viral load, and CD4 cell count. Using 1,782 samples from 709 individuals in the US infected 0.1 to 8 years and additionally 500 samples from individuals infected 8+ years, we evaluated more than 13,000 possible algorithms to identify a MAA that had a maximum window period (the median time an individual appears recently infected by the testing algorithm). We determined that using the BED-CEIA and the BioRad avidity assay with less stringent cut offs and in combination with CD4 and viral load, we had a MAA with a window period of 159 days (95%CI 134-186) where none of the 970 samples from individuals infected >5 years appeared recently infected. We compared annual incidence estimated with the MAA to incidence based on HIV seroconversion in three different longitudinal cohorts. In comparison with the HIVNET001 cohort study, annual incidence based on HIV seroconversion was 1.04% (95% CI: 0.70%-1.55%), while the incidence estimate obtained using the MAA was essentially identical: 0.97% (95% CI: 0.51%-1.71%). We did a similar analysis for HPTN064 and observed similar incidence estimates MAA-0.13% (95% CI: 0.01%-0.76%) and cohort-0.24% (95% CI: 0.07%-0.62%). We also compared incidence in HPTN 061 and found similar incidence cohort-3.0% (95%CI 2.0 4.4%) vs. MAA-3.4% (95%CI 1.8% - 6.2%). We also initiated studies on the utility of cross-sectional incidence assays in sub-Saharan Africa. Using 2193 samples from long term infected subjects from the Partners in Prevention study, we determined that the frequency of misclassification by both the BED-CEIA and BioRad avidity assay was higher in Eastern Africa (where subtypes A &D predominate) compared to Southern Africa (where subtype C is endemic). Further investigation on samples from the Rakai, Uganda, with known infecting subtype, demonstrated that the misclassification by the BED-CEIA and BioRad avidity assay was ten times higher in subtype D than in subtype A infected individuals. These finding demonstrate that cross-sectional incidence estimates can be determined effectively but may be subject to false rates in areas where subtype D predominates. We have expanded our use of ultra-deep pyrosequencing to identify HIV superinfection and describe its effects on the pandemic. As part of this research we identified the first case of a HIV superinfection associated transmission event from a positive individual to his negative partner. We used data generated examining HIV superinfection in Rakai Uganda to validate a high-resolution melting assay for the measurement of HIV diversity. In addition, we established multiple collaborations to expand our work on HIV superinfection to explore its role on HIV transmission in the HPTN 052 and PEPI clinical trials, occurrence in seroconverters from the CAPRISA 004 clinical trial, and in a group of female sex workers in Kampala Uganda. Using a combination of techniques including deep-sequencing we demonstrated that previously transmitted viral strains were preferentially selected during transmission to a new uninfected individual. We are continuing to use these techniques to assist in the identification of transmission linkage as part of the ongoing follow-up to the HPTN 052 trial. We examined the pathophysiology of HIV in patients with impaired liver function. This study found that HIV infected individuals with measurable liver disease had higher levels of soluble CD14, which is a marker for monocyte activation. However, the opposite was true for HIV uninfected individuals with liver disease. We further examined the role of HIV subtype on pathogenesis. This study examined what viral genomic regions influenced the difference between subtypes A and D in terms of pathogenesis. Our work found that polymorphisms in the pol gene influenced rates of disease progression and was associated with differences in replication capacity.
Showing the most recent 10 out of 299 publications
