Risk factors analysis of longitudinal viral dynamics and immunological responses among HIV-discordant couples

Abstract

In their 2011 paper “Prevention of HIV-1 infection with Early Antiretroviral Therapy” (Cohen, et al, New England Journal of Medicine, 2011), an interim analysis was performed for the HPTN 052 Study to address both personal and public health benefits from early antiretroviral therapy (ART) among HIV-infected patients and their HIV-uninfected partners. It was found that at least 96% of genetically linked infections caused by human immunodeficiency virus type I (HIV-1) in serodiscordant couples were prevented by ART [1]. Furthermore, as described in their 2016 paper “Antiretroviral Therapy for the Prevention of HIV-1 Transmission” ((Cohen, et al, New England Journal of Medicine, 2016), offering the antiretroviral therapy to all patients with HIV-1 infection including 5 years of follow-up proves the durability of ART for the prevention of HIV-1 transmission comparing early and delayed ART. As a result, they conclude that the early initiation of ART results in meaningful decrease in genetically linked HIV-1 infections in sexual partners [2]. In this thesis, we intend to perform risk factors analyses to identify key prognostic factors that are associated with the longitudinal progression of viral dynamics and immunological response, reflected by viral loads and CD4+ counts, respectively, via repeated measurements analysis with Linear Mixed-Effects Models and Generalized Estimating Equations (GEE), for the HIV-infected patients receiving early and delayed ART for the HPTN 052 Study. Such analyses provide more insights into how HIV-1 viral dynamics and patient’s immunological response are progressing over time, when the patients are experiencing or yet to experience HIV disease symptoms with ART [2]. After performing all the analyses, it was found that it is better to use Antiretroviral therapy as early as possible based on multivariate analyses for both immediate arm and delayed arm. In addition, we also found that baseline CD4 count, baseline viral load and region are the key risk factors when predicting how the CD4 count and viral load change over time in either immediate arm or delayed arm.