UNDERSTANDING THE IMMUNOPATHOGENESIS OF PERINATAL HIV-1 INFECTION: TOWARDS LONG-TERM REMISSION

Background: If left untreated, HIV-1 infection in infants is detrimental to the immune system with lifelong perturbation in the generation of effective cellular and humoral immune responses leading to increased morbidity and mortality. The underdeveloped immune system of infants leads to high levels of viremia (>100,000 copies/mL plasma), with prolonged time to reach the viral load set point and —compared with adult infection— a faster pace of disease progression. Therefore, the health benefits of combination antiretroviral treatment (cART) that effectively suppresses HIV-1 replication are enormous to a perinatally infected infant and more pronounced when started early (within three months of life) in the course of infection. However, despite the clear benefits of an early reduction of viral load, HIV-1 infection is accompanied by a cascade of inflammatory processes that are not normalized with cART. The inflammatory cascade is hypothesized to be driven in part by HIV-1-mediated mucosal injury and microbial translocation. In HIV-1-infected adults, the heightened inflammatory milieu has been associated with increased morbidity, reduced vaccine responses, and increased mortality. The extent of inflammation and its ill effects, including the responses to immunizations, are undefined in perinatal HIV-1 infection and are the focus of the first objective of my thesis work. An important consideration for treatment of HIV-1 infection is that while cART can control HIV-1 replication for decades, it is not curative due to an early established HIV-1 reservoir in latently infected memory CD4+ T cells (the latent reservoir), necessitating lifelong treatment. There are now intense research efforts to shrink these viral reservoirs in order to achieve a state of HIV-1 remission where infected individuals can potentially stop their cART without viremic rebound. Heightened inflammation and immune activation in infected adults also appear to contribute to maintenance of the latent reservoir, highlighting the importance of an inflammatory milieu in HIV-1 remission and curative efforts. It appears, however, that treatment of HIV-1 during acute infection in adults restricts the size and composition of the latent reservoir towards lower infection rates of long-lived central memory cells. Achieving low reservoir size (considered < 2.3-log10 or 199.5 HIV-1 DNA copies/million peripheral blood cells) through early treatment has emerged as an important prerequisite to enable HIV-1 remission, but is likely not the only contributor to this outcome. With perinatal HIV-1 infection, unlike infection in adults, early treatment is more feasible due to knowledge of timing of exposure from maternal infection. This enables early testing and treatment. Understanding the dynamics of the generation and clearance of HIV-1-infected cells with early, long-term cART in perinatal infection is critical for guiding strategies to achieve virologic remission, but has not been intensively investigated. This is the focus of the second aspect of my thesis work. Objectives: 1) To test the hypothesis that the level of inflammation and immune activation induced in infants by perinatal HIV-1 infection reduces the efficacy of the humoral responses to oral, live pentavalent rotavirus vaccine (RV5). This study was performed in the context of a double-blind, placebo-controlled clinical trial (IMPAACT P1072) sponsored by the NIH Clinical Trials Network of pentavalent rotavirus vaccine (RV5) in HIV-1 infected and exposed-uninfected African infants. This study was conducted before the widespread roll-out of rotavirus vaccines in Sub-Saharan Africa. 2) To test the hypothesis that children who receive an early initiation of cART will increase the clearance of infected cells and reduce the proviral reservoir size compared to children who have a delay in beginning cART. A study of HIV-1-infected cell frequencies before and following cART in the first two years of life was conducted in perinatally infected infants enrolled in a clinical trial of early therapy (IMPAACT P1030). This was a Phase I/II clinical trial of pharmacokinetic properties of the HIV-1 protease inhibitor lopinavir/ritonavir in infants. A second study on HIV-1 decay was performed in older children and youth enrolled in a prospective cohort study in the U.S. (PHACS Cohort) of the long-term outcomes of perinatal HIV-1 Infection. Methods: For the first objective, we determined inflammation and immune activation in infants enrolled in the IMPAACT P1072 trial by measuring plasma cytokines (IFNγ, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, TNFα) and a monocyte/macrophage activation marker (soluble CD14) before and following receipt of RV5. Humoral responses were assessed by measuring both serum neutralizing antibodies (SNA) to the RV5 viral capsid proteins (G1, G2, G3, G4 and P1[A]), in addition to serum and copro anti-rotavirus IgA. Cytokines and sCD14 results were analyzed by HIV-1 infection status, receipt of vaccine, age, exposure to breastmilk and humoral responses to RV5. For the second objective, we optimized a highly sensitive and precise method of quantifying HIV-1 DNA and cell-associated HIV-1 RNA in circulating PBMC using droplet digital PCR (ddPCR). We determined the concentrations of HIV-1-infected cells occurring during perinatal infection, their transcriptional activity (measured by cellular HIV-1 RNA species [multiply-spliced (ms) and unspliced (us)] and the effects of early cART on the short- and long-term decay dynamics of HIV-1-infected cells with perinatal infection. A group of control subjects who did not achieve effective virologic suppression until after infancy was also studied to examine the effects of delayed virologic control on HIV-1 reservoir formation and clearance. Results: For objective 1, we found that by 3 months of age and prior to vaccination, acquiring HIV-1-infection led to higher levels of cytokines, mostly inflammatory (IFNγ, IL1-β, IL-2, IL-6, IL-8, IL-10 and TNFα), in addition to the monocyte/macrophage activation marker, sCD14, when compared with infants who were HIV-1 exposed but uninfected (HEU). After adjusting for age and exposure to breast milk, concentrations of all cytokines but IL-8 remained significantly higher in the HIV-1-infected infants compared with the HEU infants. Among HIV-1-infected infants, lower WHO weight-for-age Z-scores and higher levels of plasma viral load were significantly correlated with elevated concentrations of ≥1 biomarkers. Importantly, overall antibody responses, including serum IgA to RV5, were comparable in the HIV-1-infected infants compared with the HEU infants. Having higher concentration of IL-10 at pre-vaccination was correlated with lower concentration of SNA to virus capsid antigens G1 and P1[A] post-vaccination. Similarly, higher concentration of IFNγ was also associated with lower concentration of SNA to G1. The lack of effect of inflammation and immune activation on short-term responses to RV5, particularly the IgA responses, is promising, although whether there are long-term effects on the durability of immune response is unknown and requires further study. For the second objective, we found that CD4+ T cells in infancy—despite having sub-optimal immune response to pathogen-associated molecular patterns (PAMP)—are quite permissive to HIV-1 infection. By the time the infant is 2 months old, HIV-1-infected cells reach a median concentration of 3.2 log10 copies per million PBMCs, analogous to concentrations reported in adults. We also found that before cART, HIV-1-infected cells have high transcriptional activity with high concentrations of both ms- and us-HIV-1 RNA species. With cART, a substantial proportion of transcriptionally active cells were cleared (99%) within the first 24 weeks of cART, supporting clearance of the shorter-lived, productively-infected cells. The estimated decay rates did not change after adjusting for concurrent CD4+ T-cell percentages, suggesting the decrease observed in HIV-1-infected cell concentrations was due to clearance of infected cells rather than to a dilutional effect from immune reconstitution. Importantly, the concentration of HIV-1-infected cells before cART was associated with time to undetectable plasma viral load, suggesting ongoing viremia from long-lived cells generated pre-cART. Early reservoir establishment is also supported by our finding that the concentrations of HIV-1-infected cells persisting under cART through two years of age was highly correlated with the starting concentration. With 96 weeks of cART, HIV-1 DNA concentrations exceeded replication-competent genomes by 148-fold, emphasizing early and large contributions of defective genomes in the pathogenesis of HIV-1 infection; however, low-levels of unspliced HIV-1 transcripts remained detectable in 75% of the infants. The mechanisms governing the nature and dynamics of persistent HIV transcription during effective cART requires further investigation. In the second study, we investigated the decay dynamics of HIV-1-infected cells and proviral reservoir size following a decade of suppressive cART in children who receive an early initiation of cART and achieve prompt virologic suppression compared to children who have a delay in cART initiation and virologic suppression. The findings from this study illustrated that HIV-1-infected cells are cleared differentially in the first two years following virologic control based on the age at which a child achieves successful control of HIV-1 replication. HIV-1-infected cells are cleared at a significantly faster rate in the first two years of virologic suppression among participants who achieve virologic control by 1 year of age compared to between 1 and 5 years of age. This observation suggests a greater clearance of productively-infected cells with early treatment. After 2 years of virologic suppression, there was no significant difference in the clearance of infected cells between the two groups. With a decade of effective virologic suppression, HIV-1-infected cell concentrations reached an estimated mean of 0.99 log10 copies per million PBMC in the children who controlled virus replication by 1 year of age. We also examined whether the substantial ongoing decay of HIV-1 infected cells with durable cART might be due to dilution of infected cells as a consequence of immune reconstitution and possibly lymphoid expansion due to somatic growth. However, this is less likely as the HIV-1 DNA decay rates did not change after adjusting for CD4+ T cells count per µL—although mathematical modeling will be required to fully address this question. Conclusions: Perinatal HIV-1 infection is associated with early onset of dysregulation of inflammation and immune activation pathways that are associated with levels of viremia and low weight. High plasma IFNγ and/or IL-10 may impair the generation of neutralizing IgG, but not IgA, antibody responses to RV5 vaccines antigens after cART is instituted. Early initiation of cART, i.e., by two months of age, did not prevent establishment of a large pool of transcriptionally active HIV-1-infected cells. The level of HIV-1-infected cells generated during primary infection influences the time to undetectable viral load and reservoir size at 2 years of cART. Importantly, with over a decade of suppressive cART, HIV-1-infected cell concentrations continually decreased, irrespective of age at cART initiation. HIV-1 DNA slope estimates did not change with adjustments for CD4+ T cell counts and percentages, suggesting that immune reconstitution with naïve CD4+ T cells or lymphoid expansion from somatic growth are likely not the primary mechanisms for the observed decrease of HIV-1 DNA concentrations over time in perinatal infection. Moreover, infants who initiate cART early and achieve faster virologic control have a rapid initial decay of HIV-1 DNA that results in an extremely small HIV-1 proviral reservoir that is substantially smaller than that observed in adults initiating cART during acute infection. Data from the above two studies (P1030 and PHACS) on the concentration of HIV-1-infected cells will inform clinical trials aimed at achieving HIV-1 remission and cure for this population