Biomarkers of Inflammation and Immune Activation and HIV-1 Disease Progression in the Multicenter AIDS Cohort Study
McKay, Heather S.
MetadataShow full item record
Background: Progression of HIV-1 disease is characterized by chronic inflammation and immune activation, the precise mechanisms of which remain unclear. While the introduction of highly active anti-retroviral therapy has vastly improved the survival of HIV-infected individuals, individuals infected with HIV experience excess risk of non-AIDS-related morbidities and mortality relative to uninfected individuals. The putative role of inflammation and immune activation in early HIV infection is incompletely understood but may offer insights into HIV pathogenesis and suggest novel ideas for vaccine research. Objectives: This dissertation was nested in the Multicenter AIDS Cohort Study (MACS). The overall objectives of this dissertation were as follows: Aim 1) to ascertain the long-term within-person reliability of 22 serological biomarkers of inflammation and immune activation as measured by the Meso Scale Discovery and Luminex multiplex assay platforms; Aim 2) to evaluate the association of non-HIV-specific sociodemographics, risk behaviors, and morbidities with serological measurements of these 22 biomarkers and C-reactive protein and investigate underlying interrelationships among these biomarkers; Aim 3) to determine the effect of untreated HIV infection on inflammation and immune activation and evaluate the longitudinal effect of these biomarker concentrations early after infection on subsequent HIV RNA setpoint; and, finally, Aim 4) to evaluate the impact of inflammation and immune activation early in the course of HIV infection on long-term non-progression of clinical HIV disease among men who have sex with men. Methods: In Aim 1, the Meso Scale Discovery and Luminex assay platforms were assessed using external control samples run across plates. Linear mixed models were used to obtain age-adjusted intraclass correlation coefficients for biomarkers with > 80% detectability using 4 visits from 250 HIV-uninfected men. Generalized gamma models were used to estimate associations between fixed and modifiable host factors and individual biomarkers in Aim 2. Exploratory factor analysis was conducted to investigate the interrelationships among these biomarkers, and linear mixed models were used to examine the associations of host characteristics with the resulting factors. To examine the effect of HIV infection, as stated in Aim 3, serum concentrations of the biomarkers were measured in longitudinal specimens obtained from 273 men observed to seroconvert to HIV prior to 1996. Linear mixed models with piecewise linear spline terms were employed to examine the change in biomarker concentrations from pre- to post-seroconversion time points. Multiple linear regression was then used to determine the association between biomarker concentrations obtained 0 – 9 months after seroconversion and prior to infection with subsequent HIV RNA setpoint levels. Finally, in Aim 4, we compared biomarker concentrations early after HIV infection between 54 long-term non-progressors and 189 HIV-infected progressors and 193 HIV-uninfected men using relative percentiles obtained from generalized gamma models. Results: Most markers were detectable in ≥ 80% of control samples with median intra-assay CVs < 15%. Among the HIV-uninfected men, most biomarkers showed fair to strong within-person correlation (ICC > 0.40) up to 15 years between measurements. Significantly (p < 0.002) higher biomarker concentrations were documented as follows: Eotaxin (non-black race), IL-8 (persistent and uncontrolled hypertension), IP-10 (age, obese BMI, cleared and chronic hepatitis C), MCP-1 (non-black race), MCP-4 (age, black race), MIP-1β (black race), TARC (black race), CXCL13 (normal/underweight BMI, low total cholesterol), IL-2 (A.M. blood draw), IL-10 (HCV, use of stimulants), IL-6 (age, BMI, HCV, persistent and uncontrolled diabetes, uncontrolled hypertension), TNF-α (depression), BAFF (HCV), CRP (BMI), sIL-2Rα (HCV), sCD27 (age, HCV, low total cholesterol), sgp130 (HCV), and sTNF-R2 (age, HCV). Two biomarker factors were identified: immune activation (sTNF-R2, sCD27, sIL-2Rα, BAFF, sgp130, IP-10) and inflammation (IL-6, GM-CSF, IL-1β, IL-8, TNF-α, IL-2, IL-10, IL-12p70). Immune activation scores were significantly elevated with age, obesity, and hepatitis C, and significantly depressed with binge drinking and hypercholesterolemia. Inflammation scores were significantly elevated with age, obesity, uncontrolled hypertension, and depressive symptoms. Within the first year after seroconversion, significant (p < 0.002) increases in concentrations of IP-10, CXCL13, TNF-α, sCD27, sTNF-R2, sIL-2Rα, sCD14, BAFF, and CRP and significant decreases in GM-CSF, MIP-1β, IL-8, and Eotaxin were observed. Heterogeneous patterns of increasing, decreasing, and unchanging marker levels were observed from 1 to 3 years after seroconversion. Concentrations of IP-10, CXCL13, sCD27, sIL-2Rα, and sTNF-R2 0 – 9 months after seroconversion were positively associated with subsequent HIV RNA setpoint levels, as was BAFF in sensitivity analyses (0 – 6 months after seroconversion). Pre-infection concentrations of sCD27 and CXCL13 were associated with subsequent setpoint levels. Relative to progressors, long-term non-progressors had higher concentrations of Eotaxin (+22%; p = 0.02), MIP-1β (+44%; p = 0.03), and IFN-γ (+53%; p = 0.05) 12 – 24 months after HIV infection. Conclusions: Our results suggest that the Meso-Scale Discovery and Luminex assays showed acceptable performance for quantitating the concentrations of most of the studied inflammatory and immune biomarkers. Additionally, most biomarkers displayed moderate-to-excellent within-person reliability over the long term (up to 15 years). Host characteristics associated with inflammation and immune activation provide potential targets for disease prevention and treatment. Summary factors may be useful for evaluating the role of inflammatory burden in future etiologic studies. HIV infection yields a complex mixture of early inflammatory and immune activation responses. The identification of biomarkers associated with subsequent HIV RNA setpoint levels has implications for current vaccine and therapeutic strategies. Long-term non-progression may be associated with elevated concentrations of MIP-1β and Eotaxin early after HIV infection. Long-term non-progressors were distinct from HIV-uninfected men, exhibiting higher levels of T cell activation and dysregulated inflammatory responses.