Viremia copy-years: measuring the effect of cumulative HIV burden on outcomes

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): For over a decade, the plasma HIV viral load biomarker has proven invaluable for the provision of clinical care and the conduct of HIV research. Although this biomarker is measured serially over time, it has largely been used in a cross-sectional manner analytically, focusing upon viral load measurement at a single time point (e.g., pre-treatment or most recent measure). While such approaches have demonstrated clear prognostic value of viral load in relation to key clinical outcomes, they fail to capture an individual's cumulative HIV burden over time, which is likely to be related to inflammation and immune activation, and to contribute to non-AIDS related morbidity and mortality. This application proposes development of a novel, cumulative measure of plasma HIV RNA exposure referred to as viremia copy-years. The investigators will estimate cumulative HIV burden following antiretroviral therapy (ART) initiation in treatment-naove patients, and evaluate the effect of viremia copy-years on clinical outcomes. It is hypothesized that viremia copy-years will yield demonstrable independent prognostic value beyond traditional cross-sectional approaches to measuring HIV viral load. A retrospective cohort study of treatment-naove HIV-infected patients initiating ART between 1 Jan 2000 and 31 Dec 2008 will be conducted through the 9-site, nationally distributed, CFAR Network of Integrated Clinical Systems. The specific aims are to: Develop a measure of cumulative plasma HIV viral load exposure, viremia copy-years, among treatment-naove patients initiating ART (aim 1); Evaluate the relationship between patient sociodemographic and clinical characteristics and viremia copy-years (aim 2); and Estimate the effect of viremia copy-years on all-cause mortality independent of pre-treatment and time-updated viral load (aim 3). Measurement of viremia copy-years will be approximated with a time-weighted sum using the trapezoidal rule (aim 1). Other methods will be explored including alternative use of a Riemann sum and kernel smoothers on scatter plots of viral load-by-time to define the area under the curve non-parametrically. Linear regression will be used with viremia copy-years as the dependent variable for aim 2, and as an independent variable using survival methods for aim 3. A semi-Bayes approach to covariate selection will be employed, and inverse probability weighted marginal structural models will be used to account for confounders on the causal pathway. Development of a cumulative measure of HIV viral load burden with demonstrable independent prognostic value has implications for future studies of viremia copy-years in relation to inflammatory biomarkers, immune activation and clinical events, as well as for risk stratification for clinical events among patients in HIV care. As patients in developed countries are increasingly living with HIV infection into the 7th and 8th decades of life and experiencing morbidity and mortality from non-AIDS defining clinical conditions, many of which are associated with inflammation and immune activation, the development of a novel measure capturing cumulative HIV viral load exposure is timely and desirable, and may transform the role of viral load in research and practice. PHS 398/2590 (Rev. 09/04, Reissued 4/2006) Page Continuation Format Page PUBLIC HEALTH RELEVANCE: HIV viral load measures the amount of virus in the blood of someone with HIV. This research will develop a new way to calculate the quantity of HIV virus someone with HIV has in their blood over time. This new HIV viral load measurement may allow doctors to better determine a patient's risk for dying.