A variety of therapeutic strategies are being pursued to try to eradicate or constrain the latent HIV reservoir such that high-level viral rebound does not occur when antiretroviral therapy is discontinued. Testing the efficacy of these strategies currently requires analytical treatment interruption (ATI), during which a suppressed patient is taken off ART and carefully monitored for viral rebound. Having reliable biomarkers that could accurately predict the effectiveness of the therapeutic approach as assessed by time to rebound after treatment interruption would greatly accelerate the pace of HIV cure research, reduce the number of costly and logistically challenging ATI trials, and provide greater protection to patients participating in these trials. From a translatable perspective, these biomarkers should be readily detectable in blood and not require tissue biopsies. Although viral reservoir size is a predictor of time-to-rebound post-ATI, it is not a robust biomarker. Currently, there are no known immunological biomarkers at the time of ATI that can predict the duration of viral control. In this project, biomarkers predicting shorter or longer times to rebound will be identified in well-suppressed HIV-infected patients undergoing treatment interruption. Specifically, mass cytometry, or cytometry by time-of-flight (CyTOF), will be used to identify phenotypic and functional biomarkers predicting time to viral rebound. Blood samples obtained at the time of ATI, from patients who were treated during chronic as well as acute infection, will be analyzed by CyTOF deep-phenotyping for immunological signatures of CD4+ T cells, CD8+ T cells, B cells, monocytes, neutrophils, conventional DCs, plasmacytoid DCs, and NK cells that predict time-to-rebound. Seven panels of ~40 parameters each will be used. These signatures will include not only cellular phenotype but also the functional activity of these cells in response to ex vivo stimulations including treatment with latency reversal agents (LRAs). Correlations will be analyzed by the Citrus algorithm that identifies features of cellular subsets that significantly associate with disease outcome, in this case time-to-rebound. Finally, productively infected cells in these patients at the time of viral rebound will be characterized to help chart the latent reactivatable reservoir in these patients.
The aims i n this project will take advantage of the powers of high-dimensional CyTOF phenotyping and its analysis tools to identify novel biomarkers predicting time to viral rebound after treatment interruption. Such biomarkers might ultimately prove helpful in the evaluation of various cure therapies.