Multidimensional and longitudinal immune profiling of sepsis in Uganda

PROJECT SUMMARY The global burden of sepsis is overwhelmingly concentrated in sub-Saharan Africa (SSA), where nearly 40% of all sepsis cases occur and up to 65% of all deaths are sepsis-related. While sepsis outcomes have improved over time in high-income countries (HICs), outcomes remain poor in SSA, with short-term mortality between 20- 50%. Further, sepsis treatment protocols developed in HICs have repeatedly shown harm when implemented in SSA. We argue that an imprecise understanding of sepsis pathobiology in SSA is a fundamental barrier to the development of effective treatment strategies. In contrast with HICs, sepsis in SSA occurs primarily in younger adults, inciting pathogens are diverse, and HIV/AIDS is a major driver of poor outcomes. Despite this distinctive landscape, locally relevant models of sepsis pathobiology are scarce. Our overarching hypothesis is that unique demographic and pathogen profiles inherent to sepsis in SSA, including a high burden of comorbidities such as HIV/AIDS, underlies heterogenous sepsis pathobiology and treatment responses. In this project, we will apply and improve upon precision medicine strategies used in HICs to establish pathobiologically-driven and therapeutically relevant sepsis subtypes (“endotypes”) in SSA. Using RNAseq and machine learning in a pilot cohort, our Uganda-based team has demonstrated, for the first time, the presence of two sepsis endotypes in SSA that are reflective of host-pathogen features unique to the region and differentiated by targetable pathobiological mechanisms and risks of multiorgan dysfunction and death. In a multicenter prospective cohort study in Uganda, we will validate these findings and infer relevant biological pathways to pinpoint mechanisms driving sepsis outcomes in Aim 1. Leveraging this cohort in Aim 2, we will apply high-dimensional proteomics to define longitudinal immune response trajectories underlying sepsis outcomes beyond hospital mortality, including functional disability in sepsis survivors. Our preliminary data from Uganda also suggests there are unique aspects of sepsis pathobiology in persons living with HIV (PLWH), including exaggerated production of pro-inflammatory cytokines and alarmins, expansion of T-reg and aberrant NK cells, and upregulation of immune checkpoint T-cell exhaustion pathways. In Aim 3, we will perform in-depth immunophenotyping to develop a working mechanistic understanding of sepsis in PLWH. In all three Aims, we will leverage established sepsis research infrastructure in Uganda and multidisciplinary collaborations between critical care and infectious disease physician-scientists, immunologists, and bioinformaticians/biostatisticians at Columbia University and Uganda Virus Research Institute. Results from this project will (i) illuminate crucial domains of sepsis pathobiology in the global hotspot of SSA and (ii) translate these insights to classify individual sepsis patients into pathobiologically-driven and therapeutically relevant subgroups. Ultimately, our data will inform design of the first clinical trial of precision, host-directed treatment for sepsis in SSA, results of which will enhance delivery of effective sepsis therapeutics worldwide.