Background: The high-density lipoprotein (HDL) hypothesis of atherosclerosis has been challenged by clinical trials of cholesteryl ester transfer protein (CETP) inhibitors which failed to show significant reductions in cardiovascular events. Plasma levels of HDL-cholesterol (HDL-C) decline drastically during sepsis and this phenomenon is explained, in part, by the activity of CETP, a major determinant of plasma HDL-C levels. We tested the hypothesis that genetic or pharmacologic inhibition of CETP would preserve HDL levels and decrease mortality in clinical cohorts and animal models of sepsis. Methods: We examined the effect of a gain-of-function variant in CETP (rs1800777, p.Arg468Gln) and a genetic score for decreased CETP function on 28-day sepsis survival using Cox proportional hazard models adjusted for age and sex in the UK Biobank (n=5,949), Identification of SNPs Predisposing to Altered Acute Lung Injury Risk (iSPAAR; n=882), Copenhagen General Population Study (n=2,068), Copenhagen City Heart Study (n=493), Early Infection (n=200), St. Paul's Intensive Care Unit 2 (n=203), and Vasopressin versus Norepinephrine Infusion in Patients with Septic Shock studies (n=632). We then studied the effect of the CETP inhibitor anacetrapib in adult, female APOE*3-Leiden mice with or with human CETP expression using the cecal-ligation and puncture model of sepsis. Results: A fixed-effect meta-analysis of all 7 cohorts found that the CETP gain-of-function variant was significantly associated with increased risk of acute sepsis mortality (hazard ratio [95% confidence interval]: 1.44 [1.22-1.70], p<0.0001). In addition, a genetic score for decreased CETP function was associated with significantly decreased sepsis mortality in the UK Biobank (hazard ratio [95% confidence interval]: 0.77 [0.59-1.00] per 1 mmol/L increase in HDL-C) and iSPAAR cohorts (hazard ratio [95% confidence interval]: 0.60 [0.37-0.98] per 1 mmol/L increase HDL-C). APOE*3-Leiden.CETP mice treated with anacetrapib had preserved levels of HDL-C and apolipoprotein-AI and increased survival relative to placebo treatment (70.6% vs 35.3%, Log-rank p=0.03), while there was no effect of anacetrapib on the survival of APOE*3-Leiden mice which do not express CETP (50.0% vs 42.9%, Log-rank p=0.87). Conclusions: Clinical genetics and humanized mouse models suggest that inhibiting CETP may preserve HDL levels and improve outcomes for individuals with sepsis.