D17-03 Genetically Regulated Angiopoietin-like-4 Is Causally Associated With CAP-Related Outcomes  

Abstract Rationale Community acquired pneumonia (CAP) remains a leading cause of death world-wide and therapies that target a dysregulated host response are limited. Angiopoietin-like-4 (ANGPTL4) is a secreted glycoprotein involved in capillary leak, angiogenesis, lipid metabolism, and in pre-clinical models inhibition of ANGPTL4 leads to improved survival. This study measured plasma ANGPTL4 in two prospective patient cohorts of CAP and employed Mendelian randomization (MR) to elucidate the causal relationship between ANGPTL4 and CAP-related outcomes. Methods We analyzed ICU patients with COVID-19 from two prospective cohorts (Discovery: CHROme, N = 198 and the Validation: SARI-PREP, N = 100). Plasma was collected within 48 hours of ICU admission and ANGPTL4 levels were quantified using an aptamer-based platform from Somalogic. Multivariable ordinal regression analysis was performed to assess the association of ANGPTL4 with admission NIH ordinal scale, and logistic regression for hospital mortality, adjusting for age, sex, BMI, diabetes, heart failure, and COPD/asthma. Next, we performed genome-wide genotyping using the Illumina Infinium Global Diversity Array in patients with available DNA in these cohorts (n = 258) and analyzed single-nucleotide polymorphisms (SNPs) within 50kbp proximity to the coding sequence of the ANGPTL4 gene to conduct a 1-sample MR to assess for a causal relationship between ANGPTL4 and mortality. Results Doubling in ANGPTL4 levels was associated with higher odds of more severe NIH ordinal scale in the discovery (adjusted odds ratio (aOR), 3.3; 95% confidence interval 95% CI, 2.3 to 4.9; p=5.2x10-10) and validation cohorts (aOR, 2.9; 95% CI, 1.6 to 5.3; p=3.2x10-4) (Figure 1A and 1B). ANGPTL4 levels were also associated with greater hospital mortality in both cohorts (discovery: aOR, 2.3; CI, 1.5-3.5; p=1.6x10-4; validation: aOR, 3.2; CI, 1.7-6.0; p=3.6x10-4). In combined data from CHROme and SARI-PREP, we found SNP rs74829860 demonstrated significant association with hospital mortality (CT, minor allele OR 3.4, 95% CI: 1.5-7.9; FDR=0.03, p=0.004), correcting for local linkage disequilibirum-based FDR (Figure 1C and 1D). This SNP was associated with ANGPTL4 plasma levels (1.35-fold change per T allele, 95% CI 1.04-1.75, p=0.025). In a 1-sample MR analysis, positive causality was observed for rs74829860 and hospital mortality (Wald test p=0.003). SNP rs74829860 is a noncoding intergenic variant that mapped 32kbp in the 3’ region downstream of ANGPTL4 in an epigenetic regulatory ZBTB6 transcription factor binding site. Conclusions Plasma ANGPTL4 levels are genetically regulated and associated with worse clinical outcomes. MR suggested a causal association between ANGPTL4 levels and hospital mortality in CAP. These results support development of ANGPTL4-targeted therapeutics for CAP. This abstract is funded by: T32 HL007287; T32 GM121290; R01HL172872