Wechsler, M E

Abstract Rationale Asthma and obesity are common comorbidities, both driven by inflammation and oxidative stress. Oxylipins are derivatives of polyunsaturated fatty acids that regulate the onset and resolution of inflammation. Oxylipin profiles are altered in obesity and have been shown to differentiate asthma from non-asthma; however, the association between oxylipins and asthma outcomes and the impact of obesity is not known. This study aims to identify differences in serum oxylipins associated with asthma morbidity and evaluate the effect modification by BMI. Methods We analyzed 150 participants in the Best African American Response to Asthma Drugs (BARD) clinical trial. Ninety oxylipins were quantified using liquid chromatography tandem mass spectroscopy from serum obtained at the time of randomization. Oxylipins found to be above the limit of detection in > 80% of samples were log-transformed and included in the analysis. Asthma severity measures included past 12-month systemic corticosteroid use for asthma exacerbation, spirometry, asthma control and asthma quality of life questionnaires (AQLQ). The associations between oxylipins and asthma outcomes were tested using multivariable regression models adjusted for sex, age, BMI, absolute eosinophil count, percent African ancestry, and child vs adult/adolescent cohort. A false discovery rate (FDR) of < 0.1 is used to determine statistical significance. Results Of the 90 oxylipins measured, 42 (46.7%) were detected in > 80% of all samples and included in the analysis. In multivariable models, the pro-inflammatory arachidonic acid (AA) derivative 5,6-DiHETrE was associated with decreased odds of past 12-month systemic corticosteroid use (FDR-adjusted p-value 0.0628). AA derivatives TXB2, PGF2a, and 6-α-prostaglandin were associated with increased FEV1/FVC (FDR-adjusted p-values=0.0925). Thirteen oxylipins were associated with decreased asthma-related quality of life in multivariable models, including AA metabolites 5S-HETE, 5-KETE, 14,15-EET, LTB4, and Lipoxin A4; dihomo-gamma-linoleic acid metabolites 5-HETrE and 15-HETrE; linoleic acid metabolites 3-OxoODE and 9-OxoODE; and eicosapentaenoic acid metabolites 15-HEPE and 17,18-EpETE, all with FDR-adjusted p-values<0.05. Oxylipin-BMI interaction was significant only in the AQLQ models. In all but LTE4, increased concentrations of oxylipins were associated with a decreased quality of life in individuals with higher BMI (Figure). Six of these oxylipins are products of lipoxygenase activity. Conclusions In this study, we find decreased quality of life with increasing concentrations of pro-inflammatory and resolving oxylipins among those with obesity. These oxylipins are primarily produced via lipoxygenases and may represent an important inflammatory pathway in obesity-associated asthma. Quantification of sputum oxylipins and correlation with serum levels will be important in understanding potential mechanisms and is forthcoming. This abstract is funded by: K12TR004412, R01HL123385, U10HL098075, Parker B. Francis Fellowship, 2L30HL159725

Abstract Rationale Higher African genetic ancestry has been associated with lower lung function, among individuals from specific racial-ethnic groups. The influence and interactions of genetic variation and ancestry in lung function is poorly understood. This study investigates ancestry-associated loci influencing lung function via admixture mapping and fine-scale allelic association. Methods 512 African American patients with asthma (≥12 years) from the Severe Asthma Research Program and Collaborative Study on the Genetics of Asthma (SARP/CSGA) cohorts were genotyped on the OmniExpress and HumanHap1M arrays or had whole-genome sequencing (WGS). For admixture mapping, global ancestry was inferred using ADMIXTURE and local ancestry estimated using RFMix v1.5.4 with CEU (N = 99) and YRI (N = 108) references across 15,095 local ancestry segments. Peak regions and contiguous segments (p < 10-3) were fine mapped in SARP using local ancestry estimated from WGS with MAF ≥0.03. The Local Ancestry Adjusted Allelic Association (LAAA) model was used to fine map local genetic effect, local ancestry effect, and the interaction between the two, adjusted for age, sex, height, BMI, cohort, and percentage global African ancestry. SNPs with local genetic or interaction p < 10⁻³ were tested for replication in 282 adolescent/adult participants from the Best African Response to Asthma Drugs (BARD) cohort genotyped with the MEGA array imputed to TOPMed (MAF ≥0.03). Results Mean African genetic ancestry was 81.3% in SARP and 80.1% in BARD. In SARP, African genetic ancestry was inversely associated with lung function (FEV₁:β = −7.9, p = 0.005; FVC:β = −9.0, p = 0.004), with stronger effects in females (FEV₁: β = −9.3, p = 0.003; FVC: β = −11.4, p = 0.001) than males (FEV₁:β = −6.4, p = 0.237; FVC:β = −5.7, p = 0.340). Similar total and sex-specific associations were identified in BARD. Admixture mapping in SARP/CSGA identified significant peaks for maximum FVC (p < 1.87 × 10⁻4) on chromosomes 8 and 17 near RP1 and RBFOX3, and near-significant peaks on chromosome 5 near IQGAP2, SNRPCP2, SV2C and for the maximum FEV₁/FVC ratio on chromosome 10 near PCDH15. Fine mapping identified the top SNP on chromosome 8 (rs73587686; p-allele-dose=9.0x10-5; RGS20; q11.23), with 62 additional SNPs overlapping RGS20. SNP rs7825816 was also identified in SARP (β = 2.35, p-allele-dose =1.03 × 10⁻4) and replicated in BARD (β = 1.15, p-allele-dose =0.03). In BARD, LAAA also identified eight additional SNPs (p ≤ 0.05) in the same LD block (chr8:53914976-53918297) overlapping with RGS20, ATP6V1H, and TCEA1. Conclusions We identified multiple chromosomal regions associated with lung function in asthma patients of predominantly African genetic ancestry. Lung function associations were strongest at genetic variation near a key that was replicated in an independent asthma cohort. This abstract is funded by: R01 K08 HL118128, R01 HL142992, R01 NR013700, R01 HL69167, R01 HL161674, HL69116, HL69167, HL69170, HL69174, U10 HL109164, U10 HL109257, U10 HL109146, U10 HL109172, U10 HL109250, U10 HL109250, U10 HL109250, U10 HL109168, U10 HL109152, U10 HL109086-04, U10 HL098115, HL098102, HL098096, HL098075, HL098090, HL098177, HL098098, HL098107, HL098112, HL098103, and K12 HL89992