Plasma metabolomics of primary open-angle glaucoma in three prospective US cohorts and the UK Biobank.

PurposeTo better understand the etiologic pathways in glaucoma, we aimed to identify pre-diagnostic plasma metabolites associated with glaucoma risk. MethodsIn a case-control study from the Nurses Health Study (NHS), NHSII and Health Professionals Follow-Up Study (HPFS), 599 incident primary open-angle glaucoma (POAG) cases (mean time between blood draw and diagnosis was 10.3 years) were 1:1 matched to 599 controls. Plasma metabolites were measured with LC-MS/MS at the Broad Institute (Cambridge, MA, USA); 367 metabolites from 17 metabolite classes passed quality control analyses. For comparison, in a cross-sectional study in the UK Biobank, 168 NMR metabolites (Nightingale, Finland; version 2020) were measured in serum samples from 2,238 prevalent glaucoma cases and 44,723 controls. Metabolites were probit-score transformed for normality; multiple logistic regression was used to identify metabolites associated with POAG in NHS/NHSII/HPFS and glaucoma in UK Biobank. In NHS/NHSII/HPFS, we also used Metabolite Set Enrichment Analysis to identify metabolite classes associated with POAG. All analyses adjusted for established glaucoma risk factors. False discovery rate (FDR) and number of effective tests (NEF) were used to adjust for multiple comparisons. ResultsNine metabolite classes were associated (FDR<0.05) with POAG in NHS/NHSII/HPFS: triglycerides, diglycerides, two lysophospholipids classes [lysophosphatidylcholines and lysophosphatidylethanolamines], and two phospholipid class [phosphatidylethanolamines and phosphatidylcholines] were positively associated, while cholesteryl esters, carnitines, and organic acids and derivatives were inversely associated with POAG risk; further adjustment for covariates minimally altered the results. These associations were particularly stronger for POAG with paracentral visual field loss. In the UK Biobank, notably, triglycerides and phospholipids (from which lysophospholipids are derived through hydrolysis), were confirmed to be associated (p<0.05) with higher glaucoma risk. Also, in the UK Biobank, the metabolites of tyrosine, glucose, and glutamine were positively associated (NEF<0.2) while 3-hydroxybutyrate, acetate, citrate, pyruvate, and lactate (the latter 4 being anionic organic acids) were inversely associated with glaucoma (NEF<0.05). ConclusionsHigher levels of glycerides (diglycerides and triglycerides) and phospholipids were adversely associated with glaucoma in both the NHS/NHSII/HPFS and the UK Biobank, suggesting that they play an important role in glaucoma pathogenesis. PRECISHigher glyceride and phospholipid levels in pre-diagnostic plasma was associated with glaucoma risk in three cohorts and were associated with prevalent glaucoma in the UK Biobank. Altered lipid metabolism may be etiologically important in glaucoma.