Lipidome analysis of pseudoexfoliation and other forms of glaucoma, and control aqueous humor

Abstract

Purpose : Comprehensive lipid profiling of the aqueous humor (AH) in patients with pseudoexfoliation glaucoma (PEXG) and primary open angle glaucoma (POAG) compared to controls with the aim of determining lipidome based disease group predictability.

Methods : The AH samples were collected from human donors following tenets of the declaration of Helsinki, under IRB exempted protocols. The following samples were collected 23 non-glaucomatous control, 19 POAG, 9 pseudoexfoliation syndrome but without glaucoma (PEX), and 14 PEXG AH. The samples were subjected to Bligh and Dyer lipid extraction. Untargeted lipidomic analysis was performed with 13 deuterated lipid internal standards for normalization among the lipid classes. Machine learning prediction was performed using three supervised logistic regression binary classification tasks stratified by patients: 1) POAG vs control, 2) PEXG vs control, and 3) PEX vs control vectors. Data are presented as mean peak intensity ± standard deviation.

Results : Lipidomic analysis resulted in the combined identification of 489 lipid species within 26 lipid classes across PEX, PEXG, POAG, and control AH. The mean total lipid content demonstrated lipid content to be highest among control AH (13.54±56.1) compared to the remaining PEX (4.21±10.90), PEXG (9.08±25.97), and POAG (5.66±15.75) samples. Notably, multiple cholesterol esters (ChE), phosphatidylcholines (PC), triglycerides (TG), and ceramides (Cer) were present in higher concentrations for the PEXG AH samples: ChE(16:0), ChE(20:3), ChE(18:1), ChE(18:3), ChE(22:6), ChE(18:2), ChE(20:4), PC(16:0/16:0), PC(16:0/18:2), TG(18:1/18:1/20:4), and Cer(t18:0/24:0). The PC (18:0/18:2), PC (36:2), and PC (34:1e) lipid clases are in low concentrations for PEX AH but highly concentrated in PEXG AH samples. Machine learning prediction yielded accuracy as follows: 1) POAG vs control, with 86% accuracy 2) PEXG vs control, with 71% accuracy and 3) PEX vs control, with 86% accuracy.

Conclusions : Despite the similarity in material deposition, several PC species were found in low concentration in PEX AH samples and found in high concentration in PEXG AH samples, suggesting the composition of the materials are fundamentally different in composition. These differences in lipid composition may help to distinguish them. Machine learning prediction has demonstrated the ability to differentiate all three groups and control, mostly with 86% accuracy.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.