Abstract
Purpose :
Exfoliation glaucoma (XFG) exhibits genetic and environmental heterogeneity in etiology; however, distinct etiologic endotypes of XFG have not been delineated. We aimed to use unsupervised machine learning on prediagnostic plasma metabolites to characterize distinct etiologic XFG endotypes.
Methods :
We collected blood samples from participants of the Nurses’ Health Study in 1989-1990, and the Health Professionals Follow-up Study in 1993-1995. Liquid chromatography-mass spectrometry was used to profile 379 plasma metabolites. During a mean follow-up period of 11.8 years until 2016, we identified 205 incident XFG cases, confirmed with medical record information. In this case-only analysis, after preprocessing of pre-diagnostic metabolites with adjustment for season, time of blood draw, and fasting status, we computed a distance matrix using Pearson distance. We then computed gap statistics to determine the optimal number of endotypes. XFG risk factors, clinical presentations, and metabolomic profiles were compared across endotypes. False discovery rate (FDR) was used to account for multiple comparisons in Metabolite Set Enrichment Analyses.
Results :
We identified three distinct endotypes. Compared with the reference group (endotype 2, the most common endotype (n=90; 43.9%), endotype 1 (n=56; 27.3%) tended to be male southern US residents with greater UV exposure and were least likely to have cardiovascular disease; among women, there was a higher percentage of postmenopausal status. Endotype 3 (n=59; 28.8%) was associated with being a male northern US resident and having a higher body mass index, a higher prevalence of cardiovascular disease, diabetes, hypertension, and dyslipidemia, and the lowest genetic susceptibility (e.g., lowest occurrence of the XFG risk allele rs7329408 in POMP). There were no differences in ophthalmic characteristics (e.g., maximum intraocular pressure, bilaterality, age at diagnosis) across endotypes (P≥0.6). In metabolite class enrichment analyses, compared with endotype 2, organic acids and carnitines were positively associated with endotype 1 while diglycerides and triglycerides were positively associated with endotype 3 (FDR<0.05).
Conclusions :
Integrated metabolomic profiling identified distinct XFG etiologic endotypes, suggesting different pathobiological mechanisms and susceptibilities to genetic and environmental XFG risk factors.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.