Large-Scaled Proteomics Analysis for Glaucoma: Integrated Genetics, Multi-Omics, UK Biobank and Therapeutic Analysis
Investigative Ophthalmology & Visual Science, 2026
Xu J., Gao Y., Yu J., Xu Z., Zhang Y., Pang C., Tham C., Yam J., Chen L.
| Disease area | Application area | Sample type | Products |
|---|---|---|---|
Ophthalmology | Pathophysiology Patient Stratification | Plasma |  Olink Explore 3072/384 |
Abstract
Purpose: To identify plasma proteins associated with glaucoma and assess the translational potential of key proteins as both biomarkers and therapeutic targets.
Methods: Genome-wide association study data were obtained from the UK Biobank Pharma Proteomics Project, FinnGen, and the Million Veteran Program. We used a four-stage analytical framework: Stage 1 applied Mendelian randomization and Bayesian colocalization to evaluate associations between 2923 plasma proteins and glaucoma; Stage 2 used summary-based Mendelian randomization to explore transcriptomic and epigenomic associations of the identified proteins with glaucoma risk; Stage 3 involved a prospective association analysis of protein levels and incident glaucoma in the UK Biobank cohort, including 40,170 glaucoma-free participants; and Stage 4 systematically evaluated the druggability of the prioritized protein targets.
Results: We identified 26 plasma proteins with putative causal associations with glaucoma, six of which were novel: COL24A1, KAZALD1, EBAG9, CSNK1D, AZI2, and AXIN1. COL24A1 (odds ratio [OR] = 0.85; 95% confidence interval [CI], 0.80–0.90; PFDR < 0.001; PP.H4 = 0.95) and EFEMP1 (OR = 0.88; 95% CI, 0.83–0.92; PFDR < 0.001; PP.H4 = 0.98) emerged as the most compelling candidates. To further elucidate the regulatory mechanisms, multiomics analyses indicated that epigenetic modifications and alternative splicing events affecting these genes were associated with elevated glaucoma risk. Notably, EFEMP1 was significantly associated with glaucoma incidence in the prospective cohort analysis (fully adjusted Cox model: hazard ratio = 1.61; 95% CI, 1.29–2.00; PFDR = 0.002), demonstrating strong predictive performance (C-index = 0.811, area under the curve = 0.806) and representing a promising therapeutic target.Conclusions: Our findings provide new insights into the proteomic basis of glaucoma and highlight promising opportunities for developing targeted therapies.