ESCRS - FP03.14 - Non-Invasive Corneal Cytology Sampling For Proteomics Biomarker Discovery In Keratoconus

Non-Invasive Corneal Cytology Sampling For Proteomics Biomarker Discovery In Keratoconus

Published 2025 - 43rd Congress of the ESCRS

Reference: FP03.14 | Type: Free paper | DOI: 10.82333/5vth-2d98

Authors: Janvi Karia* 1 , Saurabh Jain 1

1Ophthalmology ,Royal Free Hospital,London,United Kingdom

Purpose

To evaluate a novel, non-invasive method for collecting corneal epithelial cytology samples in ophthalmology clinics and its potential in identifying new proteomics biomarkers for keratoconus.

Setting

Keratoconus is a progressive corneal ectasia lacking reliable molecular biomarkers for early diagnosis and disease monitoring. Current diagnostic methods may not detect the earliest stages of the disease or predict progression accurately. The need for minimally invasive techniques to explore keratoconus at the molecular level has driven research into a novel cytology-based approach combined with untargeted proteomics analysis to improve early detection and personalized management.

Methods

This multicenter, prospective observational clinical study was conducted across 6 clinical centers in Spain. Corneal epithelial cytology samples were collected from 63 individuals, including patients with mild, moderate, and severe keratoconus, as well as controls. The samples underwent proteomics analysis using mass spectrometry to identify differentially abundant proteins. Statistical analysis included normalization and linear modeling to detect significant protein variations between groups. Proteins were considered relevant if their abundance met the criteria of adjusted p-value <0.05 and Log Fold-Change >|1.5|.

Results

Corneal cytology sampling was performed in outpatient settings (56%) and operating rooms (44%) with no reported adverse effects. This enabled proteomics analysis in 92% of cases. A total of 5,753 proteins were identified and quantified, with 712 showing significant differential abundance between keratoconus and controls. 20 proteins exhibited differential abundance exclusively in mild keratoconus, and 5 correlated with disease progression across severity stages. Functional analysis linked these proteins to cytoskeleton organization, extracellular matrix remodeling, protein transport, autophagy, immune response regulation, and cellular stress response. 20 candidate biomarkers were selected for verification in a new cohort of 60 individuals.

Conclusions

This non-invasive technique provides a reliable and well-tolerated method for investigating keratoconus and other corneal diseases at the molecular level. It has the potential to enhance biomarker discovery, leading to improved diagnostic and therapeutic strategies.