Personalized Model to Predict Keratoconus Progression From Demographic, Topographic, and Genetic Data

Area of Science:

• Ophthalmology
• Genetics
• Biostatistics

Background:

• Keratoconus is a progressive corneal disease affecting vision.
• Predicting keratoconus progression is crucial for timely intervention with corneal crosslinking (CXL).

Purpose of the Study:

• To develop and validate a prognostic model for predicting keratoconus progression.
• To identify key predictors of progression that may necessitate CXL.

Main Methods:

• Retrospective cohort study of 5025 early keratoconus patients (8701 eyes).
• Utilized Royston-Parmar method to model time-to-event for keratometric progression and CXL.
• Investigated age, keratometry, corneal thickness, and genetic data (SNPs) as covariates.

Main Results:

• Corneal crosslinking (CXL) served as a more robust endpoint than keratometric progression for the prognostic model.
• The final model explained 33% of variation; significant predictors included age, maximum anterior keratometry, and minimum corneal thickness.
• Identified SNPs associated with keratoconus did not significantly contribute to progression risk.

Conclusions:

• A validated prognostic model can improve patient empowerment, triage, and healthcare service provision for keratoconus.
• Age at presentation is the primary predictor of keratoconus progression risk.
• Genetic factors (SNPs) do not appear to influence keratoconus progression in this cohort.