Identifying Predictors of Anti-VEGF Treatment Response in Patients with Neovascular Age-Related Macular Degeneration through Discriminant and Principal Component Analysis.

OBJECTIVE

AURA was an observational study that monitored visual acuity outcomes following ranibizumab use in neovascular age-related macular degeneration patients over 2 years. The aim of this analysis was to identify factors that were predictive of visual acuity outcomes in AURA.

METHODS

The correlation between the baseline characteristics, the use of resources and the visual acuity outcomes in AURA was explored using principal component analysis (PCA) and partial least-squares-discriminant analysis (PLS-DA). The response variables analysed were mean change in visual acuity over 2 years (analysed via PCA) and no decline in visual acuity at 2 years compared with baseline (analysed via PLS-DA).

RESULTS

The AURA dataset comprised 2,227 patients and 132 variables. Using PCA and PLS-DA, we found that the number of ranibizumab injections, clinic and monitoring visits, number of optical coherence tomography scans and ophthalmoscopies correlated with a change in visual acuity at Years 1 and 2, and are therefore key drivers of treatment success.

CONCLUSION

This is a novel approach to graphically explore relationships between multiple correlated covariates and outcomes in real-life ophthalmology studies. It identified a number of variables that are positively linked with treatment outcomes.