Predicting Visual Acuity Responses to Anti-VEGF Treatment in the Comparison of Age-related Macular Degeneration Treatments Trials Using Machine Learning

To evaluate multiple machine learning (ML) models for predicting 2-year visual acuity (VA) responses to anti-vascular endothelial growth factor (anti-VEGF) treatment in the Comparison of Age-related Macular Degeneration (AMD) Treatment Trials (CATT) for neovascular AMD (nAMD) patients. Secondary analysis of public data from a randomized clinical trial 1029 CATT participants who completed 2 years follow-up with untreated active nAMD and baseline VA between 20/25 and 20/320 in the study eye. Five ML models [support-vector machine (SVM), random forest, extreme gradient boosting (XGBoost), multi-layer perceptron (MLP) neural network, and lasso] were applied to clinical and image data from baseline, weeks 4, 8, and 12 for predicting 4 VA outcomes (≥15-letter VA gain, ≥15-letter VA loss, VA change from baseline, actual VA) at 2 years. The CATT data from 1029 participants were randomly split for training (n=717), from which the models were trained using 10-fold cross-validation, and for final validation on a test dataset (n=312). Main Outcome Measures: Performances of ML models were assessed by R2 and mean absolute error (MAE) for predicting VA change from baseline and actual VA at 2 years, by the area under the receiver operating characteristic (ROC) curve (AUC) for predicting ≥15-letter VA gain and loss from baseline. Using training data up to week 12, the ML models from cross-validation achieved mean R2 of 0.24-0.29 (MAE=9.1-9.8 letters) for predicting VA change and 0.37-0.41 (MAE=9.3-10.2 letters) for predicting actual VA at 2 years. The mean AUCs for predicting ≥15-letter VA gain and loss at 2 years was 0.84-0.85 and 0.58-0.73, respectively. In final validation on the test dataset up to week 12, the models had an R2 of 0.33-0.38 (MAE=8.9-9.9 letters) for predicting VA change, an R2 of 0.37-0.45 (MAE=8.8-10.2 letters) for predicting actual VA at 2 years, and AUCs of 0.85-0.87 and 0.67-0.79 for predicting ≥15-letter VA gain and loss, respectively. ML models have the potential to predict 2-year VA response to anti-VEGF treatment using clinical and imaging features from the loading dose phase, which can aid in decision-making around treatment protocols for nAMD patients.