Prognostic Model Predicts Therapeutic Responses of Eyes With Center-Involved DME to Anti-VEGF Therapy

Prognostic model can predict the therapeutic responses of eyes with center-involved diabetic macular edema (CI-DME) to anti-vascular endothelial growth factor (VEGF) therapy, according to a study presented at the Association for Research in Vision and Ophthalmology (ARVO) Annual Meeting held in New Orleans, April 23-27, 2023.

The researchers developed an integrated deep-learning (DL) model based on pretreatment optical coherence tomography (OCT) images and clinical variables to predict the therapeutic response of eyes with CI-DME after 3 monthly anti-VEGF injections.

The prognostic model was trained, tuned, and validated using retrospective data, including 705 pre-treatment OCT volumes, from 457 eyes with CI-DME of 366 patients who received anti-VEGF treatment at 3 retina clinics in Hong Kong.

The integrated DL model classified eyes as having “good response”, defined as ≥ 1-line gain on the Snellen visual acuity (VA) chart or >10% reduction in central subfield thickness (CST), or “no response”, defined as less than 1-line gain on the Snellen VA chart and ≤ 10% reduction in CST, after 3 months of treatment.

Performance of the prognostic model was evaluated with external testing using 3 independent datasets from Australia and India.

In the internal validation, the integrated DL model achieved an area under the receiver operating characteristic curve of 0.731 (95% confidence interval [CI], 0.619-0.844), sensitivity of 70.0% (95% CI, 55.6-87.4), specificity of 70.2% (95% CI, 50.5-79.3), and accuracy of 69.5% (95% CI, 62.4-77.3).

In the external testing using 3 independent datasets, the DL model achieved area under the receiver operating characteristic curve of 0.677, 0.652, and 0.669, and accuracies of 76.3%, 78.1% and 68.4%, respectively.

“Our proposed DL system may be useful as a clinical decision support tool to classify eyes with CI-DME into “likely” or “unlikely” to have good prognosis when treated with anti-VEGF agents for assisting therapeutic management decision,” according to the researchers.