โก Quick Summary
This study explored the use of artificial intelligence to identify latent phenotypic subgroups of diabetic macular edema (DME) and assessed how treatment responses to anti-VEGF and dexamethasone (DEX) therapies vary among these groups. The findings revealed three distinct clusters of DME, each with unique characteristics and treatment responses, highlighting the potential for tailored therapeutic approaches.
๐ Key Details
- ๐ Sample Size: 114 eyes from 82 treatment-naรฏve patients
- ๐งฉ Metrics Analyzed: Intraretinal fluid (IRF) and subretinal fluid volumes, hyperreflective foci counts, ellipsoid zone (EZ) disruption
- โ๏ธ Methodology: Gaussian finite mixture modeling and mixed-effects models
- ๐ Main Outcomes: Changes in visual acuity (VA) and OCT parameters post-treatment
๐ Key Takeaways
- ๐ Three distinct DME clusters were identified, each with unique structural and functional characteristics.
- ๐ Cluster 1: Localized central IRF, moderate structural damage, better VA (mean LogMAR 0.29).
- ๐ Cluster 2: Diffuse IRF, highest IRF volume, significant structural disruption, poorest VA (mean LogMAR 0.63).
- ๐ Cluster 3: Intermediate fluid levels, minimal structural damage (EZ disruption: 0.5%).
- ๐ก Anti-VEGF therapy showed the greatest VA improvement in Cluster 2.
- โ๏ธ No significant VA differences between DEX and anti-VEGF in Clusters 1 and 3.
- ๐ DEX achieved a greater reduction in central subfield thickness compared to anti-VEGF in Cluster 3.
- ๐ AI-derived OCT metrics could support personalized treatment strategies for DME.
๐ Background
Diabetic macular edema (DME) is a common complication of diabetes that can lead to significant vision loss. Understanding the underlying phenotypic variations in DME is crucial for optimizing treatment strategies. Traditional approaches often fail to account for the heterogeneity in DME presentations, which can influence patient outcomes. This study leverages artificial intelligence to uncover latent phenotypic clusters, paving the way for more personalized treatment options.
๐๏ธ Study
Conducted as a retrospective analysis, this study included 114 eyes from 82 patients with treatment-naรฏve DME. The researchers utilized advanced optical coherence tomography (OCT) metrics to identify distinct phenotypic subgroups and evaluate treatment responses to anti-VEGF and DEX therapies. The study aimed to enhance our understanding of DME and improve therapeutic outcomes through tailored approaches.
๐ Results
The analysis revealed three phenotypic clusters of DME, each exhibiting unique characteristics. Cluster 1 had localized IRF and better visual acuity, while Cluster 2, with diffuse IRF, showed the highest fluid volume and poorest visual acuity. Cluster 3 presented intermediate features. Notably, anti-VEGF therapy resulted in the most significant improvement in visual acuity for patients in Cluster 2, while DEX demonstrated superior efficacy in reducing central subfield thickness in Cluster 3.
๐ Impact and Implications
The findings from this study underscore the importance of recognizing the latent heterogeneity in DME presentations. By employing AI-derived OCT metrics, clinicians can better tailor treatment strategies to individual patient profiles, potentially leading to improved outcomes. This research highlights a significant step towards personalized medicine in ophthalmology, suggesting that understanding phenotypic variations can enhance therapeutic efficacy and patient care.
๐ฎ Conclusion
This study illustrates the potential of artificial intelligence in identifying distinct phenotypic clusters of DME and their differential responses to treatment. By leveraging advanced OCT metrics, healthcare providers can adopt more personalized therapeutic approaches, ultimately optimizing patient outcomes. The future of DME management looks promising, with AI playing a pivotal role in enhancing treatment strategies.
๐ฌ Your comments
What are your thoughts on the use of AI in identifying phenotypic variations in DME? We would love to hear your insights! ๐ฌ Leave your comments below or connect with us on social media:
What Lies beneath Diabetic Macular Edema: Latent Phenotypic Clustering and Differential Treatment Responses to Intravitreal Therapies.
Abstract
PURPOSE: To identify latent phenotypic subgroups of diabetic macular edema (DME) using artificial intelligence-based OCT metrics and evaluate whether treatment responses to anti-VEGF and dexamethasone (DEX) therapies differ across these phenotypic clusters.
METHODS: Retrospective study including 114 eyes (82 patients) with treatment-naรฏve DME. Quantitative OCT metrics, including intraretinal fluid (IRF) and subretinal fluid volumes, IRF % distribution within central 0-1, 1-3, and 3-6 mm, hyperreflective foci counts, and ellipsoid zone (EZ) % disruption, were analyzed before and after treatment.
MAIN OUTCOME MEASURES: Gaussian finite mixture modeling was used to identify distinct DME subgroups. Changes in visual acuity (VA) and OCT parameters following anti-VEGF or DEX therapy were analyzed using linear and generalized linear mixed-effects models, with false discovery rate correction applied to account for multiple comparisons.
RESULTS: Three phenotypic clusters of DME were identified, each demonstrating distinct structural and functional characteristics: cluster 1 (29%, 95% confidence interval [CI]: 20.0%-38.4%), characterized by localized central IRF (mean 0.34 mm3, 32% in the 0-1 mm zone), moderate structural damage (EZ disruption: 13%), and better VA (mean logarithm of the minimum angle of resolution [LogMAR] 0.29); cluster 2 (49%, 95% CI: 39.6%-57.9%), with diffuse IRF (60% in the 3-6 mm zone), the highest IRF volume (mean: 3.33 mm3), significant structural disruption (EZ disruption: 46%), and the poorest VA (mean LogMAR: 0.63); and cluster 3 (22%, 95% CI: 13.9%-31.2%), showing intermediate fluid levels and minimal structural damage (EZ disruption: 0.5%). Anti-VEGF therapy led to the greatest VA improvement in cluster 2 (-31.5%, standard deviation: 28.6). Pairwise contrasts showed no significant VA differences between DEX and anti-VEGF in cluster 1 (-26.6%, 95% CI: -64.7 to 11.6) or in cluster 3 (-12.4%, 95% CI: -58.2 to 33.4), although the direction of effect suggested a trend toward greater improvement with DEX. In contrast, cluster 2 showed a nonsignificant difference favoring anti-VEGF (+25.0%, 95% CI: -4.6 to 54.6). For central subfield thickness, DEX achieved a significantly greater reduction than anti-VEGF in cluster 3 (-20.9%, 95% CI: -37.0 to -4.9) and was also associated with a relative increase in peripheral IRF distribution in cluster 3 (+26.7%, 95% CI: 6.5 to 46.9), supporting phenotype-dependent treatment effects.
CONCLUSIONS: Latent heterogeneity in DME presentations may influence treatment responses. Artificial intelligence-derived spectral-domain OCT metrics could support tailored therapeutic approaches to optimize patient outcomes.
FINANCIAL DISCLOSURES: Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.
Author: [‘Cicinelli MV’, ‘Leonardo B’, ‘Maiucci G’, ‘Martino G’, ‘Ziafati M’, ‘Bousyf S’, ‘Frizziero L’, ‘Lattanzio R’, ‘Midena E’, ‘Bandello F’]
Journal: Ophthalmol Sci
Citation: Cicinelli MV, et al. What Lies beneath Diabetic Macular Edema: Latent Phenotypic Clustering and Differential Treatment Responses to Intravitreal Therapies. What Lies beneath Diabetic Macular Edema: Latent Phenotypic Clustering and Differential Treatment Responses to Intravitreal Therapies. 2026; 6:100975. doi: 10.1016/j.xops.2025.100975