Can Artificial Intelligence Lead to Personalized Macular Edema Treatment?

Using automated spectral-domain optical coherence tomography (OCT), researchers found the presence of subretinal fluid (SRF) in individuals with diabetic macular edema (DME) is associated with lower baseline best-corrected visual acuity (BCVA). However, the results, published in JAMA Ophthalmology, also show SRF is associated with good response to anti-vascular endothelial growth factor (VEGF) therapy.

The findings of the retrospective study suggest automated quantification of intraretinal fluid (IRF) and SRF may be an objective approach to assess DME treatment.

Although some of the most important signaling molecules involved in the process of diabetic retinopathy (DR) have been identified, “the pathomechanisms of DME, a common and visually detrimental complication of DR, are still not completely understood,” authors wrote.

To better examine the volumetric change of IRF and SRF in DME during anti-VEGF treatment, researchers developed deep learning algorithms based off of data collected as part of the Diabetic Retinopathy Clinical Research Network (protocol T) trial.

“Artificial intelligence (AI) has recently become available to a larger community of researchers worldwide owing to improvements in computer hardware and software and represents a breakthrough in big data management,” authors said. “Because IRF and SRF are associated with DME, automated and objective quantification is essential not only for a more profound understanding of the disease but also for objective evaluation of treatment efficacy (ie, resolution of IRF and SRF).”

The post hoc analysis, conducted between December 2017 and January 2020, included 6945 spectral-domain OCT volume scans from 570 eyes of the 570 participants presenting with DME. All patients were treated between 2012 and 2018 with aflibercept (n = 190), ranibizumab (n = 188), or bevacizumab (n = 192), with or without deferred laser.

Of the 570 participants, the majority (53%) were male and white (65%) with an average age of around 43 years. At baseline all 570 eyes (100%) had IRF and 235 (41%) had SRF, while at a 12-month follow up, a total of 452 of 514 eyes (88%) still had IRF, and 82 of 514 eyes (16%) had SRF.

Analyses found:

• Presence of SRF at baseline was associated with a worse baseline BCVA Early Treatment Diabetic Retinopathy Study (ETDRS) score of 63.2 (95% CI, 60.2-66.1) (approximate Snellen equivalent of 20/63 [95% CI, 20/50-20/63]) in eyes with SRF vs 66.9 (95% CI, 65.7-68.1) (approximate Snellen equivalent, 20/50 [95% CI, 20/40-20/50]) without SRF (P&thinsp;<&thinsp;.001)
• SRF at baseline was also associated with a greater gain in ETDRS score (0.5; 95% CI, 0.3-0.8) every 4 weeks during follow-up in eyes with SRF at baseline vs 0.4 (95% CI, 0.3-0.5) in eyes without SRF at baseline (P&thinsp;=&thinsp;.02) when adjusted for baseline BCVA
• Aflibercept was associated with greater reduction of IRF volume compared with bevacizumab after the first injection (difference, 79.8 nL; 95% CI, 5.3-162.5 nL; P&thinsp;<&thinsp;.001) and every 4 weeks thereafter (difference, 10.4 nL; 95% CI, 0.7-20.0 nL; P&thinsp;=&thinsp;.004)
• Ranibizumab was associated with a greater reduction of IRF after the first injection compared with bevacizumab (difference, 75.2 nL; 95% CI, 1.4-154.7 nL; P&thinsp;<&thinsp;.001)

“With use of AI, an automated and objective profiling of the efficacy of different substances was possible by automated fluid quantification,” authors wrote. “Despite the limited correlation with visual function found in our study, the presence and amount of IRF and SRF served as measures of individual treatment response. Thus, automated fluid segmentation could support the clinical applicability of individualized medicine.”

Although SRF had a negative association with BCVA at baseline but indicated a good response to anti-VEGF treatment, the presence of SRF is also associated with good visual prognosis in a variety of retinal diseases, including age-related macular degeneration. This suggests a potential protective association between SRF and photoreceptors independent of underlying disease.

“Artificial intelligence may be used to automatically measure fluid volumes in a reliable way and thus may facilitate a comparison of the differential efficacy of anti-VEGF agents,” researchers concluded.

Reference

Roberts PK, Vogl W, Gerendas BS, et al. Quantification of fluid resolution and visual acuity gain in patients with diabetic macular edema using deep learning. JAMA Ophthalmol. Published online July 23, 2020. doi:10.1001/jamaophthalmol.2020.2457