Diabetic macular edema (DME) represents a principal cause of vision loss in patients with diabetic retinopathy, with global prevalence estimated at 19 million cases and projected to increase to 29 million by 2045.^1-3 When left untreated, DME can progress from subtle blurring to severe, irreversible central vision loss, making early identification and individualized management essential.⁴

The therapeutic landscape has evolved substantially over the past two decades. Intravitreal anti-VEGF therapy has replaced laser photocoagulation as the gold standard for center-involving DME (CIDME), supported by landmark trials including DRCR.net Protocols I and T.^5,6 Newer agents such as faricimab (Vabysmo; Genentech) and high-dose aflibercept (Eylea HD; Regeneron) offer extended injection intervals without compromising efficacy.⁷ However, resistance to anti-VEGF treatment is estimated to occur in approximately 40% of DME patients, potentially due to differing biological profiles, where elevated inflammatory cytokines may contribute to poor response to anti-angiogenic monotherapy.^8,9 For these patients, intravitreal corticosteroids provide an alternative mechanism targeting inflammatory pathways. Their use is tempered by risks of cataract formation and elevated intraocular pressure, positioning them primarily as second-line agents.

Contemporary evidence suggests that conventional systemic risk factors, while important, are insufficient to explain the marked heterogeneity in treatment response observed in clinical practice. This has resulted in greater interest in imaging biomarkers to improve disease stratification and guide therapeutic decision-making.

Overview of Imaging Modalities for DME

Multimodal imaging is central to the diagnosis, monitoring, and treatment decision-making in DME. Optical coherence tomography (OCT) has emerged as a cornerstone of clinical assessment, providing noninvasive, high-resolution, cross-sectional imaging of retinal microstructure. OCT enables precise measurement of central subfield thickness (CST), characterization of fluid morphology, and identification of structural features such as hyperreflective foci and disruptions of the outer retinal layers, findings that directly inform treatment planning and response monitoring.^4,10

Fluorescein angiography (FA) remains the gold standard for evaluating retinal vascular perfusion, detecting leakage from microaneurysms, and identifying macular ischemia, offering information that OCT alone cannot provide.¹¹ However, FA’s invasive nature, requirement for intravenous dye, and associated risk of adverse reactions limit its routine use. Optical coherence tomography angiography (OCTA) has also emerged as a powerful noninvasive adjunct, visualizing retinal and choroidal microvasculature through motion contrast imaging without dye injection.¹² OCTA enables assessment of capillary nonperfusion, foveal avascular zone enlargement, and vessel density, which are parameters increasingly relevant to treatment stratification. Widefield swept-source OCTA further extends this capability, correlating with functional outcomes such as contrast sensitivity in DME.¹³

In clinical practice, OCT images are obtained at every visit for quantitative monitoring, while FA and OCTA are used selectively to characterize vascular pathology, evaluate ischemia, or investigate poor treatment response. Together, these modalities provide complementary structural and vascular information to guide individualized care.

Predictors of Anti-VEGF Response

Several OCT biomarkers have demonstrated predictive value for improvements in visual acuity (VA) with longitudinal anti-VEGF therapy. Improvement in disorganization of the retinal inner layers (DRIL) at the fovea—encompassing the ganglion cell layer–inner plexiform layer (GCL-IPL), inner nuclear layer (INL), and outer plexiform layer (OPL)—is associated with improved visual outcomes.⁴ DRIL is thought to reflect impaired visual signal transmission.⁴

Disorganization of the outer retinal layers, including the ellipsoid zone (EZ) and external limiting membrane (ELM) is associated with worse visual function improvement following anti-VEGF therapy.¹⁴ Restoration of these outer reflective bands, including the cone outer segment tip (COST) correlates with improved visual outcomes.

Presence of cystoid macular edema (CME) is predictive of greater reductions in CST following anti-VEGF therapy.⁴ Diffuse retinal thickening is also predictive of greater CST reduction over time.⁴ Reduction in outer nuclear layer (ONL) cyst burden is correlated with improved visual function, likely due to decreased vascular permeability.¹¹ Eyes with serous retinal detachment demonstrate greater decrease in CST following anti-VEGF injection.⁴

Absence of hyperreflective retinal foci (HRF) predicts improved visual outcomes following anti-VEGF.^4,11 These HRF are thought to represent lipoprotein extravasation and microglial cell activation. Such markers can provide insight into the level of retinal and choroidal inflammation present.⁴

Increased choroidal thickness is a biomarker correlated with responsiveness to anti-VEGF therapy, with the thickening possibly reflecting VEGF-induced vasodilation.^11,15 Further ELM and EZ disruption further predispose to choroidal involvement.⁴ Further choroidal characteristics such as presence of hyper-reflective foci are predictive of increase in visual function after anti-VEGF therapy.^4,11

Predictors of Steroid Responders

EZ continuity is another marker of more optimal visual function following steroid implant therapy.^16,17 However, direct comparison to anti-VEGF therapy is limited. Improvements via anti-VEGF therapy have been documented as well. ELM continuity is also predictive of improved response to either treatment approach.¹⁸ In cases of EZ or ELM discontinuity, there are potential benefits to transition to corticosteroid therapy for DME from anti-VEGF due to the anti-inflammatory benefits.¹⁸

Among eyes with subfoveal subretinal fluid, corticosteroid implants are associated with greater improvement in DRIL compared to anti-VEGF therapy.^4,19 However, these structural improvements do not seem to correlate to visual function improvements.^4,19

HRF is a predictor for DME recurrence following steroid therapy via implant, although steroids may achieve greater resolution of HRF compared to anti-VEGF. Further, absence of HRF is associated with higher visual function following steroid implant therapy. Taken together, HRF may provide a prompt for closer monitoring of DME recurrence after treatment.¹¹

Subfoveal hard exudates may predict improved responsiveness to steroid therapy via implants or intravitreal injection compared to anti-VEGF therapy.²⁰ Intraretinal cysts are associated with improvement in DME following steroid implant therapy.²¹

The clinical utility of subretinal fluid (SRF) in steroid treatment decision-making remains highly debated. Some studies support improved response with steroid implants and more optimal visual function outcomes in eyes with SRF.^11,16,17 However, direct comparisons with anti-VEGF therapy remain limited.

Combination Therapy in Refractory DME

In cases of DME that are refractory to monotherapy with intravitreal steroids or anti-VEGF, a combination approach is often considered. A 2018 Cochrane Review article found no significant change in visual acuity or central macular thickness after a year in combination therapy vs anti-VEGF or steroid treatment monotherapy, though comparisons to steroid monotherapy were more limited.²² However, more recent evidence suggests that combination therapy is associated with improvements in central retinal thickness and visual acuity when compared in refractory cases of DME.²³

Role of Structural Factors

Structural factors may also play a role in mediating the relationship between imaging biomarkers and responsiveness to steroids or anti-VEGF therapy. Vitreomacular interface abnormalities, including epiretinal membrane (ERM) and vitreomacular traction (VMT) are associated with reduced responsiveness to anti-VEGF therapy.²⁴ These factors are also associated with lower reduction in CMT and less improvement in visual acuity.²⁴ Lastly, chronic DME may result in improved responsiveness to corticosteroid therapy over anti-VEGF due to the increased inflammatory contribution.²¹

OCT-Based Evaluation of Response to Therapy

Response to therapy can be evaluated with multiple quantitative and qualitative metrics. The Early Treatment Diabetic Retinopathy Study (ETDRS) defines clinically significant macular edema based on (1) central thickening of the retina in proximity of the fovea, (2) presence of exudates, or (3) an extensive area of retinal thickening greater than one disc area.²¹ Additionally, distinction between CIDME and non-CIDME provides insights into disease severity, location of edema, and, ultimately, the treatment approach.²¹

OCTA provides additional information regarding the retinal and choroidal vasculature in DME. DME eyes with lower macular vessel density on OCTA have demonstrated more substantial decreases in central retinal thickness following anti-VEGF therapy—possibly reflecting improved response to treatment.²⁵ Expanded-field and ultrawidefield OCTA do correlate with contrast sensitivity in DME.¹³

Clinical Decision-Making

Therefore, beyond just CST, there are multiple factors to evaluate in the clinical setting. These include DRIL, EZ/ELM integrity, presence or absence of HRF, and presence or absence of intraretinal and/or subretinal fluid or cystic pockets.

Often DME is VEGF-driven, which supports anti-VEGF as the primary treatment modality for resolution. Such cases often involve CME, diffuse retinal or choroidal thickening.

Meanwhile, steroid therapy is considered in cases refractory to anti-VEGF or with characteristics suggesting a significant inflammatory component. Such features include HRF, serous retinal detachment, poor outer retinal integrity, and chronic or recurrent DME. Based on these factors, corticosteroids may be considered for adjunctive or monotherapy. However, in all patients, structural factors such as ERM or VMT should be factors to consider when counseling on treatment efficacy and visual function prognosis.

Determining follow-up timing requires individualized assessment. Response to therapy can be variable and may differ by retinal sector. Therefore, thorough assessment longitudinally is critically important. Biomarkers visualized on imaging may suggest complex overlapping mechanisms. While useful, these biomarkers should be taken together alongside other individual health and social factors in determining optimal treatment approaches.

Conclusions

Overall, multimodal imaging biomarkers in DME can be valuable in predicting response to anti-VEGF or corticosteroid therapy. Determining the relative contributions of VEGF or inflammatory-driven disease to the DME is useful to consider when initiating therapy. Furthermore, consistent monitoring of treatment response with modern retinal imaging tools is critical. RP

Kailynn M. Barton, BS, is a medical student at Tufts University School of Medicine currently pursuing a research year at the Harvard Retinal Imaging Laboratory under the guidance of Dr. John B. Miller. She reports no disclosures.

Muhammad Abidi, BS, is a medical student at Harvard Medical School and a member of the research team at the Harvard Retinal Imaging Laboratory. He reports no disclosures.

Rajiv M. Sastry, BA, is a teaching fellow at Harvard University and a member of the research team at the Harvard Retinal Imaging Laboratory. He reports no disclosures.

John B. Miller, MD, is an associate professor of ophthalmology at Massachusetts Eye and Ear of Harvard Medical School in Boston and director of the Harvard Retinal Imaging Laboratory. He discloses support from Adaptive Sensory Technology and Intalight, and consulting work for Alcon, Boehringer, Genentech, Sumitomo Pharma America, Topcon, and Carl Zeiss Meditec. He can be reached at john_miller@meei.harvard.edu.

References

1. International Diabetes Federation. IDF Diabetes Atlas 11th Edition. 2025. Accessed April 13, 2026. https://diabetesatlas.org/atlas/

2. Yau JWY, Rogers SL, Kawasaki R, et al. Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care. 2012;35(3):556-564. doi:10.2337/dc11-1909

3. Teo ZL, Tham YC, Yu M, et al. Global prevalence of diabetic retinopathy and projection of burden through 2045: systematic review and meta-analysis. Ophthalmology. 2021;128(11):1580-1591. doi:10.1016/j.ophtha.2021.04.027

4. Szeto SK, Hui VWK, Tang FY, et al. OCT-based biomarkers for predicting treatment response in eyes with centre-involved diabetic macular oedema treated with anti-VEGF injections: a real-life retina clinic-based study. Br J Ophthalmol. 2023;107(4):525-533. doi:10.1136/bjophthalmol-2021-319587

5. Bressler SB, Qin H, Beck RW, et al. Factors associated with changes in visual acuity and central subfield thickness at 1 year after treatment for diabetic macular edema with ranibizumab. Arch Ophthalmol. 2012;130(9):1153-1161. doi:10.1001/archophthalmol.2012.1107

6. Jampol LM, Glassman AR, Bressler NM. Comparative effectiveness trial for diabetic macular edema: three comparisons for the price of 1 study from the Diabetic Retinopathy Clinical Research Network. JAMA Ophthalmol. 2015;133(9):983-984. doi:10.1001/jamaophthalmol.2015.1880

7. Wykoff CC, Abreu F, Adamis AP, et al. Efficacy, durability, and safety of intravitreal faricimab with extended dosing up to every 16 weeks in patients with diabetic macular oedema (YOSEMITE and RHINE): two randomised, double-masked, phase 3 trials. Lancet. 2022;399(10326):741-755. doi:10.1016/S0140-6736(22)00018-6

8. Arima M, Nakao S, Yamaguchi M, et al. Claudin-5 redistribution induced by inflammation leads to anti-VEGF-resistant diabetic macular edema. Diabetes. 2020;69(5):981-999. doi:10.2337/db19-1121

9. Mao J, Zhang S, Zheng Z, et al. Prediction of anti-VEGF efficacy in diabetic macular oedema using intraocular cytokines and macular optical coherence tomography. Acta Ophthalmol. 2022;100(4):e891-e898. doi:10.1111/aos.15008

10. Virgili G, Menchini F, Casazza G, et al. Optical coherence tomography (OCT) for detection of macular oedema in patients with diabetic retinopathy. Cochrane Database Syst Rev. 2015;1(1):CD008081. doi:10.1002/14651858.CD008081.pub3

11. Markan A, Agarwal A, Arora A, Bazgain K, Rana V, Gupta V. Novel imaging biomarkers in diabetic retinopathy and diabetic macular edema. Ther Adv Ophthalmol. 2020;12:2515841420950513. doi:10.1177/2515841420950513

12. Guo J, Shen Y, Gu H, et al. Recent advances and applications of optical coherence tomography angiography in diabetic retinopathy. Front Endocrinol. 2025;16:1438739. doi:10.3389/fendo.2025.1438739

13. Baldwin G, Vingopoulos F, Garg I, et al. Structure-function associations between contrast sensitivity and widefield swept-source optical coherence tomography angiography in diabetic macular edema. Graefes Arch Clin Exp Ophthalmol. 2023;261(11):3113-3124. doi:10.1007/s00417-023-06086-1

14. Hatz K, Ebneter A, Tuerksever C, Pruente C, Zinkernagel M. Repeated dexamethasone intravitreal implant for the treatment of diabetic macular oedema unresponsive to anti-VEGF therapy: outcome and predictive SD-OCT Features. Ophthalmologica. 2018;239(4):205-214. doi:10.1159/000485852

15. Rayess N, Rahimy E, Ying GS, et al. Baseline choroidal thickness as a predictor for response to anti-vascular endothelial growth factor therapy in diabetic macular edema. Am J Ophthalmol. 2015;159(1):85-91.e913. doi:10.1016/j.ajo.2014.09.033

16. Moon BG, Lee JY, Yu HG, et al. Efficacy and safety of a dexamethasone implant in patients with diabetic macular edema at tertiary centers in Korea. J Ophthalmol. 2016;2016:9810270. doi:10.1155/2016/9810270

17. Zur D, Iglicki M, Busch C, et al. OCT biomarkers as functional outcome predictors in diabetic macular edema treated with dexamethasone implant. Ophthalmology. 2018;125(2):267-275. doi:10.1016/j.ophtha.2017.08.031

18. Cavalleri M, Cicinelli MV, Parravano M, et al. Prognostic role of optical coherence tomography after switch to dexamethasone in diabetic macular edema. Acta Diabetol. 2020;57(2):163-171. doi:10.1007/s00592-019-01389-4

19. Vujosevic S, Torresin T, Bini S, et al. Imaging retinal inflammatory biomarkers after intravitreal steroid and anti-VEGF treatment in diabetic macular oedema. Acta Ophthalmol. 2017;95(5):464-471. doi:10.1111/aos.13294

20. Shin YU, Hong EH, Lim HW, Kang MH, Seong M, Cho H. Quantitative evaluation of hard exudates in diabetic macular edema after short-term intravitreal triamcinolone, dexamethasone implant or bevacizumab injections. BMC Ophthalmol. 2017;17(1):182. doi:10.1186/s12886-017-0578-0

21. Dadkhah PA, Taheri H, Noroozi M, et al. Therapeutic approaches to diabetic macular edema assessed using optical coherence tomography and optical coherence tomography angiography. Int J Ophthalmol. 2026;19(1):160-174. doi:10.18240/ijo.2026.01.20

22. Mehta H, Hennings C, Gillies MC, Nguyen V, Campain A, Fraser-Bell S. Anti–vascular endothelial growth factor combined with intravitreal steroids for diabetic macular oedema. Cochrane Database Syst Rev. 2018;4(4):CD011599.

23. Heinke A, Warter A, Cavichini M, et al. Combination intravitreal steroid and anti-VEGF therapy for double-monotherapy-resistant chronic diabetic macular edema. Ophthalmic Surg Lasers Imaging Retina. 2025;56(11):664-671. doi:10.3928/23258160-20250813-03

24. Gong Y, Wang M, Li Q, Shao Y, Li X. Evaluating the effect of vitreomacular interface abnormalities on anti-vascular endothelial growth factor treatment outcomes in diabetic macular edema by optical coherence tomography: a systematic review and meta-analysis. Photodiagnosis Photodyn Ther. 2023;42:103555. doi:10.1016/j.pdpdt.2023.103555

25. Hein M, Vukmirovic A, Constable IJ, et al. Angiographic biomarkers are significant predictors of treatment response to intravitreal aflibercept in diabetic macular edema. Sci Rep. 2023;13(1):8128. doi:10.1038/s41598-023-35286-2