Epidemiology

Table 1 summarizes the prevalence of DR and DME in the Asia-Pacific region. The prevalence of DR in China (18.0–27.9 %), Hong Kong (11.0–39.0 %), Nepal (19.4–23.4 %), Fiji (27.2 %), and Bangladesh (21.6 %) were comparable to global (22.3 %) DR prevalence in the diabetic population. However, studies from Singapore (28.2–33.7 %), Australia (28.5–32.4 %), Sri Lanka (27.4 %), New Zealand (22.5–48.2 %), Indonesia (43.1 %), and the Pacific Islands (40–69 %) displayed a higher prevalence of DR, while the regions of India (10.3–21.7 %), South Korea (15.9 %), Malaysia (8.5 %), Taiwan (3.75 %), and Thailand (5.0 %) revealed lower DR prevalence in diabetic population when compared globally. Variations in prevalence may result from differences in population characteristics and sample methodology. We hypothesize that data from national health insurance or security databases may be less susceptible to selection bias than hospital-based studies where the participation of potentially sicker patients (with more complications) may be less representative of the general population, therefore potentially overestimating DR/DME prevalence. This may account for high DR prevalence in hospital-based studies from China (27.9 %) or Hong Kong (39.0 %) and low DR prevalence in studies using national health insurance or security databases from South Korea (15.9 %), Taiwan (3.75 %), and Thailand (5.0 %) in diabetic population. However, differences in the availability of accessible and effective healthcare services in various regions likely play a larger role in the prevalence of DR. According to a review of Diabetic Retinopathy in the Asia-Pacific by Chua et al., the authors hypothesized that despite a high rate of diagnosed diabetes in South Korea, a low prevalence of DR (15.6 %) in the region is potentially attributed towards an effective diabetes screening and treatment program. However, when compared to Singapore, the prevalence of DR (28.2–33.7 %) is twice as high, despite having comparable diagnosis rates and healthcare systems to South Korea.

The prevalence of DME in Asia-Pacific regions is also summarized in Table 1 . Prevalence of DME in patients with DR in China (6.7–11.3 %), Singapore (7.6 %), India (2.1–13.8 %), Australia (7.6–9.4 %), Nepal (6.9 %), Indonesia (17.1 %), and Sri Lanka (5.3 %) are all greater or comparable to global estimates (4.1–5.5 %).

Urban-rural differences

DR prevalence in rural India was reported to be lower when compared to the urban Indian population (10.3 % vs 15.4 %, respectively), even though the prevalence of diabetes mellitus in rural and urban India was essentially similar (10.4 % vs 11.4 %, respectively). Studies from China displayed similar trends, where the prevalence of DR was reported as 18.0 % among urban Chinese residents, compared to 15.6 % in rural China. A slightly higher prevalence of DR in urban China is supported by previous literature reporting the prevalence of DR as 19.1 %, 15.4 %, and 16.1 % in urban, suburban, and rural residents of Shanghai, respectively. The reason for this urban-rural discrepancy may be the results of rapid urbanization, causing changes to physical activity, eating habits, and drug use, which further leads to the development of risk factors, such as increased insulin resistance and hyperglycemia, that contribute to a higher prevalence of diabetic complications, such as DR and DME.

Ethnic differences

The potential variation of DR and DME prevalence by ethnicity is reported in studies from Singapore and Australia. The population of Singapore consists of 3 major ethnic groups of Indian, Malay, and Chinese origin. Tan et al. reported a higher prevalence of DR amongst the Indian population of 30.7 % compared to 26.2 % in Chinese and 25.5 % in Malays, along with similar findings in the prevalence of DME with 7.6 % in Indians compared to 5.9 % in Chinese and 5.0 % in Malays. An ethnic difference in the prevalence of DR is further seen in the indigenous and non-indigenous Australian populations, where the prevalence of DR was higher amongst indigenous (39.4 %) and lower amongst non-indigenous Australians (28.5 %). However, the variation between the prevalence of DR in indigenous and non-indigenous Australians is inconsistent, with another study reporting a lower prevalence of DR in indigenous Australians (23.6 %) compared to non-indigenous Australians (32.3 %), respectively. This suggests the limited role of ethnicity in predicting DR and DME prevalence in a given population.

The role of ethnicity in relation to the prevalence of DR is largely overshadowed by lifestyle and cultural factors in a region. When compared, a notably lower prevalence of DR was observed in Indians living in India (16.9 % ) than in Indians living in Singapore (30.7 % ). Chua et al. suggested that the higher prevalence of DR reported in Indian Singaporeans may result from the effect of immigration and acculturation. Second-generation ethnic Indian immigrants in Singapore had a higher prevalence of type 2 diabetes and DR compared to first-generation immigrants. Zheng et al. hypothesized that the change from a traditional lifestyle in India to a “Western” lifestyle in Singapore may be a direct determinant of diabetes-related complications, such as DR and DME.

Glycemic control

The development of DR strongly correlates with a longer duration of diabetes and greater hyperglycemia. A higher HbA1C is associated with both increased incidence as well as the progression of DR. Wong et al. reported the OR of DR increased by 1.07 ± 0.2 per year of the duration of the disease. The Action to Control Cardiovascular Risk in Diabetes (ACCORD) trial, a continuation of the Diabetes Control and Complications Trial (DCCT) study, also confirmed the effects of intensive glycemic treatment (target HbA1c <6.0 %) compared with standard treatment (target HbA1c of 7.0–7.9 %) and found a decreased rate of progression of DR in the intensive treatment group. (7.3 % vs. 10.4 %; odds ratio [OR] = 0.67; 95 % confidence interval [CI], 0.51–0.87; P = 0.003). The LALES study found that each year of increased history of diabetes was associated with an 8 % increased risk of having DR, and each 1 % increase in HbA1C was associated with a 22 % increase in the prevalence of DR. A meta-analysis showed that the prevalence of any DR increased with diabetes duration (21.1 vs. 76.3 %, comparing <10 with ≥20 years) and HbA1C (18.0 vs. 51.2 %, comparing levels ≤7.0 with >9.0 %).

Hypertension

Hypertension has consistently been demonstrated to have a positive association with the development of DR. The LALES study found an OR of 1.26 ( P = 0.002) for every 20 mm Hg increase in blood pressure (BP). The UKPDS 69 study demonstrated that tight control of BP with a target level of 150/85 mm Hg, rather than lose control of less than 180/105 mm Hg, statistically significantly decreased the development of microaneurysms (RR = 0.66; P < 0.001), hard exudates (RR = 0.53; P < 0.001), and cotton-wool spots (RR = 0.53; P < 0.001). The beneficial role of angiotensin-converting enzyme inhibitors and drugs acting on the Renin-Angiotensin system in preventing progression of DR has also been well established. A meta-analysis showed that the prevalence of any DR increased with BP (30.8 vs. 39.6 %, comparing BP ≤140/90 or >140/90 mm Hg).

Hyperlipidemia

Studies showed inconsistent results in the relationship between hyperlipidemia and DR. A meta-analysis reported that elevated total serum cholesterol was associated with a higher prevalence of DME and vision-threatening DR. In a large-scale study of 2954 patients by Cheng et al., elevated triacylglycerol levels were significantly associated with DR (OR = 1.29; 95 % CI, 1.05–1.58; P < 0.05). However, the Hoorn study found no relationship between total cholesterol level, serum triglyceride, and incidence of DR. In another large-scale study of 2535 patients with type 2 diabetes, DR was associated with triglycerides and high-density lipoprotein cholesterol in matched analysis but not significantly after additional adjustment. In the Singapore Malay Eye Study, a higher total cholesterol level was even reported to be protective of DR (OR = 0.73, per 1 mmol/L increase). Fenofibrate has been shown to likely reduce progression in patients with overt retinopathy. Patients are being recruited for protocol-AF by the DRCR.net to study the protective role of fenofibrates in patients with DR but no CI-DME.

Obesity

Obesity is another risk factor commonly associated with DR. Raum et al. found that vision-threatening DR was associated with obesity (OR = 3.29; 95 % CI, 1.504–7.206; P = 0.0029). A meta-analysis also showed that obesity was associated with DR risk (RR = 1.34; 95 % CI, 1.06–1.68). Greater waist-hip ratio is also positively associated with DR. In the study by Rajalakshmi et al., the OR of DR is 1.28 per 5-cm increase in waist circumference ( P = 0.014).

Chronic kidney disease

In a large database study with 15,409 individuals, Park et al. found that chronic kidney disease (CKD) (OR = 2.34; 95 % CI, 1.04–5.28) was significantly associated with DR in the diabetic population. Multiple studies have demonstrated the association between DR and clinical aspects of CKD such as lower estimated Glomerular filtration rate (e-GFR), albuminuria, and elevated urine albumin to creatinine ratio.

Sex

Male sex is an independent risk factor for DR. A study performed by Zhang et al. revealed that male sex was independently associated with the presence of DR (OR = 2.07; 95 % CI, 1.39–3.10). The LALES study demonstrated that men had a 50 % higher risk of having any DR when compared with women (OR = 1.50; P = 0.006). The UKPDS 50 study also showed that women had a lower relative risk of progression of DR (RR = 0.54; CI, 0.37–0.80; P = 0.0016).

Myopia

Although myopia is generally harmful to ocular health, it has been observed that the prevalence of DR is lower in myopic patients. A meta-analysis demonstrated that each millimeter increase in axial length (AL) was associated with a significantly decreased risk of DR (OR = 0.75; 95 % CI, 0.65–0.86; P < 0.001); myopic eyes showed a lower risk of DR (OR = 0.70; 95 % CI, 0.58–0.85; P < 0.001); a greater degree of myopic refractive error revealed a decreased risk of DR (OR = 0.89; 95 % CI, 0.85–0.93; P < 0.001).

Pregnancy

Pregnancy itself is an independent risk factor for worsening DR. The DCCT research group found that compared to non-pregnant women, pregnant women had a 1.63-fold and 2.48-fold greater risk of any worsening of retinopathy from before pregnancy to during pregnancy in the intensive treatment group ( P < 0.05) and the conventional treatment group ( P < 0.001), respectively.

Fundus photography

Fundus photography is the most often used imaging tool for clinical studies including standard view and wide field. Historically, the Early Treatment Diabetic Retinopathy Study (ETDRS) Group proposed the standard for diagnosis and classification of severity of DR and DME using the standard stereoscopic, 7-field, 30-degree, film-based, color retinal photographs. Although this classification was adopted as the standard method for classification of DR, it is rarely used in real-world settings mainly due to impracticality. Single-field, or up to 3-field, 45-degree, color retinal photography, instead, has been used, particularly for screening purposes. Digital color retinal photography has finally replaced the film-based photography for diagnosis of DR and DME.

Standard fundus photography provides a 30- to 50-degree image that includes the macula and visualizes lipid exudate, retinal hemorrhage, and vitreous hemorrhage ( Fig. 1 A, B, and C). Multiple images may be manually overlapped to create a montage, such as when the 7 standard 30-degree fundus images are combined to create a 75-degree field of view, which is labor and time-consuming. Instead, ultra-widefield (UWF) imaging can produce a 200-degree retinal view extending to the peripheral retina ( Fig. 1 D). Although UWF has disadvantages such as image distortion, eyelash artifacts, and false color representation of the fundus, it visualizes over 80 % of the total surface retinal area. Recently, artificial intelligence (AI) using fundus photography, machine learning, and deep learning has been adopted by various groups to develop automated DR screening algorithms. The sensitivity and specificity of these algorithms to diagnose DR ranged from 86 % to 100 % and 54–99.17 %, respectively. Recent advance in AI for image analysis, such as Vision Transformer models, generative AI and large models, such as ChatGPT or RETFound, should provide better screening results, better prediction of DR progression, and better recommendation to patients. Challenges for AI in DR or DME rest on deployment rate which is still slow.

Fundus photography, fluorescein angiography, optical coherence tomography (OCT), and OCT angiography (OCTA) in patients with diabetes. Hard exudates appear as yellow lipid exudates in the posterior pole on fundus photography (A). Microaneurysms and intraretinal hemorrhages (B). Neovascularization of the disc area with vitreous hemorrhage in high-risk proliferative diabetic retinopathy (PDR) (C). Fibrovascular proliferation with traction demonstrated with ultra-wide field fundus photography (D). Fluorescein angiography showing focal hyperfluorescent spots representing microaneurysm in non-proliferative diabetic retinopathy (NPDR) (E). Fluorescein angiography showing hyperfluorescent lesions representing neovascularization of the disc and elsewhere in PDR (F). Images of optical coherence tomography (OCT) and OCT angiography (OCTA) (3 ×3 mm, Spectralis OCT, Heidelberg Engineering, Germany) of a healthy normal subject (G, H, and I). NPDR without diabetic macular edema (J, K, and L). Severe NPDR with diabetic macular edema (M, N, and O). OCTA images show superficial capillary plexus (H, K, and N) and deep capillary plexus (I, L, and O).

Fluorescein angiography

Fluorescein angiography (FA) has been the gold standard to evaluate retinal microvasculature changes in DR, including microaneurysms, capillary nonperfusion from non-proliferative diabetic retinopathy (NPDR), neovascularization from PDR, and breakdown of the blood-retinal barrier shown as DME ( Fig. 1 E and F). Although the standard classifications for DR severity do not include FA in their criteria, FA remains valuable for assessing capillary nonperfusion and neovascularization. Recently, ultra-widefield fluorescein angiography (UWF FA) has been introduced, enabling imaging of up to 200 degrees of the retinal surface in a single shot, which can assist researchers in quantifying peripheral ischemia in DR.

Optical coherence tomography

Optical coherence tomography (OCT) has become the gold standard imaging modality to diagnose DME ( Fig. 1 G, J, and M). Current commercial spectral-domain OCT can visualize high-resolution, cross-sectional, and en face images of the retina with axial resolution up to 1μm. For DME, several classifications are suggested based on OCT morphology, such as diffuse retinal thickening, cystoid macular edema, serous subretinal fluid without posterior hyaloid traction, DME with posterior hyaloid traction, and mixed patterns. These patterns are useful to predict the treatment response. For example, eyes with posterior hyaloid traction may not respond to anti-VEGF and require pars plana vitrectomy to release the traction. Furthermore, disorganization of the retinal inner layers in the fovea correlated with VA in eyes with current or resolved DME.

Optical coherence tomography angiography

Optical coherence tomography angiography (OCTA) is a recent imaging modality to visualize the retinal and choroidal flow without dye injection. Instead of using dye, motion contrast is utilized to create high-resolution, depth-resolved, volumetric, angiographic flow images. Most devices have automated segmentation of the retina, with optional manual adjustment of automated lines. The Heidelberg Spectralis (Heidelberg Engineering, Heidelberg, Germany) displays segmented images in four layers: superficial capillary plexus, deep capillary plexus, outer retinal layer, and choroidal circulation. While FA evaluates only the superficial capillary plexus, OCTA identifies the superficial and deep capillary plexuses and the choroidal circulation. For example, patients with DR show a larger mean foveal avascular zone compared with control eyes ( Fig. 1 K, L, N, and O). Furthermore, cystoid macular edema appears as dark round spaces with smooth borders which is different from foci of capillary nonperfusion, with a grayer hue and irregular border ( Fig. 1 N and O). Due to the automatic algorithms, many studies reported the correlation of quantitative analysis with clinical phenotypes of DR; specifically, the capillary density index decreased with worsening DR severity.

Glycemic control

The DCCT, UKPDS, and ACCORD trials support that intensive control of hyperglycemia (HbA1C: 7 %) reduces the risk of development and progression of DR and every percent reduction in HbA1C lowers the risk of retinopathy by 30–40 %. However, in ACCORD trial, more intensive control of hyperglycemia (HbA1C: <6 %) increased all-cause mortality (HR = 1.22; 95 % CI, 1.01–1.46; P = 0.04) and was therefore not recommended.

Blood pressure control

It has been reported that every 10 mm Hg increase in systolic BP is associated with roughly a 10 % excess risk of early DR and a 15 % excess risk of PDR. The UKPDS 69 study showed that in people with type 2 diabetes, tight BP control (150/85 mm Hg) reduced the risks of DR (RR = 0.7). In EUCLID trial, lisinopril demonstrated a reduction in the risk of DR progression by 50 % and PDR by 80 % in normotensive patients with type 1 diabetes. In the study by Schrier et al., patients with normotensive type 2 diabetes treated with nisoldipine or enalapril demonstrated less progression of DR compared to those treated with placebo (34 % vs. 46 %, P = 0.019). One meta-analysis showed that renin-angiotensin system inhibitors reduce the risk of DR (RR = 0.87; 95 % CI, 0.80–0.95; P = 0.002), and increase the possibility of DR regression (RR = 1.39; 95 % CI 1.19–1.61; P = 0.00002). A recent large population base study verified that both poorly controlled hypertension (OR = 1.97; 95 %CI, 1.39–2.83) and untreated hypertension (OR = 2.01; 95 %CI, 1.34–3.05) were significantly associated with any DR.

Lipid-lowering therapy

The FIELD trial showed that fenofibrate reduced the need for laser treatment of vision-threatening DR by 31 % in patients with type 2 diabetes ( P = 0.0002). In the ACCORD trial, the rates of progression of DR were lower when treated with fenofibrate (OR = 0.60; 95 % CI, 0.42–0.87; P = 0.006). In the most recent study conducted in the UK, 1151 patients with non-referrable DR or DME were randomized to fenofibrate or placebo. Progression to referable DR or DME occurred in 131 (22.7 %) of 576 participants in the fenofibrate group and 168 (29.2 %) of 575 in the placebo group during a median of 4-year follow-up, (HR = 0.73; 95 % CI, 0.58–0.91; P = 0.006). For any progression of DR or DME, the frequencies were 185 (32.1 %) vs. 231 (40.2 %) (HR = 0.74; 95 % CI, 0.61–0.90); for the development of DME they were 22 (3.8 %) vs. 43 (7.5 %) (HR = 0.50; 95 % CI, 0.30–0.84). The safety profiles were not different between the two groups.

Kang et al. demonstrated that patients with type 2 diabetes treated with statin had a significantly lower rate of DR (HR = 0.86; 95 % CI, 0.81–0.91), vitreous hemorrhage (HR = 0.62; 95 % CI, 0.54–0.71), tractional retinal detachment (HR = 0.61; 95 % CI, 0.47–0.79), and DME (HR = 0.60; 95 % CI, 0.46–0.79) than those not treated with statin. One meta-analysis showed that lipid-lowering drugs were associated with a reduced risk of DR progression (OR = 0.77; 95 % CI, 0.62–0.96; P = 0.02), and similar results were echoed by Vail et al. (HR = 0.60; 95 % CI, 0.55–0.65; P < 0.01) and Kawasaki et al. (HR = 0.80; 95 % CI, 0.70–0.92; P = 0.002).

Bevacizumab

Bevacizumab is a recombinant humanized monoclonal IgG1 antibody that binds to all VEGF-A isoforms whose role in the treatment of DME has been established in multiple trials. Bevacizumab remains an off-label but common initial intravitreal treatment of DME. No new trials evaluating the safety or efficacy of bevacizumab alone were published in the last 5 years.

Ranibizumab

Ranibizumab, a recombinant humanized IgG1 monoclonal antibody fragment, was approved for the treatment of DME following the results of DRCR.net, RESTORE, RESOLVE, RIDE, and RISE trials. The recent LUMINOUS trial reaffirmed the results of real-world efficacy and safety of ranibizumab 0.5 mg in treatment-naïve patients with DME. At 1 year, the mean visual acuity (VA) ETDRS letter score improved by + 3.5 (n = 502) from a baseline of 57.7 with a mean of 4.5 injections. VA letter score gains were 0.5 (n = 264) in patients requiring ≤ 4 injections and 6.9 (n = 238) in patients ≥ 5, from baseline letter scores of 56.6 and 59.0, respectively. Over 5 years, the incidence of ocular/non-ocular adverse events and serious adverse events was 7.2 %/10.1 % and 0.3 %/5.8 %, respectively.

Aflibercept

Aflibercept is a 115 kDa soluble decoy receptor with VEGF receptors 1 and 2 fused to the Fc portion of human immunoglobulin G1 to enable binding to all isoforms of VEGF-A, VEGF-B, and placental growth factor. Following beneficial outcomes from VIVID-VISTA studies, multiple recent trials have reaffirmed its use in DME. A treat-and-extend study evaluated the change in BCVA and central subfield macular thickness (CSMT) using 2 mg aflibercept with a treat-and-extend (T&E) protocol. Eyes initially received 5 monthly injections and additional treatment was based on the change in CSMT. At one year, there was a statistically significant increase in BCVA ( P < 0.001) with an average of 9.1 letters and a decrease in CSMT of −171.7μm from 489.4 to 317.7μm ( P < 0.001). Over 1 year, the mean number of injections was 8.5, and 74 % of eyes tolerated extension to 12-week intervals. This study demonstrated that at 1 year, anatomic and functional outcomes of T&E with aflibercept were comparable to those of the fixed dosing regimen.

The PERMEATE study evaluated treatment-naïve eyes with foveal involving retinal edema due to DME and vein occlusions. Participants received 2 mg of aflibercept every 4 weeks for the first 6 months, followed by 2 mg every 8 weeks. UWF FA and OCT were used to assess change in pan-retinal leakage at 12 months. The use of aflibercept was associated with improvement in both VA (mean gain 18.4 ± 21.4 letters, P < 0.0001) and reduction in CSMT (301.3 ± 250.3μm, P < 0.0001). Additionally, in DME, aflibercept monthly was associated with a decrease in leakage index from 3.5 ± 2.7 % at baseline to 1.6 ± 0.8 % at month 12 ( P = 0.018), but there was no statistically significant change in overall pan-retinal ischemia at 1 year.

The AQUA trial specifically looked at vision-related quality of life in patients with DME treated with 2 mg intravitreal aflibercept every 8 weeks for 52 weeks after 5 initial monthly loading doses. Patients completed the 25-item National Eye Institute Visual Function Questionnaire (NEI VFQ-25) at baseline and 24 weeks. A mean of 8.8 injections was administered over 52 weeks. At week 52, the mean improvement from baseline in the NEI VFQ-25 total score was + 6.11. This improvement was explained by improvements in quality of life pertaining to both near (+11.37) and distance (+7.33) activities. These functional outcomes corresponded to improvements in BCVA (mean improvement +10.0 letters) and reduction in CSMT (mean reduction −175.38 μm).

Anti-VEGF Comparison studies

Protocol T of DCDR.net randomly assigned eyes to 6 monthly injections of 2.0 mg aflibercept, 1.25 mg bevacizumab, or 0.3 mg ranibizumab and patients were followed for two years. The study itself, as well as planned secondary studies, offered multiple insights into DME management.

The Protocol T Extension Study evaluated anatomic and functional outcomes in long-term follow-up of center involving DME 3 years after the Protocol T completion. Patients were managed at clinical discretion between 2-year and 5-year visits. Of the 68 % of eligible patients who were seen for the 5-year visit, 68 % of patients required at least 1 anti-VEGF injection. When baseline VA was 20/50–20/320, the mean 5-year VA was 11.9 letters (95 % CI, 9.3–14.5) better than baseline but 4.8 letters (95 % CI, 2.5–7.0) worse than 2 years. When baseline VA was 20/32–20/40, the mean 5-year VA was 3.2 letters (95 % CI, 1.4–5.0) better than baseline but 4.6 letters (95 % CI, 3.1–6.1) worse than 2 years. Mean CSMT decreased from baseline to 5 years by 154μm (95 % CI, 142–166) and was stable between 2 and 5 years (-1 μm; 95 % CI, −12–9). These results suggest that anatomic and functional outcomes in these patients did not correlate.

A secondary analysis of DCDR.net Protocol T results included an evaluation of eyes with persistent DME greater than 24 weeks despite initial treatment of 6 monthly injections. Persistent DME for 24 weeks was more common with bevacizumab (65.6 %) than aflibercept (31.6 %) or ranibizumab (41.5 %); (aflibercept vs bevacizumab, P < 0.001; ranibizumab vs bevacizumab, P < 0.001; and aflibercept vs ranibizumab, P = 0.05). Additionally, of patients with persistent DME at 24 weeks despite treatment, chronic persistent DME through 2 years was found in 44.2 % of eyes treated with aflibercept, 68.2 % with bevacizumab (aflibercept vs bevacizumab, P = 0.03), and 54.5 % with ranibizumab (aflibercept vs ranibizumab, P = 0.41; bevacizumab vs ranibizumab, P = 0.16). However, despite persistent DME at 24 weeks or two years, there were no statistically significant differences in vision gain between treatment arms and only 3 eyes with chronic persistent DME lost at least 10 letters.

Brolucizumab

Brolucizumab is a humanized monoclonal single-chain variable fragment that binds and inhibits vascular endothelial growth factor A. The phase 3 HAWK and HARRIER evaluated brolucizumab for use in neovascular age-related macular degeneration and reported a 2.1 % incidence of retinal vasculitis with retinal vascular occlusion. A post-hoc analysis of this data subsequently reported a 4.6 % risk of intraocular inflammation (IOI), including a 3.3 % risk of IOI with associated retinal vasculitis and a 2.1 % risk of these in addition to retinal artery occlusion with the use of the medication in neovascular age-related macular degeneration. Although the real-world data have not exactly mirrored the trials in regards to IOI, careful case selection and prompt management is recommended in the event of IOI.

Subsequently, Phase III studies were recently completed to assess the efficacy and safety of brolucizumab compared to aflibercept in DME. The KESTREL study randomized patients to brolucizumab 3 mg, brolucizumab 6 mg, or aflibercept 2 mg intravitreal therapy. KITE enrolled eyes to either brolucizumab 6 mg or aflibercept 2 mg. Brolucizumab groups received 5 loading doses every 6 weeks followed by 12-week dosing, with optional adjustment to every 8 weeks for active disease. Aflibercept groups received 5 doses every 4 weeks followed by fixed every 8 weeks dosing. At 52 weeks, brolucizumab 6 mg was found to be non-inferior to aflibercept in terms of BCVA (KESTREL: + 9.2 letters vs. + 10.5 letters; KITE: + 10.6 letters vs. + 9.4 letters; P < 0.001). Additionally, brolucizumab 6 mg demonstrated improved CSMT and fewer cases of persistent DME than aflibercept. With regards to ocular safety, adverse events remained more common in brolucizumab treatment groups compared to aflibercept in both studies. KESTREL reported IOI in 4.7 % of brolucizumab 3 mg, 3.7 % of brolucizumab 6 mg, and 0.5 % of aflibercept treatment arms. Retinal vasculitis was reported in KESTREL in 0.5 % of subjects in the brolucizumab 6 mg arm. There were 2 cases of retinal artery occlusion reported in KITE (1 case each in the brolucizumab 6 mg and aflibercept arms). No cases of retinal vasculitis were reported in the aflibercept arm in KESTREL or either of the treatment arms in KITE. Therefore, there was variability between the studies in the rates of ocular adverse events and, of note, both studies were powered for efficacy rather than safety endpoints.

Faricimab

Faricimab is a novel bispecific antibody targeting angiopoietin-2 and VEGF-A. Results of phase II and III trials have been published in recent years.

The BOULEVARD phase II trial compared the safety and efficacy of faricimab to ranibizumab in both treatment-naïve and previously treated DME patients. Treatment-naïve patients were randomized 1:1:1 to intravitreal 6.0 mg faricimab, 1.5 mg faricimab, or 0.3 mg ranibizumab. Patients treated previously were randomized 1:1–6.0 mg faricimab or 0.3 mg ranibizumab. Patients received injections monthly for 20 weeks and then observed for 16 additional weeks to assess durability. Amongst treatment-naïve patients, those receiving 6 mg faricimab dose demonstrated a statistically significant gain of 3.6 ETDRS letters over ranibizumab ( P = 0.03). Both therapies were associated with a reduction in CSMT and an improvement in the Diabetic Retinopathy Severity Scale Score. There were no statistically significant differences in adverse events between faricimab and ranibizumab.

YOSEMITE and RHINE were identical faricimab Phase III trials that enrolled patients with DME into one of three trial arms. The comparator was aflibercept 2.0 mg dosed every 8 weeks after 5 initial monthly doses. Patients receiving faricimab injections were dosed on either a fixed interval of 6.0 mg every 8 weeks after initial monthly treatment with 6 doses or a personalized treatment interval (PTI) dosing. PTI dosing involved faricimab 6.0 mg 4 doses at 4-week intervals followed by T&E up to every-16-week dosing. Patients were followed for 56 weeks. The primary endpoint was a change in BCVA from baseline at 1 year, averaged over weeks 48, 52, and 56. Both trials demonstrated non-inferior 1-year vision gains with faricimab every 8 weeks or PTI versus aflibercept every 8 weeks. The durability of faricimab was demonstrated as more than 70 % of patients in the PTI group achieved every 12-week dosing at 1 year. This potential to have a personalized treatment regimen suitable to individual needs would help reduce treatment burden and costs. Faricimab in real-life studies has also been shown to achieve more stable central retinal thickness and thus reduce the dynamic fluctuations known to cause an adverse effect.

Rates of adverse events were comparable between all groups with low rates of intraocular inflammation, most of which were mild or moderate severity. The ongoing long-term safety and tolerability study of Faricimab (Rhone-X Study) would shed light on these aspects in the future.

A recent network meta-analysis and systematic review studied the indirect comparison of faricimab T&E regime for DME. They found that faricimab T&E was associated with better retinal drying and needed fewer injections compared to other treatments in flexible-dose regimens. Faricimab also had better visual acuity outcomes compared to ranibizumab and bevacizumab.

Dexamethasone

While the use of dexamethasone sustained-release implant has been used alone or in combination with anti-VEGF for recalcitrant DME as demonstrated in trials such as the MEAD trial, no recent trials have been published.

Fluocinolone

The RESPOND trial was the nonrandomized phase IV pilot study that assessed the effectiveness and safety of 190 μg intravitreal Fluocinolone implant in 12 patients with DME for greater than 12 months refractory to prior treatment including anti-VEGF, steroids, or laser. The implant provides sustained, low-dose therapy for up to 36 months. At 12 months, patients with refractory DME demonstrated improved BCVA (mean +3.7 letters) and a statically significant reduction in mean CSMT (–292.83 μm, P = 0.003). Transient IOP elevations were noted but responded to topical therapy alone.

Razumab

Razumab is a cost-effective recombinant humanized IgG1 monoclonal antibody fragment that is the first biosimilar to ranibizumab. The drug is available in India at present. Recent studies have demonstrated comparable efficacy and safety between razumab and ranibizumab.

Pooled analyses for all patients from a prospective pilot study on Razumab injection in macular edemas of different causes demonstrated an interval improvement in BCVA and CSMT with a similar safety profile to ranibizumab. Chakraborty et al. performed a retrospective analysis comparing the use of razumab to ranibizumab in 333 eyes with DME. At 1-month post-injection, there were statistically significant improvements in BCVA and CSMT in both trial arms.

Risuteganib (ALG-1001)

Risuteganib is a synthetic peptide that regulates integrins that ultimately downregulate oxidative stress and angiogenesis. The stage 1 DEL MAR trial evaluated the safety and efficacy of risuteganib (ClinicalTrials.gov: NCT02348918).

138 participants were randomized into treatment with risuteganib monotherapy or bevacizumab monotherapy. Patients received either 1.0 mg, 2.0 mg, or 3.0 mg risuteganib injections dosed monthly at week 0, 4, and 8 with as-needed risuteganib injection at week 20. Eyes receiving bevacizumab received injections at weeks 0, 4, and 8, with as-needed therapy at 12, 16, or 20 weeks, for at least 3 but no more than 6 bevacizumab injections. The primary endpoint of non-inferiority in BCVA and CSMT improvement compared to bevacizumab was met. The second stage of this trial enrolled 80 patients into 5 treatment groups. Group 1 received five monthly injections of 1.25 mg of bevacizumab. Groups 2–5 received combinations of risuteganib and bevacizumab injections. Ultimately, the primary endpoint of non-inferiority of risuteganib about vision gain was met if given sequentially after bevacizumab, and the durability of these results was found up to 12 weeks after treatment. However, risuteganib administered in combination with bevacizumab was inferior to sequential therapy and to bevacizumab alone. There were no drug-related adverse events found.

Cataract surgery and DME

Cataracts often complicate DME management. Traditionally it is desirable to defer cataract surgery till the DME is stabilized. However, delaying cataracts is fraught with increased risks during the cataract surgery as it progresses to maturity. Additionally, the visual burden of a cloudy lens is undesirable to many patients and a balance in treatment with acceptable risk and benefit ratio often needs to be contemplated. Patients with no DR had nearly 37 % risk of DR post-cataract surgery in a large database study of 4485 diabetic patients. If cataract is affecting the vision, intra-operative anti-VEGF or dexamethasone implant can help reduce the risk of post-cataract surgery worsening of DME. Dexamethasone implant should be used with caution whenever the posterior capsular is torn or the zonules are weak to reduce the risk of migration into the anterior chamber.

Nanoparticles/nanotechnology

Nanoparticles (NP) have distinct advantages of smaller size and therefore easy passage through the blood-retinal barrier. Increased surface areas of action enable more precise and targeted management possibilities at a microstructural level. Various NP like Silica-based, Gold-based, silver-based, polylactic acid Glycolic acid-based NPs are being evaluated for their role in the management as well as monitoring of DR. Near future will probably see expansion of these treatment strategies as well.

Conclusions

Ophthalmologists need to understand the epidemiology of DR where they practice and the risk factors for DR. Current studies driven by biomarkers and gene polymorphisms have increased our understanding of DR pathogenesis. As new research continues to explore new targeted therapy, ophthalmologists should also keep updating clinical evidence for treating and preventing DR.

Role of surgery in diabetic retinopathy

The diabetic retinopathy vitrectomy study (DRVS) has demonstrated improved outcomes with vitrectomy performed within 1–6 months of the onset of vitreous hemorrhage compared to vitrectomy at 1 year. The American Academy of Ophthalmology (AAO) lays down the following criteria for a vitrectomy: non-clearing vitreous hemorrhage (VH); significant recurring VH, despite use of maximal PRP; dense pre-macular sub-hyaloid hemorrhage; tractional retinal detachment involving or threatening the macula; combined tractional and rhegmatogenous retinal detachment; red blood cell-induced glaucoma and “ghost cell” glaucoma; and anterior segment neovascularization with media opacities preventing PRP. For ease of understanding, we have classified cases needing surgical intervention as soon as possible and an early vitrectomy ( Table 6 ).

Table 6

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Ciliary sulcus characteristics in patients with axial myopia using ultrasound biomicroscope
2. Toward a greener vision: A review on advancing sustainability in ophthalmology
3. Changes in choroidal thickness in pre-myopic children after repeated low-level red-light therapy and their role in predicting myopia prevention and controlling myopic shift
4. From evidence to action: Public health approaches to reducing screen time and mitigating myopia risk

Stay updated, free articles. Join our Telegram channel

Join