Back to Journals » Clinical Ophthalmology » Volume 16

Real-World Evidence in the Management of Diabetic Macular Edema with Intravitreal Anti-VEGFs in Asia: A Systematic Literature Review

Authors Yuen YS , Tan GSW, Gan NY, Too IHK, Mothe RK, Basa P , Shaikh J

Received 23 June 2022

Accepted for publication 28 September 2022

Published 19 October 2022 Volume 2022:16 Pages 3503—3526

DOI https://doi.org/10.2147/OPTH.S378392

Checked for plagiarism Yes

Review by Single anonymous peer review

Peer reviewer comments 3

Editor who approved publication: Dr Scott Fraser



Yew Sen Yuen,1 Gavin Siew Wei Tan,2 Nicola Yi’An Gan,3 Issac Horng Khit Too,4 Raj Kumar Mothe,5 Pradeep Basa,5 Javed Shaikh5

1Department of Ophthalmology, National University Hospital, Singapore; 2Singapore Eye Research Institute, Singapore National Eye Centre, Ophthalmology and Visual Sciences Academic Clinical Program, Duke-NUS Medical School, Singapore, Singapore; 3Department of Ophthalmology, Tan Tock Seng Hospital, National Healthcare Group Eye Institute, Singapore, Singapore; 4Novartis Singapore Pte. Ltd., Mapletree Business City, Singapore; 5Novartis Healthcare Pvt. Ltd, Hyderabad, India

Correspondence: Issac Horng Khit Too, Novartis Singapore Pte Ltd, Mapletree Business City, 20 Pasir Panjang Road #10-25/28, 117439, Singapore, Tel +6567226189, Email [email protected]

Purpose: To evaluate the visual outcomes and safety profile of intravitreal anti-vascular endothelial growth factor (anti-VEGF) therapy in the treatment of diabetic macular edema (DME) in real-world studies in Asian countries.
Methods: A systematic review of electronic literature databases (Embase, Medline, and the Cochrane Library from January 1, 2010, to March 16, 2021) was conducted to identify observational studies that reported clinical and safety outcomes of anti-VEGF treatments for DME in Asia. We analyzed baseline patient characteristics, treatment patterns, mean number of injections, best-corrected visual acuity (BCVA), retinal thickness, and safety outcomes.
Results: Seventy-one studies were included in this review. Most studies reported treatment of DME with ranibizumab (n = 33), followed by aflibercept (n = 13), bevacizumab (n = 28), and conbercept (n = 9). At 12 months, the cumulative mean number of injections for ranibizumab, aflibercept, and conbercept was 5.2, 4.6, and 6, respectively. At the 12-month follow-up, the cumulative mean BCVA gain was 6.8 letters (ranibizumab), 4.6 letters (aflibercept), 4.9 letters (bevacizumab), and 8.3 letters (conbercept). The cumulative mean reduction in retinal thickness at 12 months was 116.9 μm (ranibizumab), 105.9 μm (aflibercept), 81.7 μm (bevacizumab), and 135.2 μm (conbercept). A strong positive correlation (r = 0.78) was observed between mean number of injections and change in BCVA at 12 months. A moderate positive correlation (r = 0.54) was observed between mean number of injections and mean reduction in retinal thickness at 12 months. A weak positive correlation was observed between baseline retinal thickness and visual acuity at 12 months. Baseline BCVA and mean number of injections were predictors of BCVA at 12 months.
Conclusion: All anti-VEGFs were effective in the treatment of DME in Asia. The data suggest that a greater number of anti-VEGF injections was associated with better improvement in BCVA and moderate reduction in retinal thickness at the 1-year follow-up.

Keywords: aflibercept, anti-vascular endothelial growth factors, bevacizumab, conbercept, DME, ranibizumab, retinal thickness, visual acuity

Introduction

Diabetic macular edema (DME), one of the most common causes of vision loss, manifests as retinal thickening caused by alteration of capillary permeability.1,2 Damage to the retinal microvasculature results in hypoxia, which stimulates the production of vascular endothelial growth factor (VEGF), as well as breakdown of the blood-retina barrier leading to edema.1,3

Epidemiological data suggest that approximately 7% of the diabetic patients may be at risk of developing DME and diabetic retinopathy.4,5 The maintenance of blood glucose levels, lipid levels, and systemic blood pressure is critical in preventing the development and progression of DME.6 Focal laser photocoagulation therapy was the gold standard for the treatment of patients with DME prior to the advent of intravitreal anti-VEGFs.7,8 In the Early Treatment Diabetic Retinopathy Study (ETDRS), the use of macular laser for clinically significant macular edema reduced the risk of progressive visual loss by approximately 50%.9 Pharmacological treatment of DME offers the opportunity to reduce central macular thickness (CMT) and results in improvement of visual acuity.10

Over the last decade, the introduction of therapeutic agents in the form of intravitreal anti-VEGFs and intravitreal steroids has significantly changed the treatment and prognosis of DME.11

Intravitreal anti-VEGFs are presently the first-line treatment option for patients with DME. Other treatment options include the aforementioned intravitreal steroids as well as non-steroidal anti-inflammatory drugs, inhibitors of multiple growth factors, and cytokine and chemokine inhibitors.7,12 Anti-VEGFs act on the VEGF receptors that mediate the breakdown of the blood-retinal barrier and reverse the vision impairment caused by macular edema13 and have demonstrated better efficacy compared with laser photocoagulation in several clinical studies.14–18

Real-world data on actual treatment patterns and outcomes in daily clinical practice for DME are limited and heterogeneous. The interpretation of this real-world data is essential for understanding how treatments work in clinical settings outside of well-controlled randomized controlled trials. Ethnic and geographical socioeconomic differences are known to affect the real-world efficacy of medical interventions. Thus, the objective of this systematic literature review (SLR) was to collate and report real-world evidence related to the clinical effectiveness and safety and treatment patterns of anti-VEGFs in DME patients in Asian countries.

Methods

Literature Search Strategy and Selection Criteria

This review was conducted by following the systematic principles of the Cochrane Handbook for Systematic Reviews of Interventions19 and was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA).20

Comprehensive literature searches were performed in Embase, Medline, and the Cochrane Library from January 1, 2010, to March 16, 2021, to identify relevant real-world studies. The search term strategy included both medical subject headings (MeSH) and free-text terms related to DME and anti-VEGFs. The detailed search strategy is presented in Table S1S3.

In addition, conference abstracts (Annual Meeting of Japanese Ophthalmological Society, International Congress of Ophthalmology and Optometry China, the Asia-Pacific Academy of Ophthalmology Congress, ISPOR Asia Pacific, Annual Meeting of The Korean Ophthalmological Society, Asia-Pacific Vitreo-retina Society, the Royal Australian and New Zealand College of Ophthalmologists, and American Academy of Ophthalmology) were hand searched from publication years 2018 to June 2021 to retrieve the latest unpublished studies in journals as well as full-text articles or supplement results of previously published studies.

Based on the predefined eligibility criteria (Table S4), titles and abstracts of all retrieved citations were screened for inclusion. Publications were included in the full-text review if they reported the clinical efficacy, safety, and treatment pattern of anti-VEGFs. Two independent reviewers were involved in the study selection process, and any discrepancies between them were reconciled by a third independent reviewer.

Data Collection and Synthesis

Study characteristics, baseline patient data, and outcomes data were collected. Data from the studies were extracted to a data extraction sheet by 1 reviewer, and quality check was performed by a second independent reviewer, with reconciliation of any discrepancies by a third independent reviewer.

The following outcomes were analyzed to support the objectives of this review: visual acuity (mean change from baseline and the proportion of patients gaining ≥10 letters or losing ≤10 letters), retinal thickness (change from baseline), injection frequency, presence or absence of retinal fluid, number of clinic visits, tolerability, switching, and safety of anti-VEGFs. Patients receiving monthly anti-VEGF injections for at least initial 3 consecutive months were classified into the loading group. Patients receiving <3 injections in the first 3 months were classified into the no loading group.

A qualitative descriptive analysis of these data, including cumulative means (mean of the means of several studies), was used to analyze the data with respect to baseline characteristics and respective outcomes. Mean and cumulative mean values are reported as mean±standard deviation (SD). A correlation analysis between injection frequency and BCVA/retinal thickness outcomes at 12 months was performed using the Pearson correlation (r). The association of baseline predictors (BCVA and central retinal thickness [CRT]) and mean number of injections at 12 months with change in the BCVA letter score at 12 months was analyzed using the multiple linear regression method.

For the purpose of analysis, the logarithm of the minimum angle of resolution (LogMAR) values were converted to ETDRS letters per the equation “ETDRS letters = 85–50LogMAR” based on the publication by Gregori et al.21 Subgroup analysis included eyes stratified by baseline BCVA of <69 (worse vision) and ≥69 letters (better vision) based on the Diabetic Retinopathy Clinical Research Network (DRCR.net) protocol.22 An additional subgroup analysis was performed to compare visual outcomes according to patient history of treatments with anti-VEGFs (naive vs pretreated/mixed treatment patients). Of the studies investigating the efficacy of anti-VEGFs, those with a follow-up of ≥12 months were included in the efficacy analysis.

Results

The literature search yielded 2775 citations, of which 168 were removed as duplicates. Screening of the titles and abstracts of all retrieved citations resulted in 366 potentially relevant references for inclusion. Of these, 75 publications (including 10 conference abstracts) were included in the present review following a detailed examination against the predefined eligibility criteria. Of these 75 publications, 71 were unique studies and 4 were linked to primary studies (Figure S1).

Study Characteristics

The study characteristics including study design, follow-up period, country, intervention, sample size, and outcomes are presented in Table S5. Of the 71 studies included, 65 were retrospective studies and 6 were prospective studies. Sixty-one studies reported data for a single treatment arm, 8 studies reported data for 2 treatment arms, and 2 studies reported data for 3 treatment arms. The studies were mostly conducted in Japan (18), followed by South Korea (15), Taiwan (14), China (12), India (4), Thailand (3), Australia (1), Malaysia (1), New Zealand (1), Pakistan (1), and Singapore (1). The time point of interest for the clinical outcomes (BCVA and retinal thickness) was 12 months. The mean follow-up period in these studies ranged from 0.523 to 6024 months.

Clinical Characteristics of the Study Population at Baseline

The mean (±SD) age of the patients ranged from 52.6 (±14)25 to 73.9 (±10)23 years. The size of the study population varied widely, ranging from 1426 to 204927 eyes. At baseline, the duration of DME ranged between 2.628 and 27.229 months, with the mean±SD duration in all studies being 14.85±8.43 months. The mean baseline BCVA ranged from 3930 to 7231 ETDRS letters. The mean baseline CRT ranged from 300.832 to 575.933 μm.

A total of 27 studies did not report details of naive status (whether patients received prior anti-VEGF therapy). Of the 44 studies that reported details of previous treatment, 31 included only anti–VEGF-naive patients, 3 included only patients who were previously treated with anti-VEGFs, and 11 included both naive and previously treated patients (mixed treatment patients).

Treatment Pattern, Regimen, and Mean Number of Injections

Treatment Pattern

Overall, 33 studies evaluated the use of ranibizumab as a treatment option for DME, followed by 13 on aflibercept, 28 on bevacizumab, and 9 on conbercept (Figure S2A).

In Japan, the use of ranibizumab and aflibercept was common for DME. In Taiwan, ranibizumab was the most prescribed anti-VEGF medication for DME. Off-label bevacizumab was used predominantly in South Korea, whereas conbercept was used only in China (Figure S2B).

Treatment Regimen

Of the 71 studies, 28 reported the practice of administering anti-VEGF loading doses, ie, ≥3 monthly injections. In 13 studies, no information was available on the type of regimen used for anti-VEGFs. The remaining 30 studies reported the use of either a single injection or a single injection followed by an as-needed approach. Of the 28 studies reporting the use of loading doses, only 1 study34 reported the use of 5 monthly loading doses for aflibercept and ranibizumab, whereas the remaining 27 studies reported 3 monthly loading doses (3q4w) of anti-VEGF injections. Of the 30 studies with no loading dose, 17 included only treatment-naive patients, 2 included only previously treated patients, and 4 included both naive and previously treated patients (Table 1). Loading doses were reported in 65% of ranibizumab studies, 50% of aflibercept studies, 37.5% of conbercept studies, and 33.3% of bevacizumab studies (Figure S3). Details of the treatment regimen with respect to loading and no loading doses are provided in Table 1.

Table 1 Number of Studies with Loading and No Loading Doses with Respect to Treatment Regimen

Mean Number of Injections

Twenty studies reported injection frequency data at 12 months. The mean number of injections in 12 months ranged from 2.935 to 7.2,36 with the cumulative mean being 5.2. The cumulative mean number of injections was higher in studies with naive patients (8 studies, cumulative mean: 5.97; mean number of injections ranged from 3.937 to 7.236) than in studies with mixed treatment patients (9 studies, cumulative mean: 4.67; mean number of injections ranged from 3.838 to 6.439). Two studies did not report data on prior treatment status (cumulative mean: 4.17; mean number of injections ranged from 2.935 to 540). The mean number of injections data at 12 months is not available for studies that included patients who were previously treated with anti-VEGFs (Table 2).

Table 2 Number of Studies with Mean Number of Injections Data at 12 Months Based on Prior Treatment Status

For ranibizumab, 11 studies reported injection frequency at the 12-month follow-up (range: 3.937 to 7.236). The cumulative mean number of injections was 5.2 (11 studies with 12 treatment arms). The mean number of injections for ranibizumab 1 dose at baseline (RAN 1q4w) followed by an as-needed approach (pro re nata [PRN]) ranged from 3.937 to 4.4,29 and the mean number of injections for ranibizumab 3q4w (RAN 3q4w) followed by PRN regimen ranged from 4.341 to 7.2.36 Only 1 study reported the mean number of injections for RAN 3q4w followed by treat and extend (T&E) regimen (7.1)42 at the 12-month follow-up (Figure S4).

For aflibercept, 4 studies reported injection frequency at the 12-month follow-up (range: 2.935 to 6.542). The cumulative mean number of injections was 4.6 (4 studies with 5 treatment arms). One study each reported mean number of injections for aflibercept 1 dose at baseline (AFL 1q4w) followed by PRN regimen (2.9),35 aflibercept 3 monthly doses (AFL 3q4w) followed by T&E regimen (6.5),42 and AFL 3q4w followed by PRN regimen (4.6)35 (Figure S5).

Two studies reported injection frequency at the 12-month follow-up for bevacizumab. Sepehr et al reported a mean of 5 injections (treatment regimen not specified),40 and Choovuthayakorn et al reported the use of BEV 3q4w followed by PRN regimen with a median of 6 injections.43

Five studies reported injection frequency at the 12-month follow-up for conbercept (range: 4.544 to 6.845). The cumulative mean number of injections was 6 (5 studies with 5 treatment arms). Two studies each reported the mean number of injections for conbercept 1 dose at baseline (1q4w) followed by PRN regimen (5.646 and 6.845) and conbercept 3q4w followed by PRN regimen (6.636 and 6.647) (Figure S6).

Efficacy Outcomes

The efficacy outcomes are summarized in Table S6. On an average, BCVA and retinal thickness parameters were reported for up to 12 months in 25 studies, while 8 studies reported data beyond the 12-month follow-up.

Visual Acuity Outcomes

Mean Change in BCVA from Baseline (ETDRS Letters) at 12 Months

Overall, 42 studies reported change in mean BCVA from baseline at various time points (Figure S7). Mean BCVA values were reported in 41 studies at baseline. In the 22 studies reporting data at 12 months, the mean BCVA ETDRS letters ranged from 4648 to 7231 at baseline and 51.441 to 74.237 at 12 months. The cumulative mean±SD gain in BCVA ETDRS letters with anti-VEGFs was 6.5±2.9 (cumulative mean±SD injections: 5.1±1.2). The cumulative mean gain with conbercept (8.3 ETDRS letters)36,44–47 and ranibizumab (6.8 ETDRS letters)29,31,36,37,39,41,48–52 showed greater BCVA improvement than that with aflibercept (4.6 ETDRS letters)35,38,53 and bevacizumab (4.9 ETDRS letters).40,43,54 In the subgroup analysis stratified by prior treatment status, studies with naive patients reported higher BCVA letter gain (8.5; mean number of injections: 6.0) than those with mixed treatment patients (5.8; mean number of injections: 4.6) (Table 3).

Table 3 Studies with Visual Activity Outcomes at 12 Months

Two studies reported subgroup data for patients with worse baseline BCVA (<69 ETDRS letters or 20/50 or worse Snellen equivalent) and better baseline BCVA (≥69 ETDRS letters or 20/32–20/40 Snellen equivalent) at the 12-month follow-up. Choovuthayakorn et al included naive patients and reported a significant BCVA letter gain with bevacizumab in the worse baseline BCVA group (mean gain: 8.4; 95% confidence interval [CI] 6.3 to 10.6; median injections: 6; p < 0.001) than in the better baseline BCVA group (mean BCVA gain: 2.0 letters; 95% CI −1.6 to 5.6; median injections: 6).43 Similarly, Li et al reported better BCVA gain with conbercept in the worse baseline BCVA group (median BCVA gain: 18 letters; mean injections: 6.7) than in the better baseline BCVA group (median BCVA gain: 7 letters; mean injections: 6.5)55 in patients with unclear prior treatment status.

Ranibizumab

Eleven studies reported change in BCVA from baseline. The mean gain in ETDRS letters ranged from 2.4 letters (baseline: 72; at 12 months: 74.4)31 to 14.8 letters (baseline: 51.3; at 12 months: 66.1)31 The highest letter gain (14.8 ETDRS letters) was observed in the RAN 3q4w followed by PRN regimen group with 6.5 mean injections in naive DME patients.49 In the subgroup analysis stratified by prior treatment status, the mean±SD change in the BCVA letter score from baseline to 1 year was 8.3±4.9 in studies with naive patients36,37,49 and 6.1±1.9 in studies with mixed treatment patients29,31,39,41,48,50–52 (Figure S8).

Aflibercept

Three studies reported change in BCVA from baseline. The mean gain in ETDRS letters ranged from 2 letters (baseline: 64; at 12 months: 66)35 to 7 letters (baseline: 62.5; at 12 months: 69.5).35 AFL 3q4w followed by PRN regimen with 4.6 mean injections reported the highest letter gain (7 ETDRS letters)35 (Figure S9). At 12 months, data for naive patients were not available, and only 1 study reported data for mixed treatment patients (mean BCVA gain: 4.5; mean injections: 3.8).38

Bevacizumab

Three studies reported change in BCVA from baseline. Mean gain in ETDRS letters ranged from 4 letters (baseline: 66; at 12 months: 70)40,54 to 6.8 letters (baseline: 50.2; at 12 months: 57).43 At 12 months, only 1 study reported data for naive patients (mean BCVA gain: 6.8; mean injections: data not available),43 and data were not available for mixed treatment patients.

Conbercept

Five studies reported change in BCVA from baseline. The mean gain in ETDRS letters ranged from 4.5 letters (baseline: 49.5; at 12 months: 54)44 to 10 letters (baseline: 57.1; at 12 months: 67.1).45 Conbercept 1q4w followed by PRN regimen with 6.8 mean injections reported the highest letter gain (10 ETDRS letters).45 In the subgroup analysis stratified by naive status, the cumulative mean±SD gain in the BCVA letter score from baseline to 1 year was 9.3±0.5 (cumulative mean±SD injections: 6.4±0.5) in studies with naive patients36,45–47,56 and 4.5 (mean injections: 4.5) in those with mixed treatment patients44 (Figure S10).

Proportion of Patients with Change in ETDRS Letters (Gain and Loss) at 12 Months

Six studies reported proportion of patients with ≥10 ETDRS letter gain.29,36,43,46,49,53 Four studies included naive patients, 1 study29 included mixed treatment patients, and prior treatment status was unclear in 1 study.53 Among naive patients, the proportion of patients who gained ≥10 ETDRS letters was high (65.9%)49 in patients treated with RAN 3q4w followed by PRN regimen (mean injections: 6.5) compared with those treated with bevacizumab 3q4w followed by PRN regimen (43.1%; median injections: 6)43 and conbercept 3q4w followed by PRN regimen (41.7%; mean injections: 6.6).36 For no loading dose regimens, the proportions of patients who gained ≥10 ETDRS letters were 45.2% and 40% for conbercept 1q4w followed by PRN (mean injections: 5.6)46 and RAN 1q4w followed by PRN regimen (mean injections: 4.4),29 respectively. In a study with aflibercept (mean injections: 5.2), 25% of patients gained ≥10 ETDRS letters and regimen details were unclear36 (Figure S11).

Only 3 studies reported data on the proportion of patients who gained ≥15 ETDRS letters. All these studies included patients naive to anti-VEGFs. At 12 months, the proportion of patients who gained ≥15 ETDRS letters was high (35.8%)43 in naive patients treated with bevacizumab 3q4w followed by PRN regimen (median injections: 6) compared with those treated with conbercept 3q4w followed by PRN regimen (19.4%; mean injections: 6.6)36 and RAN 3q4w followed by PRN regimen (15.6%; mean injections: 7.2)36 (Figure S12).

Limited evidence was available for the proportion of patients with letters lost after treatment. Naive patients treated with bevacizumab 3q4w followed by PRN regimen (median number of injections: 6) reported loss of ≥10 letters (13.8%) and loss of ≥15 letters (8.1%) at 12 months.43 In a study conducted in Taiwan, treatment-naive patients were assessed based on before (a maximum of 5 ranibizumab injections were reimbursed for each eye during the first year) and after the major reimbursement policy change (a maximum of 8 ranibizumab injections were reimbursed for each eye during the same period). The proportion of patients with loss of ≥10 letters was 15.9% before the policy change (mean injections: 4.6) and 4.8% after the policy change (mean injections: 6.5). This study used the RAN 3q4w followed by PRN regimen.49 None of the studies with aflibercept reported data for gain/loss of ≥15 letters at 12 months.

Correlation Between Injection Frequency and BCVA at 12 Months

Overall, 17 studies reported data on the mean number of injections and BCVA gain. The mean number of ETDRS letter gain showed a positive linear correlation with the mean number of anti-VEGF injections received (Pearson correlation coefficient, r = 0.78)29,31,35–41,44–50,53 (Figures 1 and 2).

Figure 1 Correlation of BCVA gain with mean injections at 12 months for overall patients.

Abbreviations: ETDRS, Early Treatment Diabetic Retinopathy Study; BCVA, best corrected visual acuity.

Figure 2 Mean injections and BCVA improvement at 12 months.

Abbreviations: AFL, aflibercept; CON, conbercept; ETDRS, Early Treatment Diabetic Retinopathy Study; RAN, ranibizumab; PRN, pro re nata; q4w, monthly dose; BCVA, best corrected visual acuity.

In the subgroup analysis stratified by prior treatment status, the correlation analysis showed a strong association between mean number of injections and mean letter gain in the naive patients (Pearson correlation coefficient, r = 0.73)36,37,45–47,49 (Figure S13). However, a moderate association was observed in mixed treatment patients (Pearson correlation coefficient, r = 0.47)29,31,38,39,41,44,48,50 (Figure S14).

Seven studies with the 3q4w followed by PRN regimen reported data for mean number of injections and BCVA. Limited evidence was available to establish a correlation between mean number of injections and BCVA gain at 12 months for conbercept36,47 and aflibercept.35 A positive linear correlation was observed between mean number of injections and BCVA gain for RAN 3q4w followed by PRN regimen (Pearson correlation coefficient, r = 0.59)36,39,41,48,49 (Figure S15).

Correlation Between Baseline BCVA and Mean Change in BCVA at 12 Months

A correlation analysis in overall patients irrespective of prior treatment status showed that higher the baseline BCVA score lesser the BCVA gain at 12 months (Pearson correlation coefficient, r=−0.42).29,31,35–41,43–54 In the subgroup analysis by prior treatment status, only the treatment-naive group corroborated this correlation with the Pearson correlation coefficient of r=−0.44.36,37,43,45–47,49 A weak correlation (Pearson correlation coefficient, r=−0.24)29,31,38,39,41,44,48,50–52 was observed between baseline BCVA scores and mean change from baseline to 1 year in the BCVA letter score in mixed treatment patients (Figure 3).

Figure 3 Correlation between baseline BCVA values and BCVA gain (ETDRS letters) at 12 months for (A) overall patients, (B) naive patients and (C) mixed treatment patients.

Abbreviations: ETDRS, Early Treatment Diabetic Retinopathy Study; BCVA, best corrected visual acuity.

Retinal Thickness Outcomes

Most studies used similar definitions for retinal thickness, although a range of terminologies were used, including CRT, central subfield thickness, CMT, and central foveal thickness.

Mean Change in Retinal Thickness from Baseline at 12 Months

Twenty studies reported change in mean retinal thickness at 12 months. Baseline retinal thickness ranged from 38152 to 538 µm,35 and thickness at 12 months ranged from 27637 to 414 µm.35 The cumulative mean±SD reduction in retinal thickness with anti-VEGFs was 115.4±31.8 µm (cumulative mean±SD injections: 5.1±1.2). The cumulative mean±SD reduction in CRT values was 116.9±30.7 μm with ranibizumab (mean injections: 5), 105.9±20.8 µm with aflibercept (mean injections: 4.1), 81.7±70.2 μm with bevacizumab, and 135.2±18.2 μm with conbercept (mean injections: 6) (Table 4).

Table 4 Change in Retinal Thickness Outcomes at 12 Months

Ranibizumab

Ten studies reported the change in mean retinal thickness. The mean reduction in retinal thickness ranged from 6229 to 149 µm.51 RAN 3q4w followed by PRN regimen reported high reduction in mean retinal thickness (149 µm) in patients with a high baseline retinal thickness value (503 µm)51 (Figure S16).

Aflibercept

Four studies reported the change in mean retinal thickness. The mean reduction in retinal thickness in patients treated with aflibercept ranged from 86.135 to 133.8 µm.35 AFL 3q4w followed by PRN regimen with 4.6 mean injections reported the highest reduction in mean retinal thickness (133.8 µm)35 (Figure S17).

Bevacizumab

Only 1 study reported thickness reduction data for bevacizumab (regimen details were unclear). The mean reduction in retinal thickness at the 12-month follow-up was 32 µm.40

Conbercept

Five studies reported the change in mean retinal thickness. The mean reduction in retinal thickness ranged from 10344 to 146.1 µm.44 Conbercept 3q4w followed by PRN regimen with 6.6 mean injections reported a high reduction (146.1 µm) in mean retinal thickness in patients with a high baseline retinal thickness (482 µm)44 (Figure S18).

In the subgroup analysis stratified by prior treatment status, the cumulative mean±SD reduction in retinal thickness with all anti-VEGFs was 141.6±5.2 μm (cumulative mean±SD injections: 6.1±1.2) in naive patients and 108.8±27.5 μm (cumulative mean±SD injections: 4.6±0.8) in mixed treatment patients. Eight studies with the 3q4w followed by PRN regimen reported data for mean number of injections and retinal thickness outcomes at 12 months. Naive patients treated with conbercept (mean injections: 6.6) and ranibizumab (mean injections: 7.2) showed a better reduction in retinal thickness of 146.1 and 145.2 µm, respectively, at the 12-month follow-up (Figure S19). Only 1 study each reported data for AFL 3q4w followed by PRN regimen (patients’ prior treatment status unclear) and bevacizumab 3q4w followed by PRN regimen (naive patients), with a retinal thickness reduction of 133.8 µm (mean injections: 4.6)35 and 131.3 µm (median injections: 6)43 at 12 months, respectively.

Correlation Between Injection Frequency and Reduction in Retinal Thickness at 12 Months

Overall, 16 studies reported data on mean number of injections and reduction in retinal thickness from baseline at 12 months. A positive correlation was observed between mean reduction in CRT and mean number of injections, irrespective of anti-VEGF received (Pearson correlation coefficient, r = 0.54)29,31,35–41,44–48,50,53 (Figure 4).

Figure 4 Correlation between reduction in retinal thickness and mean injections at 12 months for overall patients.

In the subgroup analysis stratified by prior treatment status, 5 studies reported correlation data between mean number of injections and retinal thickness outcome at the 12-month follow-up in patients naive to anti-VEGFs. A weak positive correlation was observed between mean number of injections and reduction in retinal thickness in patients naive to anti-VEGFs (Pearson correlation coefficient, r = 0.18)36,37,45–47 (Figure S20).

Eight studies reported correlation data between mean number of injections and retinal thickness outcome at the 12-month follow-up in patients with mixed treatment status to anti-VEGFs. A positive correlation was observed between mean number of injections and reduction in retinal thickness in patients with mixed treatment status to anti-VEGFs (Pearson correlation coefficient, r = 0.60)29,31,38,39,41,44,48,50 (Figure S21).

Correlation Between Baseline CRT with Baseline BCVA

While higher baseline CRT was associated with poor baseline BCVA in naive patients (Pearson correlation coefficient, r=−0.83),36,37,43,45–47 it was associated with better baseline BCVA in mixed treatment patients (r = 0.67)29,31,38,39,41,44,48,50–52 (Figure 5).

Figure 5 Correlation between baseline CRT and baseline BCVA for (A) naive patients and (B) mixed treatment patients.

Correlation Between Baseline CRT and BCVA Outcomes at 12 Months

Overall, a weak positive correlation was observed between baseline retinal thickness and BCVA letter score at the 12-month follow-up (Pearson correlation coefficient, r=0.35)29,31,35–41,43–48,50–53 (Figure 6).

Figure 6 Correlation between baseline CRT and BCVA letter score at 12 months for (A) overall patients, (B) naive patients and (C) mixed treatment patients.

Abbreviations: CRT, central retinal thickness; ETDRS, Early Treatment Diabetic Retinopathy Study; VA, visual acuity.

In the subgroup analysis stratified by prior treatment status, a strong negative correlation was observed between baseline retinal thickness and BCVA letter score at the 12-month follow-up in patients naive to anti-VEGFs (Pearson correlation coefficient, r=−0.81).36,37,43,45–47 However, contradictory positive results were observed between baseline thickness and BCVA letter score at 12 months in the mixed treatment patients (Pearson correlation coefficient, r = 0.62)29,31,38,39,41,44,48,50–52 (Figure 6).

In the overall group, no correlation was observed between baseline retinal thickness and mean change from baseline to 1 year in the BCVA letter score. However, in the subgroup analysis stratified by prior treatment status, strong positive (Pearson correlation coefficient, r=0.61)36,37,43,45–47and weak negative (Pearson correlation coefficient, r=−0.38)29,31,38,39,41,44,48,50–52 correlations were observed between baseline retinal thickness and BCVA letter gain at the 12-month follow-up in naive patients and mixed treatment patients, respectively (Figure 7).

Figure 7 Correlation between baseline CRT and BCVA gain at 12 months for (A) overall patients, (B) naive patients and (C) mixed treatment patients.

Abbreviations: CRT, central retinal thickness; ETDRS, Early Treatment Diabetic Retinopathy Study; VA, visual acuity.

Correlation Between Change in CRT and Change in BCVA at 12 Months

A weak positive correlation was observed between change in CRT and change in BCVA. At 12 months, a higher reduction in CRT was associated with a higher BCVA gain. However, this association was weak in all the groups, patients naive to anti-VEGFs (r = 0.29),36,37,43,45–47 mixed treatment patients (r = 0.11),29,31,38,39,41,44,48,50–52 and overall patients (r = 0.50)29,31,35–41,43–48,50–53 (Figure 8).

Figure 8 Correlation between reduction in retinal thickness and BCVA gain at 12 months for (A) overall patients, (B) naive patients and (C) mixed treatment patients.

Abbreviations: ETDRS, Early Treatment Diabetic Retinopathy Study; VA, visual acuity.

Correlation Between CRT and BCVA at 12 Months

Higher CRT was associated with poor BCVA in patients naive to anti-VEGFs (r=−0.81)36,37,43,45–47 at 12 months (Figure 9). However, in the mixed treatment patients (r = 0.35)29,31,38,39,41,44,48,50–52 and in the overall patients (r = 0.26),29,31,35–41,43–48,50–53 a higher CRT was associated with better BCVA at 12 months.

Figure 9 Correlation between CRT and BCVA letter score at 12 months for (A) overall patients, (B) naive patients and (C) mixed treatment patients.

Abbreviations: ETDRS, Early Treatment Diabetic Retinopathy Study; VA, visual acuity.

Regression Analysis

The multivariate linear model for predicting the BCVA gain included baseline BCVA data, baseline CRT data, and mean number of injections. The coefficient of determination R squared (R2) and adjusted R2 values for this model were 0.786 and 0.74. In this model, baseline BCVA and mean number of injections were significantly associated with final BCVA ETDRS letter gain (p = 0.02 and 0.0001, respectively). However, the relationship between baseline CRT and BCVA gain was not significant (p = 0.184) (Table 5).

Table 5 Baseline Factors (BCVA and CRT) and Mean Number of Injections for Mean Change of ETDRS Letters by Multiple Regression Models

Retinal Fluid Outcomes

Eight studies reported retinal fluid outcomes at baseline, whereas only 1 study reported fluid outcome data at the 12-month follow-up. Patients treated with bevacizumab reported a significant reduction in total intraretinal fluid (mean±SD at baseline: 0.25±0.24 mm3; at 12 months: 0.10±0.14 mm3; p = 0.001) and subretinal fluid (mean±SD at baseline: 0.04±0.18 mm3; at 12 months: 0.002±0.01 mm3; p = 0.048) within 3-mm area at 12 months.54

Clinical Visits

Only 1 study reported the number of ophthalmologic clinic visits. In this study, patients were receiving bevacizumab injection. The median number of clinic visits was higher in the first year (10; interquartile range [IQR]: 8–12) compared with that in the second (7; IQR: 4–9) and third year (6; IQR: 3–9).43

Tolerability and Switching

Treatment discontinuations were not commonly reported in real-world studies. Only 2 studies reported discontinuation data. Of 234 patients from the Kelkar study, 127/234 patients were lost to follow-up at 12 months. Of these 127 patients, 64 were from the ranibizumab group and 63 were from the bevacizumab group.57 In another study, of the 55 patients treated with ranibizumab, 6 (4 lost to follow-up and 2 withdrew consent) discontinued from the study before 12 months.52

Safety

Of the 21 studies reporting safety data, only 6 reported adverse events (AEs) (Table 6). None of the studies reported any serious ocular AEs. One study reported data for conjunctival hemorrhages, which were higher in the conbercept-treated group (15.6%, 5/32 patients) compared with the ranibizumab-treated group (10%, 3/30 patients) during 12 month follow-up.36 Two studies reported a rise in intraocular pressure (IOP) with bevacizumab injections,58,59 while 1 study reported a rise in IOP with the ranibizumab36 and conbercept36 regimens. One study reported injection-related endophthalmitis with bevacizumab (0.11%) and ranibizumab (0.42%).24 None of the included studies reported major systemic AEs with intravitreal anti-VEGF injections.

Table 6 Summary of Adverse Events with Anti-VEGF Agents

Discussion

In the current SLR, clinical evidence from real-world studies on visual outcomes and anatomical outcomes with intravitreal anti-VEGF injections were analyzed in the management of DME.

Overall, the studies were comparable in terms of baseline characteristics such as age, gender, and baseline glycated hemoglobin levels. However, the studies varied in prior treatment status/type of prior therapies, time point of assessment, number of injections, treatment regimen, and baseline BCVA. None of the studies in this SLR included patients with good BCVA (20/25 or better on the Snellen chart) at baseline, which contrasts with DRCR protocol V60 (DRCR protocol V includes patients with good BCVA, ie, 20/25 or better on the Snellen chart). This suggests that, in routine clinical practice, most physicians are not treating patients with good BCVA at baseline.

Use of prior anti-VEGFs at baseline is an important parameter that was not consistently reported across the included studies. In this review, most studies included a mixed treatment patients (both naive patients and previously treated patients). However, in randomized controlled trials, patients were mostly either treatment naive or had undergone a wash-out period before commencing DME treatment.61

A large variability was observed across studies in the mean number of anti-VEGF injections received at 12 months. In this review, the mean number of anti-VEGF injections ranged from 2.935 to 7.2,36 with a cumulative mean of 5.2. In contrast, the DRCR protocol T study reported the median number of injections as 9 or 10.62 In the current review, the mean number of injections in the first 12 months was less compared with that in DRCR protocol T, where 5–6 loading doses were given followed by continued regular treatment until stability. In this review of Asian studies, the cumulative mean number of injections of ranibizumab during first 12 months was 5.2, this is similar to other real-world studies conducted in non-Asian countries.63–65 The cumulative mean number of injections of aflibercept during first 12 months in Asian studies was 4.6, which is less compared to the mean number of injections reported from other real-world studies.66–69

In this review, <50% of the real-world studies reported an initial loading dose of 3 monthly doses, which was followed by a variation of an as-needed (PRN) or other regimens. This is in contrast to randomized controlled trials.

In this SLR, the mean improvement in the BCVA score, with at least a 7-letter gain, was greater with conbercept (80%, 4/5 studies), followed by ranibizumab (50%, 6/12 studies) and aflibercept (25%, 1/4 studies), at 1 year. At least 40% of patients who received ranibizumab,29,36,49 conbercept36,46 and bevacizumab43 injections reported a gain of ≥10 ETDRS letters compared with 25% of patients who received aflibercept.53 The improvement in BCVA varied with the baseline BCVA and mean number of injections received. In addition, treatment-naive patients were more likely to report greater improvements in BCVA over 12 months compared with patients previously treated with anti-VEGFs.

The mean improvement from baseline in BCVA gain was 6.9 ETDRS letters after 1 year of treatment in patients with DME. In this SLR, conbercept (8.3±2.2) and ranibizumab (6.8±3.2) treatment reported better cumulative mean±SD BCVA gain than aflibercept (4.6±2.1) and bevacizumab (4.9±1.6) at the 12-month follow-up. In contrast, the DRCR protocol T study reported better vision gain in change from baseline in the letter score (mean±SD) with aflibercept (13.3±11.1), followed by ranibizumab (11.2±9.4) and bevacizumab (9.7±10.1), at the 12-month follow-up.62 However, given the variation in baseline BCVA, frequency of injections, treatment-naive status, and other limitations in the analysis of real-world data, no definitive conclusions can be made regarding the relative efficacy of anti-VEGF agents. The other limitation of this review is, we did not use weighted means, weighing the outcome according to number of observations provides robust results. It was not possible to calculate weighted means with the evidence reported in these real-world studies. In addition, the outcomes in naive patients are not comparable to pre-treated patients and/patients with mixed treatment status. Hence, we have provided data for each subgroup to reflect the results for these patient subgroups. However, in some studies, the detail of previous treatment is unclear/not reported and hence we have also reported data for overall studies irrespective of their prior treatment status.

In the subgroup analysis stratified by prior treatment status, ranibizumab showed similar BCVA letter gain (mean±SD) in naive patients (8.3±4.9) in the real-world setting36,37,49 compared with that in the DRCR protocol T study (7±8.1) at the 12-month follow-up.62

In this review, at the 12-month follow-up, all studies demonstrated a mean reduction in CRT. Reduction in CRT was also observed with anti-VEGF treatment in the DRCR protocol T study.62 In this SLR, at the 12-month follow-up, the mean±SD reductions in CRT on optical coherence tomography were 135.2±18.2 μm with conbercept, 116.9±30.7 μm with ranibizumab, 105.9±20.8 with aflibercept, and 81.7±70.2 with bevacizumab.

Notably, in Taiwan, a health care policy change allowing reimbursement of more ranibizumab injections (before policy change: up to 5 vs after policy change: up to 8) was associated with improved BCVA and retinal thickness outcomes,49 thereby supporting the observation that a higher improvement in the mean BCVA score was associated with increased injection frequency over 12 months.

At 12 months, the correlation between the BCVA letter scores and retinal thickness in DME patients treated with anti-VEGF agents reported contrasting results compared with previous findings.70 Higher retinal thickness at 12 months was associated with better BCVA in overall (r = 0.26) and mixed treatment patients (r = 0.35) with DME. This weak correlation implies that improved BCVA is likely multifactorial and does not depend on the thickness of the retinal layers alone; moreover, from a clinical perspective, the correlation between retinal thickness and BCVA may be confounded by patients with ischemic maculopathy, whereby the thin atrophic macula is accompanied by poor vision.

The burden of health care visits in DME patients is high,71 although the DRCR.net studies have demonstrated a reduction in the frequency of injections after the first year with intensive treatment.62,72 PRN approach refers to monthly visit/control. However, some studies followed-up the patients monthly, and some studies employed a specific reinjection/revisit criteria based on predefined disease stability criteria as defined by the physician and some studies did not report this information. The studies included in this review did not clearly specify the evidence on clinic visits. Only 1 study reported median clinic visits for bevacizumab for the first, second, and third year, with a decrease in the number of clinic visits of 10, 7, and 6, respectively.43

All 4 anti-VEGFs (aflibercept, bevacizumab, conbercept, and ranibizumab) had comparable ocular and non-ocular safety profiles. In this review, the incidence of rise in IOP ranged from 1% to 5.6% with anti-VEGFs. In a similar type of real-world evidence literature review, the incidence of rise in IOP ranged between 10% and 20%.73

Conclusions

This SLR of real-world observational studies from Asia found that ranibizumab was the most frequently used anti-VEGF for DME compared with other anti-VEGFs. Off-label bevacizumab was also frequently used for DME treatment in several Asian countries; however, conbercept was used only in China. This might be due to the longest availability of ranibizumab, and bevacizumab compared to other anti-VEGFs in Asia. Treatment regimens varied across the studies. The loading dose regimen was the most frequently followed approach for ranibizumab, whereas for the other anti-VEGFs, no loading dose regimen was described. A loading dose followed by an as-needed (PRN) approach was more commonly used than a T&E approach. The mean number of anti-VEGF injections was lower compared with that in the randomized controlled trials at the 12-month follow-up, which may account for the lower improvement in mean BCVA. A greater letter gain was observed in naive patients compared with that in mixed treatment patients (naive and previously treated group). The correlation analysis suggested that a higher mean number of injections was associated with better BCVA gain for all the 3 anti-VEGFs (ranibizumab, aflibercept, and conbercept). This correlation was stronger in naive patients than in mixed treatment patients. Similarly, a higher mean number of injections was associated with a greater reduction in CRT for all the 3 anti-VEGFs (ranibizumab, aflibercept, and conbercept). In the multivariate regression analysis, baseline BCVA and mean number of injections were significantly associated with final BCVA ETDRS letter gain at 12 months.

Overall, a weak positive correlation was observed between baseline retinal thickness and BCVA letter score at the 12-month follow-up. There was a paucity of data for retinal fluid, clinical visits, and treatment switching outcomes across the included studies. Overall, no conclusion regarding the comparative efficacy of the anti-VEGFs could be drawn owing to heterogeneous study population, treatment posology, and frequency of injections, with few studies reporting head-to-head data.

Abbreviations

1q4w, 1 dose at baseline; 3q4w, 3 monthly loading doses; AE, adverse event; AFL, aflibercept; BCVA, best-corrected visual acuity; BEV, bevacizumab; CFB, change from baseline; CI, confidence interval; CMT, central macular thickness; CON, conbercept; CRT, central retinal thickness; DME, diabetic macular edema; DRCR, Diabetic Retinopathy Clinical Research; ETDRS, Early Treatment Diabetic Retinopathy Study; IOP, intraocular pressure; IQR, interquartile range; IVR, intravitreal ranibizumab; IVTA, intravitreal triamcinolone acetonide; LogMAR, logarithm of the minimum angle of resolution; MGP, modified grid laser photocoagulation; NR, not reported; PRN, pro re nata; q4w, monthly dose; RAN, ranibizumab; RAZ, razumab; RWE, real-world evidence; SD, standard deviation; SLR, systematic literature review; T&E, treat and extend; VEGF, vascular endothelial growth factor.

Author Contributions

All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Funding

This research was funded by Novartis Singapore Pte Ltd.

Disclosure

IHT (Novartis Singapore Pte Ltd.), PB, and JS (Novartis Healthcare Pvt. Ltd.) are Novartis employees. RKM was an employee of Novartis when the review was being conducted. YSY, GSWT, and NYG have acted as consultants for Novartis for this work and other work. YSY is a consultant for Roche and receives financial support from Bayer. GSWT is a consultant for Roche, Bayer, Allergan, Zeiss, Nikon-Optos, Topcon, and Leica, has received grants from Santen, and owns equity in Eyris. NYG receives financial support from Bayer. The authors report no other conflicts of interest related to this work.

References

1. Musat O, Cernat C, Labib M, et al. Diabetic macular edema. Rom J Ophthalmol. 2015;59(3):133–136.

2. Browning DJ, Stewart MW, Lee C. Diabetic macular edema: evidence-based management. Indian J Ophthalmol. 2018;66(12):1736–1750. doi:10.4103/ijo.IJO_1240_18

3. Elnahry AG, Elnahry GA, Cicinelli MV. Optical coherence tomography angiography of macular perfusion changes after anti-VEGF therapy for diabetic macular edema: a systematic review. J Diabetes Res. 2021;2021:6634637. doi:10.1155/2021/6634637

4. Lee R, Wong TY, Sabanayagam C. Epidemiology of diabetic retinopathy, diabetic macular edema and related vision loss. Eye Vis. 2015;2(1):17. doi:10.1186/s40662-015-0026-2

5. O’Doherty M, Dooley I, Hickey-Dwyer M. Interventions for diabetic macular oedema: a systematic review of the literature. Br J Ophthalmol. 2008;92(12):1581–1590. doi:10.1136/bjo.2008.144550

6. Gundogan FC, Yolcu U, Akay F, Ilhan A, Ozge G, Uzun S. Diabetic macular edema. Pak J Med Sci. 2016;32(2):505–510. doi:10.12669/pjms.322.8496

7. Furino C, Boscia F, Reibaldi M, Alessio G. Intravitreal therapy for diabetic macular edema: an update. J Ophthalmol. 2021;2021:6654168. doi:10.1155/2021/6654168

8. Kodjikian L, Bellocq D, Mathis T. Pharmacological management of diabetic macular edema in real-life observational studies. Biomed Res Int. 2018;2018:8289253. doi:10.1155/2018/8289253

9. ETDRS. Early photocoagulation for diabetic retinopathy. ETDRS report number 9. Early treatment diabetic retinopathy study research group. Ophthalmology. 1991;98(5 Suppl):766–785. 10.1016/S0161-6420(13)38011-7.

10. Chhablani J, Wong K, Tan GS, et al. Diabetic macular edema management in asian population: expert panel consensus guidelines. Asia Pac J Ophthalmol. 2020;9(5):426–434. doi:10.1097/APO.0000000000000312

11. Das T, Aurora A, Chhablani J, et al. Evidence-based review of diabetic macular edema management: consensus statement on Indian treatment guidelines. Indian J Ophthalmol. 2016;64(1):14–25. doi:10.4103/0301-4738.178142

12. Meduri A, Oliverio GW, Trombetta L, Giordano M, Inferrera L, Trombetta CJ. Optical coherence tomography predictors of favorable functional response in naïve diabetic macular edema eyes treated with dexamethasone implants as a first-line agent. J Ophthalmol. 2021;2021:6639418. doi:10.1155/2021/6639418

13. Stefanini FR, Badaro E, Falabella P, Koss M, Farah ME, Maia M. Anti-VEGF for the management of diabetic macular edema. J Immunol Res. 2014;2014:632307. doi:10.1155/2014/632307

14. Mitchell P, Bandello F, Schmidt-Erfurth U, et al. The RESTORE study: ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema. Ophthalmology. 2011;118(4):615–625. doi:10.1016/j.ophtha.2011.01.031

15. Nguyen QD, Brown DM, Marcus DM, et al. Ranibizumab for diabetic macular edema: results from 2 Phase III randomized trials: RISE and RIDE. Ophthalmology. 2012;119(4):789–801. doi:10.1016/j.ophtha.2011.12.039

16. Brown DM, Nguyen QD, Marcus DM, et al. Long-term outcomes of ranibizumab therapy for diabetic macular edema: the 36-month results from two phase III trials: RISE and RIDE. Ophthalmology. 2013;120(10):2013–2022. doi:10.1016/j.ophtha.2013.02.034

17. Heier JS, Korobelnik J-F, Brown DM, et al. Intravitreal aflibercept for diabetic macular edema: 148-week results from the vista and VIVID studies. Ophthalmology. 2016;123(11):2376–2385. doi:10.1016/j.ophtha.2016.07.032

18. Massin P, Bandello F, Garweg JG, et al. Safety and efficacy of ranibizumab in diabetic macular edema (RESOLVE Study): a 12-month, randomized, controlled, double-masked, multicenter Phase II study. Diabetes Care. 2010;33(11):2399–2405. doi:10.2337/dc10-0493

19. Higgins JP, Chandler J, Cumpston M, Li T, Page MJ, Welch VA, editor(s). Cochrane Handbook for Systematic Reviews of Interventions Version 6.2 (Updated February 2021). Cochrane; 2021.

20. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi:10.1136/bmj.n71

21. Gregori NZ, Feuer W, Rosenfeld PJ. Novel method for analyzing Snellen visual acuity measurements. Retina. 2010;30(7):1046–1050. doi:10.1097/IAE.0b013e3181d87e04

22. Wells JA, Glassman AR, Jampol LM, et al. Association of baseline visual acuity and retinal thickness with 1-year efficacy of aflibercept, bevacizumab, and ranibizumab for diabetic macular edema. JAMA Ophthalmol. 2016;134(2):127–134. doi:10.1001/jamaophthalmol.2015.4599

23. Shiraya T, Kure K, Araki F, Kato S, Kaiya T. Correlation between anterior chamber flare changes and diabetic macular edema after intravitreal injection of ranibizumab and aflibercept. Jpn J Ophthalmol. 2020;64(3):250–256. doi:10.1007/s10384-019-00698-z

24. Kanchanaranya N, Rojdamrongratana D, Piyasoonthorn P. Incidence of post-intravitreal anti-VEGF endophthalmitis at Thammasat University Hospital. J Med Assoc Thai. 2015;98(5):489–494.

25. Cho YJ, Lee DH, Kim M. Optical coherence tomography findings predictive of response to treatment in diabetic macular edema. J Int Med Res. 2018;46(11):4455–4464. doi:10.1177/0300060518798503

26. Ma DJ, Park KH, Woo SJ. Predicting 1-month response of macular edema to intravitreal bevacizumab from 1-hour response. Can J Opthalmol. 2014;49(3):267–272. doi:10.1016/j.jcjo.2014.03.007

27. Shimura M, Kitano S, Muramatsu D, et al. Real-world management of treatment-naïve diabetic macular oedema in Japan: two-year visual outcomes with and without anti-VEGF therapy in the STREAT-DME study. Br J Opthalmol. 2020;104(9):1209–1215. doi:10.1136/bjophthalmol-2019-315199

28. Hwang HS, Chae JB, Kim JY, Kim DY. Association between hyperreflective dots on spectral-domain optical coherence tomography in macular edema and response to treatment. Invest Ophthalmol Vis Sci. 2017;58(13):5958–5967. doi:10.1167/iovs.17-22725

29. Sheu S-J, Lee -Y-Y, Horng Y-H, Lin H-S, Lai W-Y, Tsen C-L. Characteristics of diabetic macular edema on optical coherence tomography may change over time or after treatment. Clin Opthalmol. 2018;12:1887–1893. doi:10.2147/OPTH.S173956

30. Chen Y-P, Wu A-L, Chuang -C-C, Chen S-N. Factors influencing clinical outcomes in patients with diabetic macular edema treated with intravitreal ranibizumab: comparison between responder and non-responder cases. Sci Rep. 2019;9(1):10952. doi:10.1038/s41598-019-47241-1

31. Sato S, Shinoda H, Nagai N, et al. Predictive factors of better outcomes by monotherapy of an antivascular endothelial growth factor drug, ranibizumab, for diabetic macular edema in clinical practice. Medicine. 2017;96(16). doi:10.1097/MD.0000000000006459

32. Ghosh S, Dutta J, Mukhopadhyay S, Bhaduri G. Correlation of macular thickness and posterior hyaloid change following bevacizumab and triamcinolone in diffuse diabetic macular edema in middle-aged patients. Int Ophthalmol. 2011;31(5):363–368. doi:10.1007/s10792-011-9464-6

33. Liu S, Wang D, Chen F, Zhang X. Hyperreflective foci in OCT image as a biomarker of poor prognosis in diabetic macular edema patients treating with Conbercept in China. BMC Ophthalmol. 2019;19(1):157. doi:10.1186/s12886-019-1168-0

34. John. Outcomes of Anti-Vascular Endothelial Growth Factor (anti-VEGF) for Diabetic Macular Edema (DME) by 5 loading doses followed by as needed regimen. Presented at: APVRS; 2018; Korea.

35. Hayashi Y, Tatsumi T, Oshitari T, et al. Comparisons of one to three monthly injections of aflibercept for diabetic macular edema by practical protocol. J Diabetes Res. 2021;2021:1374891. doi:10.1155/2021/1374891

36. Xu Y, Rong A, Xu W, Niu Y, Wang Z. Comparison of 12-month therapeutic effect of conbercept and ranibizumab for diabetic macular edema: a real-life clinical practice study. BMC Ophthalmol. 2017;17(1):158. doi:10.1186/s12886-017-0554-8

37. Nagai N, Suzuki M, Uchida A, et al. The area and number of intraretinal cystoid spaces predict the visual outcome after ranibizumab monotherapy in diabetic macular Edema. J Clin Med. 2020;9(5):1391. doi:10.3390/jcm9051391

38. Kaiho T, Oshitari T, Tatsumi T, et al. Efficacy of one-year treatment with aflibercept for diabetic macular edema with practical protocol. Biomed Res Int. 2017;2017:1–6. doi:10.1155/2017/7879691

39. Yoshitake T, Murakami T, Suzuma K, Dodo Y, Fujimoto M, Tsujikawa A. Hyperreflective foci in the outer retinal layers as a predictor of the functional efficacy of ranibizumab for diabetic macular edema. Sci Rep. 2020;10(1):873. doi:10.1038/s41598-020-57646-y

40. Sepehr S, Haya Husam A-A, Sophie H, Moaz A, Ehsan V, David S Efficacy of intravitreal avastin for the treatment of centre involving diabetic maculopathy in a real-world setting (Auckland DME study II). In: 51st Annual Scientific Congress of the Royal Australian and New Zealand College of Ophthalmologists. 2019:147. Available from: http://www.ranzco2019.com/wp-content/uploads/2019/11/RANZCO-2019-Abstract-Handbook-1.pdf.

41. Tsai M-J, Hsieh Y-T, Peng Y-J. Real-life experience of ranibizumab for diabetic macular edema in Taiwan. Int Ophthalmol. 2019;39(7):1511–1522. doi:10.1007/s10792-018-0970-7

42. Chujo S, Sugimoto M, Sasaki T, et al. Comparison of 2-year outcomes between intravitreal ranibizumab and intravitreal aflibercept for diabetic macular edema with “treat-and-extend” regimen—its usefulness and problems. J Clin Med. 2020;9(9):1–12. doi:10.3390/jcm9092848

43. Choovuthayakorn J, Phinyo P, Tantraworasin A, et al. Intravitreal anti-vascular endothelial growth factor therapy for diabetic macular edema in clinical practice of single center: three-year outcomes. Ophthalmic Res. 2021;64(3):483–493. doi:10.1159/000512300

44. Zhou Q, Guo C, You A, Wang D, Wang W, Zhang X. One-year outcomes of novel VEGF decoy receptor therapy with intravitreal conbercept in diabetic retinopathy-induced macular edema. Mol Vis. 2019;25:636–644.

45. Yu Q, Wang F, Zhou L, Yang J, Liu K, Xu X. Quantification of diabetic retinopathy lesions in DME patients with intravitreal conbercept treatment using deep learning. Ophthalmic Surg Lasers Imaging Retina. 2020;51(2):95–100. doi:10.3928/23258160-20200129-05

46. Xu Y, Rong A, Bi Y, Xu W. Intravitreal conbercept injection with and without grid laser photocoagulation in the treatment of diffuse diabetic macular edema in real-life clinical practice. J Ophthalmol. 2016;2016:2143082. doi:10.1155/2016/2143082

47. Xu Y, Qu Y, Suo Y, et al. Correlation of retinal layer changes with vision gain in diabetic macular edema during conbercept treatment. BMC Ophthalmol. 2019;19(1):123. doi:10.1186/s12886-019-1131-0

48. Lai IP, Huang W-L, Yang C-M, Yang C-H, Ho T-C, Hsieh Y-T. Renal biomarkers for treatment effect of ranibizumab for diabetic macular edema. J Diabetes Res. 2020b;2020:1–9. doi:10.1155/2020/7239570

49. Lai -T-T, Hsieh Y-T, Yang C-M, Ho T-C, Yang C-H. Effect of reimbursement policy on visual outcomes in patients with diabetic macular edema treated with ranibizumab. Retina. 2020;40(11):2191–2197. doi:10.1097/IAE.0000000000002716

50. Lai -T-T, Yang C-M, Yang C-H, Ho T-C, Hsieh Y-T. Treatment outcomes and predicting factors for diabetic macular edema treated with ranibizumab – one-year real-life results in Taiwan. J Formos Med Assoc. 2019;118(1):194–202. doi:10.1016/j.jfma.2018.03.009

51. Murakami T, Suzuma K, Uji A, et al. Association between characteristics of foveal cystoid spaces and short-term responsiveness to ranibizumab for diabetic macular edema. Jpn J Ophthalmol. 2018;62(3):292–301. doi:10.1007/s10384-018-0575-8

52. Seo KH, Yu S-Y, Kim M, Kwak HW. visual and morphologic outcomes of intravitreal ranibizumab for diabetic macular edema based on optical coherence tomography patterns. Retina. 2016;36(3):588–595. doi:10.1097/iae.0000000000000770

53. Yi-Sheng C. One-year outcome of aflibercept treatment for diabetic macular edema in Taiwan. In: The 13th APVRS; 2019:137. Avaialable from: https://2019.apvrs.org/wp-content/uploads/2019/10/APVRS-Abstract-Book-2019_20191113_ONLINE.pdf.

54. Lee H, Kang KE, Chung H, Kim HC. Three-dimensional analysis of morphologic changes and visual outcomes in diabetic macular edema. Jpn J Ophthalmol. 2019;63(3):234–242. doi:10.1007/s10384-019-00657-8

55. Li F, Zhang L, Wang Y, et al. One-year outcome of conbercept therapy for diabetic macular edema. Curr Eye Res. 2018;43(2):218–223. doi:10.1080/02713683.2017.1379542

56. Kimberly S, Thomas H, Andrew C. Comparison of intravitreal aflibercept, ranibizumab and dexamethasone implant in eyes with persistent diabetic macular oedema despite prior bevacizumab treatment. Clin Experiment Ophthalmol. 2018;46(1):82–132.

57. Kelkar A, Webers C, Shetty R, et al. Factors affecting compliance to intravitreal anti-vascular endothelial growth factor therapy in Indian patients with retinal vein occlusion, age-related macular degeneration, and diabetic macular edema. Indian J Ophthalmol. 2020;68(10):2143–2147. doi:10.4103/ijo.IJO_1866_19

58. Roh MI, Kim JH, Kwon OW. Features of optical coherence tomography are predictive of visual outcomes after intravitreal bevacizumab injection for diabetic macular edema. Ophthalmologica. 2010;224(6):374–380. doi:10.1159/000313820

59. Jain P, Sheth J, Anantharaman G, Gopalakrishnan M. Real-world evidence of safety profile of intravitreal bevacizumab (Avastin) in an Indian scenario. Indian J Ophthalmol. 2017;65(7):596–602. doi:10.4103/ijo.IJO_992_16

60. Baker CW, Glassman AR, Beaulieu WT, et al. Effect of initial management with aflibercept vs laser photocoagulation vs observation on vision loss among patients with diabetic macular edema involving the center of the macula and good visual acuity: a randomized clinical trial. JAMA. 2019;321(19):1880–1894. doi:10.1001/jama.2019.5790

61. Ang JL, Ah-Moye S, Kim LN, et al. A systematic review of real-world evidence of the management of macular oedema secondary to branch retinal vein occlusion. Eye. 2020;34(10):1770–1796. doi:10.1038/s41433-020-0861-9

62. Wells JA, Glassman AR; Diabetic Retinopathy Clinical Research N. Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema. N Engl J Med. 2015;372(13):1193–1203. doi:10.1056/NEJMoa1414264.

63. Massin P, Creuzot-Garcher C, Kodjikian L, et al. Real-world outcomes with ranibizumab 0.5 mg in patients with visual impairment due to diabetic macular edema: 12-month results from the 36-month BOREAL-DME study. Ophthalmic Res. 2019;62(2):101–110. doi:10.1159/000497406

64. Paul M, Soumil P, Wayne M. Effectiveness of ranibizumab for the treatment of patients with diabetic macular edema in a real-world setting: 1- and 2-year results from the LUMINOUS™ study. Invest Ophthalmol Vis Sci. 2018;59(9):3617.

65. Van Aken E, Favreau M, Ramboer E, et al. Real-world outcomes in patients with diabetic macular edema treated long term with ranibizumab (VISION study). Clin Ophthalmol. 2020;14:4173–4185. doi:10.2147/OPTH.S281501

66. Lukic M, Williams G, Shalchi Z, et al. Intravitreal aflibercept for diabetic macular oedema: Moorfields’ real-world 12-month visual acuity and anatomical outcomes. Eur J Ophthalmol. 2020;30(3):557–562. doi:10.1177/1120672119833270

67. Ciulla TA, Pollack JS, Williams DF. Visual acuity outcomes and anti-VEGF therapy intensity in diabetic macular oedema: a real-world analysis of 28 658 patient eyes. Br J Ophthalmol. 2021;105(2):216–221. doi:10.1136/bjophthalmol-2020-315933

68. Korobelnik J-F, Daien V, Faure C, et al. Real-world outcomes following 12 months of intravitreal aflibercept monotherapy in patients with diabetic macular edema in France: results from the APOLLON study. Graefes Arch Clin Exp Ophthalmol. 2020;258(3):521–528. doi:10.1007/s00417-019-04592-9

69. Plaza-Ramos P, Borque E, Garcia-Layana A, Vavvas DG. Evaluation of ranibizumab and aflibercept for the treatment of diabetic macular edema in daily clinical practice. PLoS One. 2019;14(10):e0223793. doi:10.1371/journal.pone.0223793

70. Bressler SB. 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

71. Wallick CJ, Hansen RN, Campbell J, Kiss S, Kowalski JW, Sullivan SD. Comorbidity and health care resource use among commercially insured non-elderly patients with diabetic macular edema. Ophthalmic Surg Lasers Imaging Retina. 2015;46(7):744–751. doi:10.3928/23258160-20150730-09

72. Mukkamala L, Bhagat N, Zarbin MA. Practical lessons from protocol i for the management of diabetic macular edema. Dev Ophthalmol. 2017;60:91–108. doi:10.1159/000459692

73. Bucolo C, Gozzo L, Longo L, Mansueto S, Vitale DC, Drago F. Long-term efficacy and safety profile of multiple injections of intravitreal dexamethasone implant to manage diabetic macular edema: a systematic review of real-world studies. J Pharmacol Sci. 2018;138(4):219–232. doi:10.1016/j.jphs.2018.11.001

74. Masahiko S, Mineo K, Toshiaki S, Yasuhiro T. Real-world data of intravitreal aflibercept for diabetic macular edema: 6-month outcomes of Japan postmarketing surveillance. In: The 34th Asia-Pacific Academy of Ophthalmology (APAO) Congress; 2019. Available from: https://2019.apaophth.org/wp-content/uploads/2019/01/APAO19_Abstract-Book-0301.pdf. Accessed October 3, 2022.

75. Di Y, Li Z, Ye J, Li L, Li B, Yu R. The fellow eye effect of unilateral intravitreal conbercept injections in eyes with diabetic macular edema. Acta Diabetol. 2020;57(8):1001–1007. doi:10.1007/s00592-020-01511-x

76. Terada N, Murakami T, Uji A, Dodo Y, Mori Y, Tsujikawa A. Hyperreflective walls in foveal cystoid spaces as a biomarker of diabetic macular edema refractory to anti-VEGF treatment. Sci Rep. 2020;10(1):7299. doi:10.1038/s41598-020-64332-6

77. Pongsachareonnont P, Charoenphol P, Hurst C, Somkijrungroj T. The effect of anti-vascular endothelial growth factor on retinal microvascular changes in diabetic macular edema using swept-source optical coherence tomography angiography. Clin Opthalmol. 2020;14:3871–3880. doi:10.2147/OPTH.S270410

78. Chen -N-N, Chen W-D, Lai C-H, et al. Optical coherence tomographic patterns as predictors of structural outcome after intravitreal ranibizumab in diabetic macula edema. Clin Opthalmol2020;14:4023–4030. doi:10.2147/OPTH.S264669

79. Hsieh Y-T, Alam MN, Le D, et al. OCT Angiography biomarkers for predicting visual outcomes after ranibizumab treatment for diabetic macular edema. Ophthalmol Retina. 2019;3(10):826–834. doi:10.1016/j.oret.2019.04.027

80. Muto T, Machida S. Effect of intravitreal aflibercept on corneal endothelial cells: a 6-month follow-up study. Clin Opthalmol. 2019;13:373–381. doi:10.2147/OPTH.S177506

81. Chen -Y-Y, Chen P-Y, Chen F-T, Chen Y-J, Wang J-K. Comparison of efficacy of intravitreal ranibizumab between non-vitrectomized and vitrectomized eyes with diabetic macular edema. Int Ophthalmol. 2018;38(1):293–299. doi:10.1007/s10792-017-0462-1

82. Kim TK, Shin HY, Kim SY, Lee YC, Lee MY. Factors influencing intravitreal bevacizumab and triamcinolone treatment in patients with diabetic macular edema. Eur J Ophthalmol. 2017;27(6):746–750. doi:10.5301/ejo.5000974

83. 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

84. Chang CK, Cheng CK, Peng CH. The incidence and risk factors for the development of vitreomacular interface abnormality in diabetic macular edema treated with intravitreal injection of anti-VEGF. Review. Eye. 2017;31(5):762–770. doi:10.1038/eye.2016.317

85. Chen -Y-Y, Chang P-Y, Wang J-K. Intravitreal aflibercept for patients with diabetic macular edema refractory to bevacizumab or ranibizumab: analysis of response to aflibercept. Asia Pac J Ophthalmol. 2017;6(3):250–255. doi:10.22608/APO.2016186

86. Yang HS, Woo JE, Kim MH, Kim DY, Yoon YH. Co-evaluation of peripapillary rnfl thickness and retinal thickness in patients with diabetic macular edema: RNFL misinterpretation and its adjustment. PLoS One. 2017;12(1). doi:10.1371/journal.pone.0170341

87. Sugimoto M, Ichio A, Nunome T, Kondo M. Two year result of intravitreal bevacizumab for diabetic macular edema using treat and extend protocol. Medicine. 2017;96(16). doi:10.1097/MD.0000000000006406

88. Mangunkusumo EA, Ningtyas YS. A retrospective case series: response of intravitreal bevacizumab 1.25 mg in diabetic macular edema patients based on HbA1C control. Ophthalmologica. 2016;236:41. doi:10.1159/000448911

89. Abe S, Goto S, Nishi K, Yamamoto T, Yamashita H. Effects of second-line intravitreal anti- VEGF therapy for refractory diabetic macular edema. Invest Ophthalmol Vis Sci. 2016;57(12):2069.

90. Wu P-C, Lai C-H, Chen C-L, Kuo C-N. Optical coherence tomographic patterns in diabetic macula edema can predict the effects of intravitreal bevacizumab injection as primary treatment. J Ocul Pharmacol Ther. 2012;28(1):59–64. doi:10.1089/jop.2011.0070

91. Song JH, Lee JJ, Lee SJ. Comparison of the short-term effects of intravitreal triamcinolone acetonide and bevacizumab injection for diabetic macular edema. Korean J Ophthalmol. 2011;25(3):156–160. doi:10.3341/kjo.2011.25.3.156

92. Hwang H, Lee H, Kim JY, et al. Systemic factors and early treatment response to intravitreal injection for diabetic macular edema; the role of renal function. Retina. 2020;Publish Ahead of Print. doi:10.1097/IAE.0000000000003012

93. Hu Y, Wu Q, Liu B, et al. Comparison of clinical outcomes of different components of diabetic macular edema on optical coherence tomography. Graefes Arch Clin Exp Ophthalmol. 2019;257(12):2613–2621. doi:10.1007/s00417-019-04471-3

94. Hu Y, Cheng Y, Xu X, et al. Pretreatment neutrophil-to-lymphocyte ratio predicts prognosis in patients with diabetic macular edema treated with ranibizumab. BMC Ophthalmol. 2019;19(1):194. doi:10.1186/s12886-019-1200-4

95. Namba R, Kaneko H, Suzumura A, et al. In vitro epiretinal membrane model and antibody permeability: relationship with anti-VEGF resistance in diabetic macular edema. Invest Ophthalmol Vis Sci. 2019;60(8):2942–2949. doi:10.1167/iovs.19-26788

96. Akhtar HMU, Jandan NA, Diyal SM. Effect of intravitreal bevacizumab for the treatment of diabetic macular edema. Pak J Med Health Sci. 2017;11(3):1146–1148.

97. Mori Y, Murakami T, Suzuma K, et al. Relation between macular morphology and treatment frequency during twelve months with ranibizumab for diabetic macular edema. PLoS One. 2017;12(4):e0175809. doi:10.1371/journal.pone.0175809

98. Shimizu N, Oshitari T, Tatsumi T, et al. Comparisons of efficacy of intravitreal aflibercept and ranibizumab in eyes with diabetic macular edema. Biomed Res Int. 2017;2017:1–7. doi:10.1155/2017/1747108

99. Li XQ, Meng XX, Wang FL, Fu YD. Conbercept in treating diabetic macular edema based on optical coherence tomography patterns. Biomed Res. 2017;28(21):9423–9428.

100. Lee JH, Lee WK, Kim SE. Short-term outcomes of switching to ranibizumab therapy for diabetic macular edema in patients with persistent fluid after bevacizumab therapy. J Ocul Pharmacol Ther. 2016;32(10):659–664. doi:10.1089/jop.2016.0074

101. Kang J-W, Chung H, Kim HC. Correlation of optical coherence tomographic hyperreflective foci with visual outcomes in different patterns of diabetic macular edema. Retina. 2016;36(9):1630–1639. doi:10.1097/IAE.0000000000000995

102. Lee K, Chung H, Park Y, Sohn J. Efficacy of intravitreal anti-vascular endothelial growth factor or steroid injection in diabetic macular edema according to fluid turbidity in optical coherence tomography. Korean J Ophthalmol. 2014;28(4):298–305. doi:10.3341/kjo.2014.28.4.298

103. Lee SJ, Kim ET, Moon YS. Intravitreal bevacizumab alone versus combined with macular photocoagulation in diabetic macular edema. Korean J Ophthalmol. 2011;25(5):299–304. doi:10.3341/kjo.2011.25.5.299

104. Verma L, Thulasidas M, Purohit A, Gupta A, Narula R, Talwar D. Clinical efficacy and safety of Razumab® (CESAR) study: our experience with the world’s first biosimilar Ranibizumab. Indian J Ophthalmol. 2021;69(2):347–351. doi:10.4103/ijo.IJO_2516_20

105. Hu Y, Wu Q, Liu B, et al. Restoration of foveal bulge after resolution of diabetic macular edema with coexisting serous retinal detachment. J Diabetes Res. 2020;2020:1–9. doi:10.1155/2020/9705786

106. Yew W, Eve Lyn C, Angeline G. Dropout rates among diabetic macular edema patients on anti-vascular endothelial growth factor therapy in a private eye clinic in Malaysia. In: The 33rd Asia-Pacific Academy of Ophthalmology (APAO) Congress; 2018:099. Available from: https://2018.apaophth.org/wp-content/uploads/2018/01/Abstract-Book-2018.pdf. Accessed October 3, 2022.

107. Okamoto Y, Okamoto F, Hiraoka T, Oshika T. Vision-related quality of life and visual function following intravitreal bevacizumab injection for persistent diabetic macular edema after vitrectomy. Jpn J Ophthalmol. 2014;58(4):369–374. doi:10.1007/s10384-014-0323-7

Creative Commons License © 2022 The Author(s). This work is published and licensed by Dove Medical Press Limited. The full terms of this license are available at https://www.dovepress.com/terms.php and incorporate the Creative Commons Attribution - Non Commercial (unported, v3.0) License. By accessing the work you hereby accept the Terms. Non-commercial uses of the work are permitted without any further permission from Dove Medical Press Limited, provided the work is properly attributed. For permission for commercial use of this work, please see paragraphs 4.2 and 5 of our Terms.