Article Text

Download PDFPDF

Anti-vascular endothelial growth factor agents for diabetic maculopathy
  1. A Salam,
  2. J DaCosta,
  3. S Sivaprasad
  1. Laser and Retinal Research Unit, Department of Ophthalmology, King's College Hospital, Denmark Hill, London, UK
  1. Correspondence to Sobha Sivaprasad, Laser and Retinal Research Unit, Department of Ophthalmology, King's College Hospital, Denmark Hill, London SE5 9RS, UK; senswathi{at}


The management of diabetic macular oedema is changing. The therapeutic armamentarium for diabetic macular oedema (DMO) includes a new group of drugs that inhibit vascular endothelial growth factor (VEGF). These anti-VEGF agents are already being used widely in DMO in clinical practice despite that several phase III trials on these drugs are still underway. There are no established protocols on the use of these agents in DMO, but short-term results are appealing. This review provides an update on the use of anti-VEGF agents in DMO. Although intravitreal delivery of anti-VEGF agents is a relatively safe procedure, the long-term local and systemic effects of these agents in the diabetic population remain unknown. In this regard, this review also highlights the need for close surveillance of the use of these drugs in this high-risk population.

  • Diabetic maculopathy
  • bevacizumab
  • ranibizumab
  • pegaptanib sodium
  • anti-VEGF
  • retina
  • macula
  • pharmacology
  • treatment other
  • clinical trial

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Diabetic macular oedema (DMO) is the major cause of visual loss in the working-age group in industrialised nations. In the Wisconsin Epidemiologic Study of Diabetic Retinopathy, the prevalence of DMO was 20.1% in type I diabetics, 25.4% in type 2 diabetics on insulin and 13.9% for those not on insulin.1 Despite better medical management of diabetes today, the impending epidemiological right-shift of the population and the increasing prevalence of diabetes are set to impose a huge burden on ophthalmic healthcare. It is therefore important for retinal specialists to brace up to the diabetic epidemic with new strategies aiming for better visual outcomes for DMO.

There are three subtypes of DMO that often coexist: focal oedema refers to localised areas of retinal thickening caused primarily by abnormal leakage from microaneurysms and often accompanied by hard exudates; diffuse macular oedema characterised by generalised leakage from dilated capillaries and breakdown of the retinal pigment epithelial barriers; and the ischaemic variant associated with widening and irregularity of the foveal avascular zone with occlusion of a considerable portion of capillary bed causing compensatory dilation of patent capillaries that leak diffusely.

At present, laser photocoagulation is the only evidence-based treatment available for subjects with clinically significant macular oedema, defined by The Early Treatment Diabetic Retinopathy Study (ETDRS)2 as one of more of the following:

  1. Retinal thickening at or within 500 μm of the centre of the macula.

  2. Hard exudates at or within 500 μm of the centre of the macula if associated with adjacent retinal thickening.

  3. A zone or zones of retinal thickening one disc area in size, at least part of which is within one disc diameter of the centre of the macula.

Although the exact mechanism of action of laser photocoagulation remains unclear, it is thought that this therapy increases the proliferation of retinal pigment epithelium (RPE) and vascular endothelial cells, thereby strengthening the integrity of blood–retinal barrier. Laser photocoagulation reduces moderate visual loss (defined as a loss of <15 ETDRS letters) by up to 50% in 3 years. However, the rate of visual improvement is only 17% (gain of ≥15 ETDRS letters). Eyes with diffuse DMO are poor responders to laser photocoagulation,3 and this therapy is best avoided in purely ischaemic maculopathy. In addition to the unpredictable response rate, laser photocoagulation also causes collateral damage caused by heat spread at the time of treatment and laser scars can expand up to 300% producing focal scotomas.4 Histological studies have shown that burns barely perceptible to the clinician can in fact involve the whole retinal thickness,5 compounding to permanent anatomical changes in the photoreceptors and RPE resulting in long-term visual impairment.6

Other potential adverse effects of laser therapy include reduced colour vision, choroidal neovascularisation, RPE fibrous metaplasia and inadvertent photocoagulation of the centre of the macula.4

Several modifications of laser therapy such as micropulse diode laser7 and optical coherent tomography-guided laser therapy have proven to provide similar or better results with less collateral damage. However, further refinements of these techniques are required.

Pars plana vitrectomy is another option that has been found to be useful in a few cases with thickened and taut posterior hyaloid membrane. However, in addition to it being an invasive procedure, only a minority of patients experience visual and anatomical improvement of DMO after pars plana vitrectomy.8

The recent surge and decline in the use of intravitreal tiramcinolone indicate that it may be useful in reducing the retinal thickness in the short term only with no significant effect on long-term visual outcome. Moreover, adverse events especially glaucoma limit their application.9 The outcomes of biodegradable, extended-release dexamethasone implant (Posurdex, Allergan, Irvine, California, USA) that are undergoing phase III trials will help us better understand the place of steroids in the changing paradigm of the management of DMO ( NCT00168337).

Role of vascular endothelial growth factor in DMO

The proangiogenic factor vascular endothelial growth factor (VEGF) is also a vasopermeability and proinflammatory factor. VEGF expression increases under hypoxic conditions.10 The molecule demonstrates a multiprong attack on the blood–retinal barrier that includes disruptive effects on the endothelial zona occludens and induction of fenestrations on the endothelial cells.11 12 It also causes interendothelial gaps, fragmentation of the endothelium and degenerative changes in endothelial basement membranes affecting the structural integrity of the retinal microvessels with resultant extravasation of blood plasma proteins into the extracellular space.13 14

The VEGF molecule also induces a proinflammatory state in the retina. Upregulation of intercellular adhesion molecule-1 on vascular endothelial cells leads to leucocyte adhesion to the vascular endothelium, capillary occlusion and endothelial cell apoptosis causing further destruction of the blood–retinal barrier.15 VEGF 165 is the isoform most linked to pathological retinal angiogenesis and increased vascular permeability, whereas lower-molecular-weight isoforms are mainly associated with physiological angiogenesis.16 Thus, inhibition of VEGF may represent an alternative or adjunctive remedy for DMO. Therapeutic approaches inhibit at various levels of the VEGF pathway including the genetic elaboration of the VEGF protein, its post-translational inactivation, blockage of VEGF receptors on target cells, inhibition of VEGF receptor activation and steps to counteract the cellular responses to VEGF.14 Currently, the three intravitreal anti-VEGF agents that have shown to be effective in wet age-related macular degeneration include pegaptanib sodium, ranibizumab and bevacizumab.

The objective of this review was to compile and analyse the current evidence on the functional and morphological effects of anti-VEGF agents on DMO and highlight the potential role of these agents in the management of this condition.

Sources and methods of literature search

We conducted a comprehensive search to identify studies evaluating the effects of anti-VEGF agents in the management of DMO. These included all randomised controlled studies, and prospective and retrospective case controls and case series. Studies with <3 months' follow-up and studies in which the results of DMO were not reported separately from those of other conditions (eg, central or branch retinal vein occlusion) were excluded. English and non-English-language articles with abstracts translated to English were retrieved using a keyword search of Medline (January 2003 to January 2009), Embase (up to January 2009), the National Institutes of Health Clinical Trials database and the Association for Research in Vision and Ophthalmology (2003–2008). Search terms included diabetic macular oedema, macula and diabetic, anti-VEGF, Avastin, bevacizumab, Macugen (New York, USA), pegaptanib, Lucentis (San Francisco, USA) or ranibizumab. This was supplemented by manually searching the reference lists of all major review articles.

Data extraction and study appraisal

Information on study design, outcomes and analysis was documented on a standardised data extraction form. Information entered included1 study design,2 method of randomisation and masking in randomised controlled trials, 4 diagnostic criteria (biomicroscopy/fundus fluorescein angiography (FFA), optical coherent tomography),5 length of follow-up,6 drug and doses and treatment regimen,7 outcome measures (visual outcome and change in central macular thickness (CMT)) and8 adverse events. Two investigators (AS and JDC) independently identified and grouped the studies before data entry, discussion and further analysis. If discrepancies were found, a third party (SS) was consulted, and any differences were resolved. Articles felt to be irrelevant to DMO and duplicate studies were excluded based on titles and abstracts.

The main outcomes compiled from these studies included improvements in visual outcome and changes in retinal morphology from baseline at the completion of the follow-up period.

Summary of evidence

Pegaptanib sodium (Macugen)

Pegaptanib is a 28-nucleotide aptamer (a chemically synthesised single-stranded nucleic acid) that binds specifically to the VEGF165 isomer. Pegaptanib is the first aptamer to receive Food and Drug Administration approval for use in humans,17 and it is also the first approved intravitreal anti-VEGF agent for treatment of wet age-related macular degeneration.

A double-masked multicentre controlled phase II randomised clinical trial that compared pegaptanib18 to sham in 172 subjects with DMO showed promising results at a mean follow-up of 9 months. The study showed that when the drug is given at a dose of 0.3 mg intravitreally every 6 weeks for 3 months, subjects randomised to pegaptanib had better visual outcomes (gained ≥10 letters ETDRS or two lines of vision in 34% vs 10% in sham group (p=0.003)) and were more likely to show a reduction in central retinal thickness (68 μm with pegaptanib vs an increase of 4 μm in the sham group (p=0.21)). The pegaptanib group was also less likely to need additional treatment with photocoagulation at follow-up (25% in pegaptanib group vs 48% in the sham group (p=0.04)).18

Currently, a phase III randomised controlled, multicentre trial is being conducted to compare the safety and efficacy of intravitreal injections of 0.3-mg pegaptanib given as often as every 6 weeks for 2 years to sham injections in subjects with DMO ( NCT 00605280).

The beneficial outcomes and the rate of retreatment required in real life when treatment is given on ad hoc frequency based on functional, tomographic and angiographic characteristics need to be assessed to better understand the role of this therapy when compared to macular laser photocoagulation and other anti-VEGF agents.

Ranibizumab (Lucentis)

Ranibizumab is a recombinant humanised anti-VEGF antibody fragment (Fab) of 150 kDa that inhibits all VEGF-A isoforms. Ranibizumab is currently licenced as an intravitreal agent for wet age-related macular degeneration at a dose of 0.5 mg in 0.05 ml. It is delivered monthly and the treatment is guided by the macular thickness and visual acuity.

A case series on 10 eyes of 10 subjects with DMO showed beneficial effects of ranibizumab (0.3 or 0.5 mg) when administered monthly for 3 months. At month 3, five of the ten subjects gained ≥10 letters and the mean decrease in retinal thickness was 122 μm. Although no definite conclusions could be drawn from this study, the higher dose was more effective in achieving antipermeability effect.19 A second non-randomised clinical trial (READ-1)17 studied 10 patients with chronic DME who received intravitreal injections of 0.5 mg of ranibizumab at baseline and at 1, 2, 4 and 6 months. Intraocular injections of ranibizumab significantly reduced mean foveal thickness (from mean 503 to 257 μm) and improved mean visual acuity by 12 ETDRS letters at 7 months and 8 letters at month 12. A phase II randomised controlled trial (READ-2) studied the effects of ranibizumab on 126 subjects with centre-involving DMO. The subjects were randomised to three arms: 0.5 mg ranibizumab administered monthly, laser photocoagulation or a combination of ranibizumab and laser photocoagulation. Twenty-four per cent of the ranibizumab monotherapy group gained at least three lines of vision at 6 months. The mean gain of vision was +7.26 letters in the ranibizumab group compared with −1.07 letters in the laser group and +3.8 letters in the ranibizumab and laser combination group. This study demonstrated that ranibizumab monotherapy is a better option than laser photocoagulation in the short term.

A phase II 12-month randomised control trial (Safety and Efficacy of Ranibizumab in Diabetic Macular Edema With Center Involvement; RESOLVE) demonstrated the effects of ranibizumab when administered using a retreatment regimen guided by visual acuity and tomographic response to treatment. The subjects were randomised to three groups: sham, 0.3 mg/0.05 ml ranibizumab or 0.5 mg/0.05 ml ranibizumab. After the three monthly injections of ranibizumab, all three groups could be treated with laser photocoagulation. The doses of ranibizumab were doubled to 0.6 mg/0.1 ml and 1 mg/0.1 ml, respectively, in the ranibizumab groups if the retinal thickness remained >300 μm at month 1 or if the retinal thickness in the study eye was >225 μm with a reduction in retinal oedema from the previous visit of <50 μm at any monthly visit after the baseline injection. The treatment was discontinued at any monthly visit after the third injection if the retinal thickness was ≤225 μm and the visual acuity was ≥79 letters. The therapy was reinitiated if the retinal thickness increased by ≥50 μm or the visual acuity decreased by five or more letters and is <74 letters. At 12 months, the pooled ranibizumab group (n=102) gained +10.2 letters compared with the sham group that lost a mean of 1 letter. A similar significant response was noted in the decrease of retinal thickness of 200 μm in the ranibizumab group compared with 40 μm in the sham group. This is the first study that assessed the outcome of ranibizumab based on a retreatment strategy that could be translated to real-life clinical practice.

A phase III study on ranibizumab (Efficacy and Safety of Ranibizumab (Intravitreal Injections) in Patients With Visual Impairment Due to Diabetic Macular Edema; RESTORE) is underway to assess the effects of ranibizumab (0.5 mg) as adjunctive or monotherapy to laser treatment after 12 months of treatment in subjects with visual impairment due to DMO ( NCT00687804). Two further phase III trials (A Study of Ranibizumab Injection in Subjects With Clinically Significant Macular Edema With Center Involvement Secondary to Diabetes Mellitus; RISE and A Study of Ranibizumab Injection in Subjects With Clinically Significant Macular Edema With Center Involvement Secondary to Diabetes Mellitus; RIDE studies) will provide more definitive long-term data regarding the effects of ranibizumab in DMO at 24 months ( NCT00473330 and NCT00473382).


Bevacizumab (Avastin, Genentech Inc., San Francisco, California, USA) is a recombinant full-length humanised antibody active against all isoforms of VEGF-A. It is the largest of all the intravitreal anti-VEGF agents available at present, with a molecular weight of 148 000 compared to ranibizumab, which is 48 000, and pegatanib, which is 50 000. Although it is unlicensed for intraocular use, it remains the most popular therapy among the anti-VEGF agents probably driven by cost. It is Food and Drug Administration approved for systemic use in metastatic colorectal cancer.20

The drug has shown similar beneficial effects to ranibizumab in wet age-related macular degeneration. To date, there are six published randomised controlled trials on the use of intravitreal bevacizumab in DMO.21–26 as shown in table 1. Several phase III trials are also underway (

Table 1

The outcomes of the RCTs on intravitreal bevacizumab in DMO

There are several studies that have looked at short-term effects of intravitreal bevacizumab in DMO (table 2).

Table 2

Other studies on intravitreal bevacizumab on DMO

Visual outcomes

Most studies have shown that the anti-VEGF agents may improve visual acuity in the short term, but the magnitude of response varies with the type of anti-VEGF agent, the treatment regimen and case selection. The range of visual improvement varied from no effect to +10 letters ETDRS. The definitions of DMO differ between studies. Very few studies reported effects of these drugs on treatment naive eyes, and most investigators studied the effects of these agents on subjects that have had multiple previous treatments for diffuse DMO ranging from repeated sessions of macular laser, intravitreal triamconolone to vitrectomy. So far, the studies summarised above indicate that anti-VEGF agents will be an important part of the armamentarium for the treatment of DMO. However, its role as an only agent for DMO remains unlikely. These agents may be useful in initiating treatment and in recalcitrant cases. The combination of anti-VEGF and laser therapy also failed to show any significant difference in outcome to laser monotherapy in studies involving ranibizumab and bevacizumab. Similarly, the addition of intravitreal triamcinolone has not shown any additive effect to intravitreal bevacizumab monotherapy. Two randomised controlled trials have shown that there is no dose-related response with bevacizumab. It will be interesting to compare the effects of dose-doubling of ranibizumab in the RESOLVE study.

Effect on CMT

Most studies show improvement in macular thickness after intravitreal anti-VEGF agents including studies that included ischaemic eyes and eyes that had multiple treatments previously. However, the reduction in CMT does not universally parallel the improvement in visual acuity presumably due to structural damage to the photoreceptors, RPE atrophy, lipid exudates and macular ischaemia that would often coexist in these cases. The anatomical improvement is short-lived in most cases. In the study by Arevalo et al,27 repeated intravitreal bevacizumab injections were performed after a mean of 13.8 weeks for the second injection and after 11.5 weeks for the third injection, and results showed that there was continued reduction in the mean CMT at 6 months. Therefore, it seems that repeated intravitreal bevacizumab injections for DMO at intervals longer than 4 weeks might be useful for sustained CMT reduction. Thus, it may be useful to conduct long-term studies in which repeated injections are given at longer intervals based on stringent retreatment algorithms for maximum benefit. Yanyali et al suggested that bevacizumab may clear rapidly from vitrectomised eyes and fail to maintain sustained therapeutic levels. This concept would apply to all anti-VEGF agents.

Dose and frequency of anti-VEGF treatment

The optimum dosing and sequence for intravitreal anti-VEGF in DMO remain unclear. The lower-molecular-weight ranibizumab is cleared more rapidly from the vitreous than the higher-molecular-weight bevacizumab. The terminal half-life (T1/2) has been reported as averaging 3 days for ranibizumab,36 9.8 days for bevacizumab37 and 10 days for pegaptamib.38 Most researchers used a dose of 1.25 mg/0.05 ml for bevacizumab. Higher doses did not show any significant difference in outcomes in the short term. Most studies on bevacizumb assessed the short-term effects of one, two or three injections of the drug at six weekly intervals. There are very few studies that have assessed the effects of different treatment algorithms based on visual acuity and/or CMT. Unlike bevacizumab, the dose-escalating pilot study for ranibuzamab19 administered doses of 0.5 and 0.3 mg with a total of three doses at monthly intervals and noted that the higher-dose group showed a greater decrease in retinal thickness compared with the low-dose group, showing the higher dose to be more effective in achieving the anti-permeability effect. However, it will be useful to compare the effects of 0.6 and 1.0 mg ranibizumab with the baseline doses in the RESOLVE study. The outcomes of the dose-ranging controlled trial on pegaptanib18 at varying doses of 0.3, 1 and 3 mg appeared to favour the 0.3-mg dose with respect to mean gain in visual acuity, greater decrease in retinal thickness and less need for focal/grid laser intervention studied.

At present, it is difficult to define the treatment responders and non-responders. We await the treatment outcomes of several on-going trials using retreatment algorithms that will help to understand the practicalities of using these agents in DMO.

Effect on macular ischaemia

Chung et al29 and Neubauer et al32 are the only groups who have evaluated the effect of bevacizumab on macular ischaemia over a 3-month period. Chung et al defined macular ischaemia on FFA as “an enlarged FAZ ≥100 μm or a broken perifoveal capillary ring at the border of the FAZ with a distinct area of capillary non perfusion within one disc diameter of the foveal centre in the transit phase of FFA.” The results were unfavourable for the group with macular ischaemia with 50% patients losing one or more line and 22% losing three or more lines (ETDRS) at 3-month post-treatment (p=0.042). Neubauer et al32 analysed the effect of bevacizumab on central and peripheral ischaemia by using a novel high-resolution ultra-wise-field scanning laser ophthalmoscopy system and reported an improvement in peripheral ischaemia with no correlation to visual acuity, size of FAZ or CMT. These drugs should be used with caution in eyes with macular ischaemia unless proven otherwise by well-conducted long-term studies.

Local and systemic side effects and risk factors

The incidences of endophthalmitis in these studies are no greater than those published in recent multicentre clinical trials using anti-VEGF therapy for wet age-related macular degeneration and range from 0.7% to 1.6%.39–42 Although delivered within the vitreous, anti-VEGF drugs could pass into the systemic circulation, which could potentially result in hypertension, proteinuria, increased cardiovascular events and impaired wound healing. It is important to set up surveillance on the use of these intravitreal drugs in patients with diabetes who usually suffer from other comorbidities. Ranibizumab and bevacizumab block all VEGF isoforms, whereas pegaptanib only blocks the VEGF(165) isoform, and would therefore seem the best option for avoiding systemic adverse effects in diabetic patients, although this remains to be demonstrated in clinical trials. Head-to-head studies designed to evaluate the efficacy and the systemic adverse effects of these drugs in a high-risk population such as diabetic patients are warranted.


Although VEGF is a key player in diabetic retinopathy, in which breakdown of blood–retinal barrier with or without new vessel formation is a feature, it is not the only factor. A holistic approach is required when administering these drugs to patients with multiple comorbidities. Inhibition of the action of VEGF alone is not sufficient to reverse or modify the retinopathy. Combination treatments have to be considered at a cost of compromising for maximal visual benefit.

Comparison of anti-VEGF agents is equally important. We await the outcomes of head-to-head trials on these agents. At present, the studies differ in study size, study design, inclusion/exclusion criteria, case definition, drug dosage, drug administration schedule, duration of follow-up (45% had 12-week follow- up) and lack of a control group, limiting the usefulness of these studies in providing any specific treatment recommendations.



  • Competing interests Sobha Sivaprasad has received research grants from Novartis and Pfizer.

  • Provenance and peer review Not commissioned; externally peer reviewed.