Diabetic macular edema: current and emerging therapies.

Diabetic macular edema is a leading cause of vision impairment among people within the working- age population. This review discusses the pathogenesis of diabetic macular edema and the treatment options currently available for the treatment of diabetic macular edema, including for focal/grid photocoagulation, intravitreal corticosteroids and intravitreal anti-vascular endothelial growth factor agents. The biologic rationale for novel therapeutic agents, many of which are currently being evaluated in clinical trials, also is reviewed.

INTRODUCTION

An estimated 346 million people are affected by diabetes worldwide in 2011, and the number of people with diabetes is expected to double from 2005 to 2030.1 Diabetic retinopathy (DR) is the leading cause of vision loss of working-age adults,2 and diabetic macular edema (DME) is the most frequent cause of vision loss related to diabetes. The Wisconsin Epidemiologic Study of Diabetic Retinopathy found the 14-year incidence of DME in type I diabetics to be 26%.3 Similarly the Diabetes Control and Complications Trial (DCCT) reported that 27% of type I diabetic patients develop DME within 9 years of onset.4 An even higher incidence of macular edema has been reported in older patients with type 2 diabetes (reviewed in5).

Argon laser photocoagulation has been the mainstay of treatment for macular edema since the publication of the results of the Early Treatment Diabetic Retinopathy Study, which showed an approximate 50% reduction in the rate of moderate vision loss at 3 years following laser photocoagulation compared to no treatment.6 However, for patients with center involved macular edema, the risk of moderate vision loss at 3 years remained 15% with treatment.6 Since the publication of ETDRS, the DCCT and United Kingdom Prospective Diabetes Study (UKPDS) have demonstrated that tight glycemic and blood pressure control decrease the risk of microvascular complications of diabetes, including DR and vision loss.47–10 As intensive blood pressure and blood sugar control have become the standard of care, visual outcomes have improved, but recent studies from the Diabetic Retinopathy Clinical Research Network (DRCRnet) indicate that even with the guidelines of tight glycemic and blood pressure control, 12-13% of patients with foveal centered diabetic macular edema who undergo focal/grid laser lose 10 or more ETDRS letters after 2-3 years of follow-up. Additionally, with a baseline median vision of 20/50-20/63, only 36-44% of patients gained 10 or more ETDRS letters at 2-3 years of follow-up, indicating the need for improved treatment modalities.11–13

Within the last 5 years, the use of intravitreal corticosteroids and intravitreal anti-vascular endothelial growth factor (VEGF) agents have come into common clinical practice for the management of diabetic macular edema and several recent randomized clinical trials have shown improved effectiveness of ranibizumab compared to focal/grid laser.12–14 In this review, we briefly discuss the pathogenesis of diabetic macular edema and the treatment options currently available for the treatment of diabetic macular edema. We then review the biologic rationale for novel therapeutic agents, many of which are currently being evaluated in clinical trials.

PATHOGENESIS OF DIABETIC MACULAR EDEMA

The pathogenesis of diabetic macular edema involves the breakdown of the blood-retinal barrier (BRB), which is composed of an inner BRB and an outer BRB. The inner BRB is comprised of tight junctions between retinal vascular endothelial cells as well as retinal glial cells (astrocytes and Müller cells), creating a barrier that is normally impermeable to proteins.15 The outer BRB is formed by tight junctions between retinal pigment epithelial (RPE) cells. Diabetic macular edema is thought to be caused primarily by the breakdown of the inner BRB, though evidence does exist that outer BRB dysfunction may play a role in DME (reviewed in5). Breakdown of the BRB allows for extravasation of proteins and other solutes from capillaries into the extracellular space. This causes a shift in the balance of hydrostatic and oncotic pressure, favoring the accumulation of fluid within the extracellular space and the development of macular edema.15 The mechanisms leading to the breakdown of the BRB in diabetes are complex and are extensively reviewed elsewhere,516–18 but some of the key factors are described below.

Some of the earliest changes seen histologically in the diabetic retina are leukocyte adhesion to capillaries as well as accumulation of advanced glycation end products (AGEs).19 These changes contribute to activation of inflammatory mediators and endothelial cell death. Endothelial cell death contributes to the breakdown of the BRB and can lead to increasing ischemia. In addition to cell loss, breakdown of endothelial cell tight junctions also occurs (reviewed in516). Another histologic change associated with the development of diabetic retinopathy is loss of pericytes, which are cells that are associated with capillaries and are located outside of the blood-retinal barrier. Amongst their functions are the stabilization blood vessels. Loss of pericytes may be related to accumulation of AGEs and to the presence of inflammatory mediators and is associated with the formation of microaneurysms and the breakdown of the BRB.2021 A more extensive discussion of some of the molecular pathways involved with the development of DME is d further in detail with regards to specific therapeutic targets in the subsections later in this review.

CURRENT TREATMENTS OPTIONS FOR DIABETIC MACULAR EDEMA

Laser photocoagulation

As noted above, laser photocoagulation had been the mainstay of treatment for DME for almost for the past 25 years, since the publication of photocoagulation results from the ETDRS.6 The mechanisms of action of focal/grid laser are not well understood and are not discussed in this review. However, the technique involves placement of light, small (around 50 micron) laser burns only within thickened areas of the macula, including direct (focal) treatment of microaneurysms as well as spots scattered approximately two to three burn widths apart (grid) within other areas of edema not accounted for by microaneurysms. While it previously was used as a monotherapy, focal/grid laser is used in conjunction with anti-VEGF therapy, typically when DME persists and is not continuing to improve after at least 6 months of monthly injections of anti-VEGF therapy. It is added to anti-VEGF therapy if the eye has not had complete laser (focal/grid) treatment to all areas of microaneurysms within areas of edema and grid treatment to all other areas of edema, provided it has been at least 3 to 4 months since any prior focal/grid laser. Macular laser for DME likely will continue to be part of the management of DME in selected patients, especially in the developing world, as its lower cost and less intensive management requirements compared to newer treatment modalities still make laser photocoagulation a preferred therapy in some clinical settings.

In addition to the traditional argon or diode laser used for macular photocoagulation, several newer laser technologies are being evaluated. One of these is a navigated laser photocoagulator (NAVILAS) that integrates a scanning laser slit camera with fluorescein angiography and employs computer aided steering of the laser. It has recently been demonstrated that this laser system has increased accuracy for placement of laser compared to standard manual technique,22 but whether this increased accuracy will translate into better clinical results or not, has yet to be determined.

Other newer laser technologies include subthreshold diode-laser micropulse technology2324 and selective retina therapy (SRT),25 which aims at minimizing retinal and RPE tissue damage. A recent randomized clinical trial demonstrated superior results for the primary endpoint of visual acuity for high-density subthreshold diode-laser micropulse photocoagulation compared to standard modified ETDRS laser at 1 year.23 Encouraging results have also been reported for SRT,25 but this treatment modality has not been directly compared to focal/grid laser as employed by the ETDRS or more recently by the DRCR Network. Further phase II/III studies are underway to evaluate these treatment modalities.2627

Because of our understanding of the central role that VEGF plays in the development of DME and of anti-VEGF therapy for the treatment of DME (discussed below), targeted scatter laser photocoagulation to areas of capillary non-perfusion anterior to the posterior pole is under evaluation as a potential treatment for DME. While peripheral scatter laser photocoagulation was not found to be beneficial for the treatment of DME in the ETDRS study,28 in this protocol all areas of mid-peripheral retina were treated regardless of perfusion status. The development of ultra-wide field imaging with fluorescein angiography29 allows for the identification of and selective treatment to areas of ischemic retina.30 Treatment in this fashion may allow for better visual outcomes or less frequent administration of intravitreal agents.

Corticosteroids

Mounting evidence exists that inflammation plays a significant role in the development of diabetic macular edema. As noted above, leukostasis is central to the development of diabetic retinopathy. Through the release of free radicals and enzymes, leukocytes can directly damage endothelial cells and increase the permeability of the BRB.31 Leukocytes also release a variety of cytokines that act through signaling pathways, leading to increased vascular permeability. Cytokines involved in leukocyte-mediated damage include VEGF, tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6) (reviewed in16). Intercellular adhesion molecule-1 (ICAM-1) is a cell-to-cell adhesion molecule that plays a major role in leukostasis. In animal models of diabetic retinopathy, inhibition of ICAM-1 with antibodies prevents leukocyte adhesion and prevents the breakdown of the BRB, which is further evidence of the central role played by leukocytes in the development of DME.32

Given the apparent role of inflammation in the pathogenesis of DME, steroids have more recently been utilized for the treatment of DME. Their mode of action may be largely through their ability to inhibit the expression of VEGF.3334 Two recent randomized controlled trials by the DRCR.net have shed light on the role that steroids should play in the treatment of DME.11–133536 In the first of these studies, the use of focal/grid laser was compared to treatment with 1 mg or 4 mg of intravitreal triamcinolone, with retreatment possible every four months in each arm of the study. At the four month follow-up the triamcinolone arms showed superiority in terms of visual acuity, at one year the treatments appeared equivalent and at the 2-year primary endpoint laser was found to be superior to either of the intravitreal steroid arms. These results held with 3 years of follow-up.113536 In the second of these trials, focal/grid laser alone was compared to 4 mg of intravitreal triamcinolone plus laser. Two additional arms utilized intravitreal ranibizumab, an anti-VEGF agent and are discussed further in the next section below. Similar to the previous study, the triamcinolone plus laser arm showed superiority compared to laser alone in terms of visual acuity at 24 weeks follow-up. However, at one and two years, the treatments appeared fairly equivalent in terms of visual acuity outcome, but with increased rates of cataract and elevated intraocular pressure in the triamcinolone plus laser group.1213 In the subgroup of patients who were pseudophakic at baseline, the triamcinolone plus laser arm appeared superior to the laser alone treatment and equivalent to the treatment arms utilizing anti-VEGF therapy discussed in the following section.1213 In light of these results, due to the increased rate of elevated intraocular pressure and formation of cataract in the triamcinolone plus laser group, the authors of this review only consider use of intravitreal steroids plus focal/grid laser in clinical practice in pseudophakic patients who are unable to follow-up monthly for intravitreal anti-VEGF therapy. However, these results were within a subgroup in which the entire group did not show a benefit with intravitreal corticosteroids, the eyes were not randomized within the pseudophakic eyes (so that other factors could have led to outcomes which were superior to focal/grid laser and seemingly equivalent to anti-VEGF therapy), the eyes were more likely to have additional visits for management of increased intraocular pressure and the number of treatments with anti-VEGF therapy appeared to be two to three per year after the first year of treatment. In light of these results, especially the increased rate of elevated intraocular pressure and formation of cataract in the triamcinolone plus laser group, the authors of this review typically consider use of intravitreal steroids plus focal/grid laser in clinical practice in pseudophakic patients who have not responded to monthly intravitreal anti-VEGF therapy for at least 6 months who then had additional focal/grid laser and continued injections for at least another 4 months.

Sustained release steroid delivery systems offer the possibility of less frequent dosing and in a phase II trial, beneficial results have been seen at the 90-day endpoint.37 In light of the results of the DRCR.net studies above, sustained release steroid drug delivery systems may have limited benefit for the treatment of DME. They may however have a role in difficult to treat DME in post-vitrectomy eyes.38

Anti-VEGF therapy

The VEGF family is a sub-group of growth factors that include VEGF-A, B, C, D, E and placental growth factor (PlGF). VEGF-A is the member of this family that is most critical with regards to the pathogenesis of ocular disease and its signaling induces angiogenesis as well as increased vascular permeability.51639 PlGF also appears to play a lesser role.16 VEGF-A exists in at least nine isoforms due to alternative splicing, with six major isoforms. The VEGF-A165 isoform is the predominant isoform and appears to be the most important in the pathogenesis of ocular disease, including DME. The isoforms vary in their affinity to bind heparin, causing some isoforms to be strictly bound to extracellular matrix (ECM) and other forms to be freely diffusible. VEGF-A165 has intermediate heparin binding ability. The isoforms that bind heparin and therefore have affinity for ECM can undergo proteolytic cleavage and become freely diffusible.39

VEGF-A is the ligand of 2 major receptor tyrosine kinases (RTKs), VEGFR-1 and VEGFR-2. The function of VEGFR-1 is still not completely clear. In addition to VEGF-A, it also binds PlGF and appears to be involved in monocyte chemotaxis as well as the induction of matrix metalloproteinase-9 (MMP-9),39 discussed further below. The VEGFR-2 appears to be the primary receptor responsible for angiogenesis, mitogenesis and induction of vascular permeability.39 PlGF has also been implicated in the induction of increased vascular permeability40 as well as increase in the permeability of RPE cells.41 However, in other animal models, it appears to act as an anti-permeability factor.42

VEGF-A expression can be induced by hypoxia via the transcription factor hypoxia-inducible factor (HIF) and it increases vascular permeability via a number of downstream effects. VEGF-A induces phosphorylation of the tight junction proteins occludin and zonula occludens-1, leading to disruption of tight junctions.516 VEGF-A can also induce phosphorylation and subsequent destabilization of vascular endothelial cadherin, an endothelial cell-cell adhesion molecule.16 Each of these effects leads to increased vascular permeability. Additionally, VEGF has been shown to induce the formation of vesiculo-vacuolar organelles (VVOs) within endothelial cells, which can form into transendothelial cell pores, allowing for the passage of large molecules and fluid through endothelial cells.164344 In addition to its more direct effects on vascular permeability, VEGF indices the expression of ICAM-1. As noted above this is critical to leukostasis which can lead to endothelial cell loss, release of free radicals and release of numerous inflammatory cytokines.516 Further, ischemia induced by leukostasis and endothelial cell death can in turn further increase the expression of VEGF.

Because of its central role in the pathogenesis of DME, VEGF was a logical target for therapy. Several anti-VEGF agents are available. The first anti-VEGF agent used in ophthalmology was pegaptanib (OSI Pharmaceuticals, Long Island, NY, USA), which is based on a 28-nucleotide ribonucleic acid aptamer that binds to the VEGF-A165 isoform and was initially approved for the treatment of neovascular age-related macular degeneration.45 Its use has largely been supplanted by the development of ranibizumab (Genentech, Inc., South San Francisco, CA, USA), a fragment antigen binding (Fab) anti-VEGF agent that neutralizes all isoforms of VEGF-A.39 Bevacizumab (Genentech, Inc., South San Francisco, CA, USA), a humanized monoclonal antibody that binds to all isoforms of VEGF-A, was developed in 1996 and first used for the treatment of human cancers.39 It has also come into widespread clinical use in the treatment of retinal disease.

In the recent study by the DRCRnet, discussed above,1213 in addition to the laser only and laser plus triamcinolone arms of the study, two additional arms using ranibizumab with either prompt or deferred laser were included. Each of the ranibizumab arms showed superior outcomes compared to the laser alone or laser plus triamcinolone groups.1213 Results of another large randomized control trial demonstrated similar results for ranibizumab in the treatment of DME.4647 Promising results for the treatment of DME have also been reported for the use of bevacizumab in a large retrospective and one smaller prospective study48–50 and for pegaptanib in a randomized control trial.51

Another anti-VEGF agent that has been developed and is pending FDA approval for neovascular age-related macular degeneration is VEGF trap-eye. VEGF trap-eye (Regeneron, Tarrytown, NY, USA) consists of the VEGF binding portions of the human VEGFR-1 and VEGFR-2 fused to the Fc portion of the human immunoglobulin-G1. In addition to having high affinity to all isoforms of VEGF-A, it also binds to PlGF. A recent phase II study has also demonstrated superiority of VEGF trap-eye for the treatment of DME compared to laser at 24 weeks.52 KH902 (Kanghong Biotech, Chengdu, China) is a similar molecule that incorporates different domains from the VEGFR-1 and -2 and is currently in phase I/II trials for DME.53 Its likely that each of these above therapeutics may require monthly or bimonthly intravitreal injections.

AAV2-sFLT01 (Genzyme, Cambridge, MA, USA) is a replication deficient adeno-virus expressing VEGFR-1. It offers the possibility of more sustained VEGF blockade. It is currently in phase I safety trial in patients with neovascular age-related macular degeneration.54

ADDITIONAL TARGETS AND EMERGING THERAPIES

mTOR

Mammalian target of rapamycin (mTOR) is a tyrosine kinase that forms two different binding complexes, TORC1 and TORC2. Its activation induces the expression of HIFs, including HIF1-α. HIF1-α controls the expression of over 60 genes, including VEGF.55 Rapamycin (sirolimus) as well as the related drug everolimus inhibit the TORC1 mTOR complex. A recent phase I/II study of five adult patients demonstrated the safety of subconjunctival rapamycin for the treatment of DME.56 Currently, a phase II dose-ranging study is underway for the treatment of DME.57 Palomid 529 is another mTOR inhibitor that affects both TORC1 and TORC2. It is currently undergoing the phase I safety trials for the treatment of neovascular AMD with either intravitreal or subconjunctival injection.5859 Its safety and efficacy to treat DME has not yet been examined.

Protein kinase C

Protein kinase C (PKC) is a family of serine threonine kinases that are activated by growth hormones. At least 13 isoforms exist.60 The PKC-β2 isoforms appear to play an important role in increased vascular permeability due to VEGF and inhibitors of this isoform of PKC reduce VEGF induce leakage of fluorescein.61 VEGF signaling leads to membrane translocation of several PKC isoforms, including PKC-β2. PKC is also activated by diacylglycerol, whose levels are increased in diabetes due to increased glycolysis related to hyperglycemia. PKC activation leads to increased vascular permeability. In animal models, administration of ruboxistaurin (RBX), a PKC-β specific inhibitor, has been shown to block more than 95% of VEGF's effect on increased vascular permeability.61 In addition to its effects on vascular permeability, PKC-β activation also induces ICAM-1 levels, stimulating leukocyte chemotaxis and leukostasis and it also decreases blood flow via induction of endothelin expression.560

The PKC-β specific inhibitor RBX, has been evaluated in several large, multicenter, randomized control trials with regards to its effect on diabetic retinopathy.6263 In one of these phase III studies 685 subjects with advanced non-proliferative diabetic retinopathy were randomized to placebo or 32 mg/day of RBX and followed for at least 3 years.6064 With consultation with the FDA, the primary end point was changed from progression of diabetic retinopathy to the incidence of sustained moderate vision loss (3 ETDRS lines on two consecutive study visits). The rate of moderate visual loss was 9% in the placebo group and 5.5% (P=0.034) in the RBX group. Improvement in visual acuity from baseline was more common in the treatment group and focal laser was less frequent in the treatment group. The second large, double-masked, randomized, multicenter trial examined the efficacy of RBX on DME progression.6062 In this study 686 subjects with DME outside of the fovea and no previous laser were randomized to 3 different doses of RBX or placebo. In this study, the primary end point, decrease in the progression of DME into the center of the macula or laser treatment, was not met. Due to the modest results of the first study and failing to meet the study endpoint in the second of these studies, RBX was not approved by the FDA for DME. In subgroup analysis, however, when considering eyes with long-standing, severe DME, use of RBX decreased visual acuity loss by 30% compared to placebo (17 versus 24 letter loss).6065

RBX and other PKC inhibitors still may play a role in the treatment of DME in certain clinical situations or possibly as part of combination therapy. Further investigation is required.

Ras/Raf/Mek/Erk

Also downstream of VEGFR activation is the Ras/Raf/Mek/Erk signaling cascade that partly mediates VEGF's effects on vascular permeability. A number of inhibitors of this cascade, including Vatalanib (Novartis), Pazopanib (GlaxoSmithKline), AL39324 (Alcon) and AG013958 (Pfizer) have undergone phase 1 and phase II testing for the treatment of neovascular age related macular degeneration and iCo-007, an antisense oligonucleotide targeting c-Raf, has undergone phase I trial for the treatment of DME.

ICAM-1

As detailed above, ICAM-1, plays a critical role in leukostasis in diabetic retinopathy and its inhibition in animal models of diabetes blocks increased vascular permeability.32 It exerts this action through binding with lymphocyte function antigen-1 (LFA-1). SAR 1118 is a LFA-1 inhibitor. In animal models of diabetes, topical administration of this drug has been shown to reach the retina and decrease leukostasis and breakdown of the BRB.66 Topical administration has been shown to be well tolerated in phase I trials.67 Further investigation is needed to determine if SAR 1118 or other ICAM-1 or LFA-1 inhibitors will be safe and effective for the treatment of DME.

MMPs

MMPs are a family of proteinases that cleave structural components of ECM as well as other proteins, including growth factors.16 As noted above MMPs can release ECM-bound VEGF-A. Their activation can contribute to the breakdown of the BRB by degrading ECM proteins as well as by releasing growth factors. Independent of their action as antibiotics, the tetracycline family of antibiotics can inhibit MPPs.68 In addition to inhibiting MMPs, minocycline has also been shown to be neuroprotective against excitotoxicity6970 and has been shown to normalize levels of aquaporin 4, VEGF and interleukin 1β levels in a rodent model of DR.71 Three separate studies have shown beneficial effect of minocylcine in terms of decreasing the permeability of the RBR and reducing diabetes related cytokine production in animal models of DR.686971 It is not clear from these studies whether these effects of minocylcine are mediated by inhibition of MMPs. A phase I/II clinical trial is currently recruiting patients to assess the safety and efficacy of minocyline in patients with DME.72

Nicotinic acetylcholine receptor

Nicotinic acetylcholine receptors are ligand gated ion channels that are made up of following five subunits, α1-10, β1-4, γ, and ε (reviewed in73). As homomeric and heteromeric pentamers can form, numerous subtypes exist. They can be activated by either acetylcholine or nicotine. Vascular endothelial cells contain nAChRs and can also produce acetylcholine. Nicotine has been shown to cause increased vascular permeability in the brain, largely through activation of endothelial α7-Nicotinic acetylcholine receptors (nAChRs).74 Campochiaro et al., recently published reports of a multicenter phase I/II clinical trial assessing the safety and efficacy of topical mecamylamine, a nonspecific nAChR antagonist, for the treatment of chronic DME.75 Mecamylamine was first approved for use as an oral antihypertensive in the 1950s and has a good safety profile. In this study, a topical eye drop was used to minimize systemic side effects. High levels of mecamylamine can be detected in the rabbit retina after topical application of this drop. In this study of 23 patients, 8 patients showed improvement, 9 showed equivocal results and 4 patients showed worsening that could be convincingly attributed to the drug used in this study. The authors hypothesize that differential expression of nAChR subtypes, whose actions can oppose each other, may be responsible for these varied results.75 Mecamylamine may prove to be beneficial to a subset of patients with DME and more specific nAChR antagonist may provide more uniformly beneficial results.

Receptor for Advanced Glycation End-Products

Chronic hyperglycemia leads to the formation of AGEs. AGEs bind to Receptor for Advanced Glycation End-Products (RAGE) which can lead to expression of VEGF, ICAM-1, TNF-α, RAGE and a number of other inflammatory mediators whose pathogenesis in DME are described above (reviewed in76). Several RAGE inhibitors have become available. PF-04494700 (TTP488) is an oral formulation of a small molecule inhibitor of RAGE. Phase II trials have been completed in patients with Alzheimer's and have demonstrated its safety.77

Renin-angiotensin system

The components of the Renin-angiotensin system (RAS) have been identified in the retina (reviewed in51678). The RAS appears to be upregulated in diabetic retinas and angiotensin II (ANG II) can directly stimulate the secretion of VEGF. Angiotensin converting enzyme (ACE)-inhibitors can reduce the overexpression of VEGF seen in animal models of diabetes.51678 ANG II, through binding of the angiotensin type 1 (AT1) receptor, can also induce the expression of P-selectin, ICAM-1, VCAM-1, monocyte chomoattractant protein type 1.78 Based on animal models, the renin inhibitor, aliskiren (Novartis, East Hanover, NJ, USA) may have beneficial effects on DME and DR.79

Despite the potential of medications targeting the RAS, results of the Diabetic Retinopathy Candesartan Trials (DIRECT) were disappointing with regards to the effect of candesartan, an AT1 receptor blocker, on DME.8081

Combined Therapies

It is likely that combinations of the above therapies that target different pathways or different steps of the same pathways will yield the best results for the treatment of DME. Several therapeutics have been developed, which alone can inhibit multiple targets in these signaling cascades. TG100572 and TG100801 a prodrug of TG100572 (TargeGen/Sanofi-Aventis, Paris, FRA) is a multi-targeted kinase inhibitor that targets the VEGFR 1 and 2, fibroblast growth factor receptor 1 and 2, platelet derived growth factor receptor β, as well as a numerous members of the Src family of kinases, some of which are downstream in the VEGF signaling pathway.82 TG100801 when administered topically to rabbit eyes was shown to achieve high levels of the drug in the retina and choroid. Topical application was shown to significantly reduce fluorescein leakage in rat model of branch retinal vein occlusion.82 This compound or other multi-targeted inhibitors may offer the benefits of combined therapies with administration of a single drug.

CONCLUSIONS

The etiology of DME is multifactorial. Understanding of key aspects of the pathogenesis and molecular pathways involved in the development of DME has led to the development of improved therapies for DME that have come into use in clinical practices over the past several years. Additional promising therapeutic agents are currently being evaluated in clinical trials and additional molecular targets exist. Combined therapies targeting multiple pathways may yield synergistic treatment responses.