Statistics continue to underscore the growing epidemic of diabetes worldwide. According to the World Health Organization, diabetes is the fourth most common cause of non-communicable diseases after cardiovascular disease, cancer, and respiratory disease.¹ The global prevalence of diabetes, particularly type 2, has increased rapidly.^2,3 The International Diabetes Federation estimates that more than 382 million people worldwide had diabetes in 2013, an increase of 11 million from 2012, and this figure is projected to rise to almost 600 million by 2035.² Among the almost 400 million people with diabetes, 46%, most of whom had type 2 diabetes, were undiagnosed.

In 2008, about 1 in 3 people with diabetes had diabetic retinopathy.⁴ Diabetes-related blindness costs the nation about $500 million each year.⁴ From 2000 to 2010, the number of cases of diabetic retinopathy increased almost 90%, from about 4 million to almost 8 million.⁵

The American Academy of Ophthalmology recommends annual eye examinations for everyone with diabetes, starting 3 to 5 years after diagnosis of type 1 and at the time of diagnosis for type 2 diabetes; however, in 2011, only 57% of people with diabetes had the recommended eye examinations.⁶

In what could be termed the “one-third rule,” one-third of people with diabetes develop some form of retinopathy, and one-third of patients with retinopathy have diabetic macular edema (DME), while one-third of patients with DME have clinically significant macular edema.² According to the International Diabetes Federation, about 1.5 million people have clinically significant macular edema.^2,7,8

Stages of Diabetic Eye Disease

Nonproliferative diabetic retinopathy (NPDR), an early stage of diabetic eye disease, is a chronic, progressive condition that develops over several years.⁹ Typically, no significant vision loss is associated with NPDR unless DME develops, and DME can occur with any degree of diabetic retinopathy. Significant vision loss can also occur if NPDR progresses to the proliferative stage, which is characterized by neovascularization, with vitreous hemorrhage and/or traction retinal detachment. At these later stages, the risk of severe vision loss is high.

Diabetic macular edema occurs in 13% of patients with diabetic retinopathy. The risk of DME increases as diabetic retinopathy progresses, and it is the most common cause of vision loss in people with diabetes.

Diabetic macular edema can be classified as focal or diffuse.^10,11 Focal DME is characterized by localized areas of retinal thickening, caused by leakage from micro-aneurysms, and areas of focal leakage often demarcated by a partial or complete ring of hard (lipid) exudates. Diffuse DME is characterized by more generalized areas of vascular leakage from dilated, extensively damaged retinal capillaries.

Options for treating DME include focal laser photo-coagulation, intravitreal anti-VEGF agents, corticosteroids, and pars plana vitrectomy.

Laser Photocoagulation

The goal of macular focal laser photocoagulation is to reduce edema and prevent central vision loss. High-energy light is absorbed by retinal tissue, generating heat and coagulating the blood in microaneurysms. This, in turn, reduces leakage and, therefore, retinal edema.

In the Early Treatment Diabetic Retinopathy Study, 727 eyes received focal laser for DME compared with almost 1,500 untreated eyes.¹² At 3 years, 12% of the treated eyes had moderate vision loss compared with 23% of the untreated eyes, and the treated eyes had associated decreased retinal thickening (Figure 1).

Laser photocoagulation has inherent shortcomings. It destroys retinal tissue and can create small blind spots in or near a patient’s central vision. Until recently, laser therapy had to be delivered through a slit lamp, it required manual focus, and the treating physician was unable to readily visualize areas previously identified on the fluorescein angiogram. High-tech, targeted laser is now available. The Navilas OD-OS navigated laser, approved for use in the United States in 2010, has an integrated platform that enables the physician to view imaging during the laser treatment. The patient’s angiogram and the physician’s treatment plan can be superimposed on the patient’s live-tracked fundus image.

Although the role of laser photocoagulation has diminished in recent years, it still has utility to treat focal DME in some cases, particularly DME that does not directly involve the foveal center. It may be used in combination with intravitreal pharmacologics.

Role of VEGF in Diabetic Eye Diseases

Retinal hypoxia from diabetes leads to release of vascular endothelial growth factor (VEGF) and causes leakage of nearby retinal capillaries, contributing to DME formation. VEGF also stimulates growth of new, weak capillaries. Diabetes triggers a metabolic response, which leads to microvascular damage, which in turn leads to retinal ischemia, increased levels of VEGF, increased retinal permeability, retinal vascular leakage, and finally macular edema.

Three anti-VEGF agents are currently employed to treat DME.

• ❒ Bevacizumab (Avastin, Genentech) is a full-length, humanized monoclonal antibody that binds and inhibits VEGF. It is used off-label to treat DME.
• ❒ Ranibizumab (Lucentis, Genentech) is an antibody fragment to VEGF from the same parent antibody as bevacizumab. It is FDA-approved for DME based on the phase 3 RISE and RIDE trials.¹³
• ❒ Aflibercept (Eylea, Regeneron) acts as a decoy VEGF receptor, binding to VEGF and preventing it from binding to VEGF receptors. Aflibercept is FDA-approved to treat DME based on the phase 3 VIVID and VISTA trials.¹⁴

Ranibizumab

The RISE and RIDE trials evaluated the efficacy and safety of ranibizumab for treating DME in eyes with central subfield thickness (CST) of 275 µm or greater and best-corrected visual acuity (BCVA) ranging from 20/40 to 20/320. One eye per patient was randomly assigned to monthly intravitreal injections of ranibizumab 0.3 mg or 0.5 mg or placebo. The need for rescue laser, which was allowed following strict criteria, was assessed monthly beginning at month 3. After month 24, patients receiving placebo were eligible to switch to monthly ranibizumab 0.5 mg and continue in the long-term, open-label extension trial. Patients who started the trial receiving ranibizumab continued receiving the same monthly dose through month 36.

Treated eyes gained an average of 9.1 ETDRS letters at month 6, compared with baseline (Figure 2). These improvements continued and were maintained for 3 years.¹⁵

Central foveal thickness was decreased by an average of 200 µm at month 6 in treated eyes, compared with about 35 or 40 µm in the eyes that received sham injection at month 6. Similarly, these improvements continued through 2 years and were maintained at 3 years (Figure 3).¹⁵

Aflibercept

VIVID and VISTA were randomized, multicenter double-masked trials of aflibercept in eyes with center-involving, clinically significant DME and BCVA ranging from 20/40 to 20/320.¹⁴ Patients received five loading doses of intravitreal aflibercept, 2 mg, followed by continued treatment every 4 weeks (2q4) or every 8 weeks (2q8). The control arm received focal laser photocoagulation. The primary endpoint was mean change in BCVA at 1 year. The study continued through year 3.

At 52 weeks in VISTA (Figure 4), the mean change in BCVA was 12.5 letters for aflibercept 2q4, 10.7 letters for aflibercept 2q8, and 0.2 letters for laser photocoagulation. In VIVID, the mean change was 10.5 letters for aflibercept 2q4, 10.7 letters for aflibercept 2q8, and 1.2 letters for control.

At 100 weeks in VISTA, the mean change in BCVA was 11.5 letters for aflibercept 2q4, 11.1 letters aflibercept 2q8, and 0.9 letters for control.¹⁶ In VIVID, the mean change was 11.4 letters for aflibercept 2q4, 9.4 letters for aflibercept 2q8, and 0.7 letters for control. Rescue laser was less likely in eyes receiving aflibercept either monthly or every 2 months in both trials, anywhere from 3% to 11%.

A secondary endpoint in VIVID and VISTA was change in CST from baseline. Eyes in the treatment arms showed significant thinning of the retina compared with those in the laser arm (Figure 5). Of note is the sawtooth pattern in the 2q8 group, suggesting that not everyone responds or has a full 8-week treatment window. That is in contrast to the 2q4 group where there was little variation in OCT thickness. Ultimately, the results are about the same, with stabilized vision and OCT thickness.

Researchers also found that treated eyes demonstrated a 2-step improvement in the Diabetic Retinopathy Severity Scale, suggesting a disease-modifying component to anti-VEGF therapy. [Editor’s note: FDA has approved both aflibercept and ranibizumab for the treatment of diabetic retinopathy in patients with diabetic macular edema.]

A comparison of visual outcomes versus the number of injections administered over 12 months in key trials evaluating anti-VEGF therapy to treat DME suggests that fixed, frequent dosing may lead to greater benefits in visual acuity (Figure 6).^13,14,17

In summary, VEGF plays a key role in the development and pathophysiology of DME. Anti-VEGF therapy reduces edema, decreases the diabetic retinopathy severity score, and improves visual acuity.

Insights on Anti-VEGF From Protocol T

The Diabetic Retinopathy Clinical Research (DRCR) Network’s Protocol T study, which was funded by the National Eye Institute, is the first head-to-head comparison of the efficacy and safety of the three anti-VEGF agents for the treatment of center-involving DME.18 Protocol T was an interventional, randomized, multicenter, parallel single-masked study of 660 patients who had either type 1 or type 2 diabetes and visual acuity ranging from 20/32 to 20/320 (Figure 7). One eye from each patient was randomly assigned to monthly intravitreal injections of 2 mg aflibercept, 1.25 mg bevacizumab, or 0.3 mg ranibizumab. The primary endpoint was mean change in visual acuity from baseline at 1 year.

The study drugs were administered as often as every 4 weeks, according to a protocol-specified algorithm. Starting at the 24-week visit, regardless of visual acuity and CST, an injection was withheld if there was no improvement or worsening after two consecutive injections. Treatment was resumed if the visual-acuity letter score or the CST worsened.

Focal or grid laser was initiated at or after 24 weeks only if the DME persisted without improvement after two or more injections. The maximum number of injections was 13. The mean number of injections for each anti-VEGF agent was similar for each drug, about 9.2 for aflibercept, 9.7 for bevacizumab, and 9.4 for ranibizumab. The percentage of eyes that received at least one focal grid laser treatment differed significantly among the three agents: 37% of the aflibercept-treated eyes, 56% of bevacizumab-treated eyes, and 46% of ranibizumab-treated eyes.

At week 52, regardless of baseline visual acuity, aflibercept-treated eyes gained 13 letters, the ranibizumab-treated eyes gained 11 letters, and the bevacizumab-treated eyes gained 10 letters. The difference between aflibercept gains and bevacizumab gains was statistically significant.

A key takeaway from Protocol T was the result of comparing visual acuity gains relative to initial visual acuity. Aflibercept showed the most marked mean letter score increase in eyes with worse vision compared with bevacizumab and ranibizumab. This difference was noticeable when the initial baseline visual acuity was worse than 20/40 (Figure 8). When eyes with initial visual acuity of worse than 20/40 were assessed, aflibercept-treated eyes gained a mean of 19 letters compared with 14 for ranibizumab, and about 12 for bevacizumab (Figure 9). This was statistically significant for bevacizumab and approaching statistical significance for ranibizumab, with a P value of .003.

From a different perspective, 67% of aflibercept-treated eyes with baseline visual acuities of worse than 20/40 gained 3 lines of vision, as compared with 41% of bevacizumab-treated eyes and 50% of ranibizumab-treated eyes. These outcomes suggest that aflibercept may be the anti-VEGF of choice for eyes with DME and visual acuity of worse than 20/40.

At 12 months, aflibercept-treated eyes had the greatest improvement in CST, with a reduction of 170 µm. This was significant when compared with a reduction of 100 µm in bevacizumab-treated eyes. Ranibizumab was also superior to bevacizumab, when the entire cohort was evaluated (Figure 10). Researchers also found that aflibercept-treated eyes in the subgroup that had 20/50 visual acuity or worse had the most improvement in CST. This, too, was statistically significant.

In summary, all 3 anti-VEGF agents on average produced substantial visual acuity improvement at 1 month, and the improvement was sustained through 1 year. On average, aflibercept produced the most improvement, but the relative effect varied depending on initial visual acuity. In eyes worse than 20/40, aflibercept had an advantage over the other agents, but if the visual acuity was 20/40 or better, there was little difference in the mean visual acuity at 1 year.

Bevacizumab had a lesser effect on reducing macular edema than the other 2 agents, regardless of baseline visual acuity. Few eyes in any of the groups had significant visual acuity loss. The median number of injections was about the same — 9 or 10 — in all 3 groups. Also of note, fewer eyes in the aflibercept group received focal or grid laser for DME after 24 weeks, possibly because center-involving DME resolved in a higher percentage of the eyes in that group.

Anti-VEGF Safety

More than 6,000 injections were delivered during year 1 of the DRCR Protocol T study, with no cases of endophthalmitis reported, and the incidence of injection-related cataract and IOP elevation was similar across all groups (Figure 11). Systemic adverse events occurred in 3% of patients receiving aflibercept, 4% of those receiving bevacizumab, and 5% of patients receiving ranibizumab (Figure 12).

A post-hoc analysis of Protocol T found the incidence of any cardiovascular event, excluding hypertension, was 9% in the aflibercept group, 9% in the bevacizumab group, and 17% in the ranibizumab group. Two-year data from Protocol T, expected to be released in early 2016, will further clarify this important finding.

Corticosteroids

Corticosteroids, which have long been used for ocular indications, have anti-inflammatory and anti-angiogenic properties. A variety of biochemical factors with pro-inflammatory or vascular effects have been implicated in DME (Figure 13).^11,19

The corticosteroids currently employed to treat DME include triamcinolone acetonide, which is used off label, and two intraocular implants, both of which are FDA-approved for treating DME: the biodegradable dexamethasone 0.7 mg intravitreal implant (Ozurdex, Allergan) and the non-erodible fluocinolone 0.19 mg intravitreal implant (Iluvien, Alimera Sciences).

A confounding factor in studies of DME that involved the use of triamcinolone is that many patients entering the studies were phakic, and at some point, they developed a cataract, causing their vision to decline. The DRCR Protocol I trial determined that eyes that were pseudophakic at baseline and treated with triamcinolone and prompt laser had similar gains in visual acuity as eyes treated with ranibizumab and prompt or deferred laser (Figure 14). In phakic eyes, the gain in visual acuity was less than in the steroid-treated eyes, and this was likely due to cataract formation.

A critical evaluation of these clinical trials should include differentiating whether patients were phakic or pseudophakic upon entry, and determining how outcomes were affected if the natural lens was removed during the trial.

The DRCR Protocol B study evaluated the efficacy and safety of 1-mg and 4-mg doses of preservative-free intravitreal triamcinolone compared with focal/grid photocoagulation for the treatment of DME.²⁰ Researchers found that IOP increased from baseline by 10 mm Hg or more at any visit in 4%, 16%, and 33% of eyes in the 3 treatment groups, respectively, and cataract surgery was performed in 13%, 23%, and 51% of eyes in the 3 treatment groups, respectively. Glaucoma surgery was performed in 4 eyes in the 4-mg triamcinolone group compared with zero in the 1-mg triamcinolone group and zero in the laser group.

Dexamethasone Intravitreal Implant

The dexamethasone 0.7 mg intravitreal implant provides sustained, localized release of the steroid, and biodegrades to lactic acid and glycolic acid. In the MEAD phase 3 clinical trial, more than 1,000 eyes were randomly assigned to receive the dexamethasone implant, either 0.7 mg or 0.35 mg, or a sham procedure.²¹ Eyes were assessed for retreatment eligibility every 3 months at a study-scheduled visit starting from month 6 to month 36. Up to 7 treatments were administered during the 3-year study period.

At the 3-year final visit, the proportion of eyes that gained 15 or more letters from baseline was significantly higher with dexamethasone 0.7 mg (22.2%) compared with sham (12.0%) (Figure 15). However, the occurrence of cataracts in the steroid group had an impact on visual acuity during the study. When evaluated by lens status, researchers found that pseudophakic patients receiving the dexamethasone implant gained 5.8 letters of visual acuity, while those in the sham group gained 1.4 letters (Figure 16).

Steroid-induced cataract was a manageable side effect in the MEAD trial. Among patients who were phakic at baseline, about 60% in the dexamethasone 0.7 mg group required cataract surgery, compared with 8% in the sham group. This is an important consideration when evaluating the data. As shown by Figure 17, dexamethasone-treated eyes showed improvements in visual acuity after cataract removal, and these improvements continued to the end of the study.

From the identification of the cataracts until cataract surgery, more letters were lost in the dexamethasone-treated eyes than in the sham eyes, but after cataract surgery until the end of the study, the dexamethasone-treated eyes, which were now pseudophakic, demonstrated more mean average BCVA improvement than the sham eyes.

During the MEAD study, 28% of eyes treated with dexamethasone experienced an IOP elevation greater than or equal to 10 mm Hg from baseline versus 4% of sham-treated eyes. IOP occasionally seemed to increase after each treatment and then return to baseline (Figure 18). The timing of IOP rises was predictable, and the incidence and magnitude of IOP elevations did not increase upon repeated injection or from year to year in the study.

Increases in IOP that occurred during the MEAD trial were typically manageable with topical medication. One patient treated with the dexamethasone 0.7 mg implant required incisional surgery to control steroid-induced IOP elevation.

Fluocinolone Intravitreal Implant

The non-erodible fluocinolone 0.19 mg implant is approved for the treatment of DME in patients who have been previously treated with a course of corticosteroids and did not have a clinically significant rise in IOP.

The FAME study was a randomized, multicenter, double-masked trial of eyes with DME despite having at least one previous laser treatment.²² They also had CST of 250 µm or more and BCVA ranging from 20/50 to 20/400. Eyes were randomly assigned to fluocinolone 0.2 µm or 0.5 µm or sham. The primary endpoint was percentage of eyes with 15-letter BCVA improvement at week 24; the study continued through year 3. Fluocinolone-treated eyes were more likely to gain 15 or more ETDRS letters at 24 and 36 months, compared with eyes in the sham group (Figure 19).

Researchers found that almost all phakic patients in the fluocinolone groups developed cataracts, but their visual benefit after cataract surgery was similar to that in pseudophakic patients. Incisional IOP-lowering surgery was performed in 4.8% (0.2 µm) and 8.1% (0.5 µm) of fluocinolone-treated eyes compared with less than 1% of control eyes.

In summary, studies show significant improvements in visual acuity can be achieved with the implanted corticosteroids, and this can be accomplished with relatively fewer injections than required for the anti-VEGF agents. Most increases in IOP can be managed with topical treatment. The longer exposure to corticosteroid implants was associated with cataract formation, but visual acuity was restored following cataract surgery.

Vitrectomy

Pharmacotherapy is usually the first-line choice for treating DME, but there is still a role, although limited, for vitrectomy. Vitrectomy may provide some benefit in eyes with vitreomacular traction, or sometimes even non-tractional DME. Vitrectomy surgery can decrease central retinal thickness, but it has a greater potential for complications than pharmacotherapy, including retinal detachment in particular. It is most useful in eyes that have distinct mechanical traction on the macula as detected by OCT. Also of note is the fact that the half-life of anti-VEGF agents is markedly decreased in vitrectomized eyes, while the effect of the dexamethasone sustained-release implant is likely unchanged.^23,24

Treatment of DME Continues to Evolve

In conclusion, our diabetic patients now can benefit from the wide variety of treatment options for use in treating DME that we did not have just as recently as 2 years ago.

The DRCR Protocol T 2-year data will prove to be very enlightening as they will shed light on numerous questions that were raised by the 1-year data — specifically surrounding differences in dosing frequency between the agents as well as safety concerns raised in the post-hoc analysis. We also believe that interest in combination therapy for DME will grow and research in this area will expand given the multifactorial etiology of this disease. ❒

References

1. World Health Organization. WHO Global status report on noncommunicable diseases. 2010. http://www.who.int/nmh/publications/ncd_report_full_en.pdf . Accessed January 17, 2016.
2. International Diabetes Federation. IDF Diabetes Atlas: 6th Edition. 2013. https://www.idf.org/sites/default/files/EN_6E_Atlas_Full_0.pdf . Accessed January 17, 2016.
3. International Diabetes Federation. IDF Diabetes Atlas: 5th Edition. 2012. https://www.idf.org/sites/default/files/21991_diabAtlas_5thEd.pdf . Accessed January 17, 2016.
4. National Diabetes Statistics Report, 2014. http://www.diabetes.org/diabetes-basics/statistics/#sthash.ArSXpmY1.dpuf . Accessed January 17, 2016.
5. NIH-NEI Diabetic Retinopathy. https://nei.nih.gov/eyedata/diabetic#5 Accessed January 17, 2016.
6. HEDIS Health Plan Report Card, Supplement; 2. AAO. Preferred Practice Pattern Guidelines. 2012.
7. International Diabetes Federation. Diabetes & blindness due to DME (2011). www.idf.org/sites/default/files/idf-europe/IDF%20Toolkit_Backgrounder_FINAL.pdf . Accessed January 17, 2016.
8. Wiley HE, Ferris FL. Nonproliferative diabetic retinopathy and diabetic macular edema. Retina. Ryan S, Schachat A, Wilkinson C, et al, eds. Philadelphia, PA: Elsevier; 2013.
9. NHANES database search by Genentech, Data on file.
10. Bandello F, Battaglia Parodi M, Lanzetta P, et al. Diabetic macular edema. Dev Ophthalmol. 2010;47:73-110.
11. Bhagat N, Grigorian RA, Tutela A, Zarbin MA. Diabetic macular edema: pathogenesis and treatment. Surv Ophthalmol. 2009;54:1-32.
12. Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for diabetic macular edema. Early Treatment Diabetic Retinopathy Study Report Number 1. Arch Ophthalmol. 1985;103:1796-1806.
13. Nguyen QD, Brown DM, Marcus DM, et al.; RISE and RIDE Research Group. Ranibizumab for diabetic macular edema: results from 2 phase III randomized trials: RISE and RIDE. Ophthalmology. 2012;119:789-801.
14. Korobelnik JF, Do DV, Schmidt-Erfurth U, et al. Intravitreal aflibercept for diabetic macular edema. Ophthalmology. 2014;121:2247-2254.
15. Brown DM, Nguyen QD, Marcus DM, et al.; RIDE and RISE Research Group. Long-term outcomes of ranibizumab therapy for diabetic macular edema: the 36-month results from two phase III trials: RISE and RIDE. Ophthalmology. 2013;120:2013-2022.
16. Brown DM, Schmidt-Erfurth U, Do DV, et al. Intravitreal aflibercept for diabetic macular edema: 100-week results from the VISTA and VIVID studies. Ophthalmology. 2015;122:2044-2052.
17. Mitchell P, Bandello F, Schmidt-Erfurth U, et al,; RESTORE study group. The RESTORE Study: Ranibizumab monotherapy or combined with laser versus laser monotherapy for diabetic macular edema. Ophthalmology. 2011;118:615-625.
18. Wells JA, Glassman AR, Ayala AR, et al.; Diabetic Retinopathy Clinical Research Network. Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema. N Engl J Med. 2015;372:1193-1203.
19. Rangasamy S, McGuire PG, Das A. Diabetic retinopathy and inflammation: novel therapeutic targets. Middle East Afr J Ophthalmol. 2012;19:52-59.
20. Diabetic Retinopathy Clinical Research Network. A randomized trial comparing intravitreal triamcinolone acetonide and focal/grid photocoagulation for diabetic macular edema. Ophthalmology. 2008;115:1447-1449.
21. Boyer DS, Yoon YH, Belfort R Jr, et al.: Ozurdex MEAD Study Group. Three-year, randomized, sham-controlled trial of dexamethasone intravitreal implant in patients with diabetic macular edema. Ophthalmology. 2014;121:1904-1914.
22. Campochiaro PA, Brown DM, Pearson A, et al.; FAME Study Group. Sustained delivery fluocinolone acetonide vitreous inserts provide benefit for at least 3 years in patients with diabetic macular edema. Ophthalmology. 2012;119:2125-2132.
23. Moisseiev E, Waisbourd M, Ben-Artsi E, et al. Pharmacokinetics of bevacizumab after topical and intravitreal administration in human eyes. Graefes Arch Clin Exp Ophthalmol. 2014;252:331-337.
24. Boyer DS. Dexamethasone intravitreal implant for treatment of diabetic macular edema in vitrectomized patients. Retina. 2011;31:915-23.