Updates on the Clinical Trials in Diabetic Macular Edema.

In this era of evidence-based medicine, significant progress has been made in the field of pharmacotherapeutics for the management of diabetic macular edema (DME). A. number of landmark clinical trials have provided strong evidence of the safety and efficacy of agents such as anti-vascular endothelial growth factors for the treatment of DME. Decades of clinical research, ranging from the early treatment of diabetic retinopathy study to the present-day randomized clinical trials (RCTs) testing novel agents, have shifted the goal of therapy from preventing vision loss to ensuring a maximum visual gain. Systematic study designs have provided robust data with an attempt to optimize the treatment regimens including the choice of the agent and timing of therapy. However, due to a number of challenges in the management of DME with approved agents, further studies are needed. For the purpose of this review, an extensive database search in English language was performed to identify prospective, RCTs testing pharmacological agents for DME. In order to acquaint the reader with the most relevant data from these clinical trials, this review focuses on pharmacological agents that are currently approved or have widespread applications in the management of DME. An update on clinical trials presently underway for DME has also been provided.

INTRODUCTION

Diabetic retinopathy (DR) is a leading cause of acquired blindness among young adults and the working-age population in developed countries.12 The prevalence of diabetes for all age-groups worldwide was estimated to be 2.8% in 2000 and 4.4% in 2030. The total number of individuals with diabetes is projected to rise from 171 million in 2000 to 366 million in 2030.3 Although severe vision loss may occur due to proliferative DR, the most common cause of visual loss in these patients is diabetic macular edema (DME).45 The results of the Wisconsin epidemiologic study demonstrated that the incidence of DME over the 10-year period was 20.1% in the younger-onset group, 25.4% in the older-onset group taking insulin, and 13.9% in the older-onset group not taking insulin.6

Due to the estimated rise in the number of diabetic patients, the need for ophthalmic care of patients is likely to exponentially increase in the future. In this era of evidence-based medicine, the ophthalmological management of patients with diabetes is based on the results of well-designed epidemiological studies and clinical trials. Results of landmark multicenter studies, such as the diabetes control and complications trial and the UK Prospective Diabetes Study, support that control of blood glucose, blood pressure, and serum lipids reduce the incidence and severity of DR in patients with type 1 or 2 diabetes.789

Studies on the pathophysiology of DME suggest a crucial role for vascular endothelial growth factor (VEGF) and inflammatory cytokines in disease development.101112 Such information has led to the evaluation of intravitreal corticosteroids and anti-VEGF agents in multicenter randomized clinical trials (RCTs) for the management of DME. In this review, the highlights of pioneering prospective clinical trials in DME (phase 2 and 3), such as the Early Treatment Diabetic Retinopathy Study (ETDRS), are briefly discussed. In addition, the results of various landmark RCTs that have evaluated therapeutic agents currently approved for the treatment of DME are comprehensively reviewed.

HIGHLIGHTS OF LASER TREATMENTS FOR DIABETIC MACULAR EDEMA

Three decades ago, a large multicenter RCT, the ETDRS, was launched to evaluate the role of laser therapy for preventing vision loss in DME (n = 3928). This study enrolled a wide variety of patients, ranging from those with no retinal thickening and best-corrected visual acuity (BCVA) ≥20/40 to those with retinal thickening (defined as clinically significant macular edema [CSME]) and vision ≤20/200.13 According to the results of this study, focal/grid laser therapy reduced the risk of moderate vision loss by 50% among patients with CSME at the 3-year follow-up visit.13

The Protocol A of DRCR.net (n = 263) demonstrated the efficacy of modified technique of ETDRS direct focal/grid laser photocoagulation for DME. Subsequently, another study by the DRCR.net (Protocol B) showed that 14% of patients treated with modified focal/grid laser gained ≥15 letters, whereas approximately 18% lost ≥15 letters at 2 years follow-up.1415 At the 2-year primary outcome in Protocol B (n = 693), the BCVA gain in the laser arm (+1 ± 17 letters) was significantly greater than the steroid arms (−2 ± 18 letters in the 1 mg triamcinolone arm; P < 0.02 and −3 ± 22 letters in the 4 mg triamcinolone arm; P < 0.002).1516

The ETDRS established focal/grid laser as a standard of care for DME.131718 However, a significant number of patients treated with a laser would continue to lose vision leading to suboptimal treatment outcomes.1518 Furthermore, results from recent studies employing microperimetry to assess retinal function suggest that laser therapy may be associated with worsening of macular function that is not detectable with routine BCVA testing.19 These results have led to the development of new therapeutic approaches based on our increasing understanding of the pathogenesis of DME [Figure 1]. Depicts the timeline of events in the development of newer therapeutic strategies for DME.

Figure 1

Timeline of major milestones in the field of pharmacotherapeutics for diabetic macular edema. In the figure, a number of landmark randomized clinical trials that have brought a paradigm change in the management of diabetic macular edema and led to the approval of therapeutic agents have been chronicled (based on the publication dates of the primary outcome manuscript in MEDLINE®, United States National Library of Medicine)

ANTIVASCULAR ENDOTHELIAL GROWTH FACTOR AGENTS

Ranibizumab

Ranibizumab (Lucentis®, Genentech Inc, San Francisco, CA, USA) (RBZ) is a 48 kDa recombinant humanized monoclonal antibody fragment that binds to all the isoforms of human VEGF-A.2021 RBZ molecule lacks Fc region, which allows for shorter systemic circulation and faster clearance.20 A number of RCTs have demonstrated that intravitreal RBZ reduces DME and sustainably improves vision. RBZ has thus replaced laser therapy as the standard of care for DME.

The Ranibizumab for Edema of the mAcula in diabetes-2 (READ-2) study (n = 126) was a pioneering RCT that randomized patients 1:1:1 to receive 0.5 RBZ, laser, or both.22 The results of this study provided early evidence of favorable bioactivity of RBZ in DME (BCVA gain of 7.4 letters in the RBZ arm compared to 0.5 letters in the laser arm at month 3). In addition, the study demonstrated that combining focal/grid laser with RBZ may help in decreasing the frequency of injections needed to control edema for at least 2 years.23 Three years extension of the READ-2 study revealed that monthly follow-up and aggressive retreatment with RBZ between months 24 and 36 results in sustained reduction central subfield thickness (CST) and improvement in BCVA.24 READ-3 study (Protocol 3 with high dose RBZ) was a double-masked, multicenter RCT that evaluated two doses of RBZ (0.5 mg and 2 mg) (n = 152). The study results demonstrated that high-dose RBZ (i.e. 2 mg) did not show any additional benefits over 0.5 mg dose at the primary endpoint at month 6 (+7.01 in the 2 mg group vs. +9.43 letters in the 0.5 mg group; P = 0.161).25

Another phase 2, double-masked, sham-controlled RCT (RESOLVE study) (n = 151) evaluated 0.3 mg, and 0.5 mg RBZ in three monthly doses for DME. Thereafter, the treatment could be stopped or re-initiated based on protocol defined criteria, with an option of dose-doubling in the RBZ arm. More than 60% of eyes treated with the RBZ had ≥10 letters gain compared to 18% of eyes with sham at 1 year (P < 0.0001).26

An RCT by the DRCR.net (Protocol I) evaluated the effect of 0.5 mg RBZ with prompt or deferred (>24 weeks) focal/grid laser and 4 mg intravitreal triamcinolone with a prompt laser for DME (n = 691).27 At 2-year follow-up, this study demonstrated that RBZ with a prompt or deferred laser is more effective than sham + prompt laser in improving BCVA and optical coherence tomography (OCT) outcomes. Approximately, half of the eyes treated with RBZ with either prompt or deferred laser had substantial BCVA improvement (≥10 letter), and one-third had ≥15 letter gain.28 At 3-year follow-up visit, it was found that prompt focal/grid laser treatment combined with RBZ was possibly worse than deferred laser with respect to BCVA outcomes (6.8 vs. 9.7 letters; P = 0.02).29 Expanded 5-year results of the Protocol I demonstrated that mean difference in BCVA change from baseline was 2.6 letters lesser in the RBZ + prompt laser treatment group compared to RBZ + deferred laser treatment group. Another relevant conclusion of the study was that more than half of eyes in which laser treatment was deferred may avoid additional laser for at least 5 years. However, these eyes may require a higher number of injections to achieve these results when following the protocol.30

RESTORE31 and REVEAL32 studies were phase 3 RCTs conducted with an aim to demonstrate the superiority of RBZ over laser therapy. In the RESTORE study (n = 345), BCVA gain was highest in the RBZ monotherapy arm at the primary endpoint of month 12 (+6.1 vs. +0.8 letters in laser arm; P < 0.001).31 Similar results were echoed by the REVEAL study (n = 396) (+5.9 letters in RBZ monotherapy arm vs. +1.4 letters in laser arm at month 12; P < 0.001).32 However, combining laser with RBZ did not provide additional benefit with respect to visual outcomes in both studies. The 3-year extension of the RESTORE study concluded that RBZ was effective in improving BCVA and CST outcomes with declining number of injections over 3 years using pro re nata (PRN) approach.33

RIDE and RISE were pivotal Phase 3 multicenter, RCTs that led to the approval of RBZ by the United States Food and Drug Administration (US FDA) in 2012 (n = 377 in RISE and n = 382 in RIDE) [Figure 1].34 The studies compared two doses of monthly RBZ (0.3 mg and 0.5 mg) to sham injection in patients with DME.34 In both the studies, significantly higher number of patients treated with RBZ gained ≥15 letters at month 24 compared to sham-treated group (44.8% vs. 18.1% in RISE; P < 0.0001, and 33.6% vs. 12.3%; P < 0.0001 in RIDE). The results demonstrated that the response to intravitreal inhibition of VEGF with RBZ was rapid and substantial. Patients treated with RBZ were noted to have higher rates of DR improvement on the ETDRS severity scale and were less likely to develop proliferative disease.35

In the extension phase of the RISE and RIDE studies (n = 759), the protocol allowed for patients in the sham control group to cross over and receive monthly RBZ injections in the 3rd year.36 The 36-month outcomes demonstrated that the rapid and sustained efficacy of RBZ in patients with DME is maintained over an additional 3rd year of continued monthly treatment. The group with delayed initiation of RBZ therapy gained fewer letters compared to the groups initially randomized to receive RBZ (+4.7 vs. +10.6 letters in the 0.3 mg RBZ arm). This suggests that chronic retinal edema (for an average of 4.5 years prior to RBZ therapy) may result in irreversible loss of vision.36 On the other hand, in the RESTORE extension study, the prior laser monotherapy group that crossed over to RBZ demonstrated similar BCVA gain compared to the prior RBZ (+6.0 vs. +6.7 letters). However, the authors concluded that the RESTORE study was not powered to test this hypothesis.33 Consistent with the month-24 outcomes, patients receiving RBZ in the RISE and RIDE extension studies were more likely to experience improvements in DR severity and less likely to develop the proliferative disease in the extension study. The sham crossover group also demonstrated improvements in retinopathy severity after crossover to RBZ therapy.36

In various clinical trials, continuous monthly dosing of RBZ has been recommended. While this regimen may optimize its efficacy, it may not be practical in the real world. Thus, RCTs have been designed to evaluate the long-term efficacy of PRN dosing of RBZ in DME. In the DRCR.net study, significant BCVA benefits were observed with a median of 8–9 RBZ injections in the 1st year, 2–3 injections in the 2nd year, 1–2 injections in the 3rd year, 0–1 in the 4th year, and no injection in the 5th year in the RBZ + prompt laser and RBZ + deferred laser arms, respectively. Thus, 54% and 45% of eyes during year 4, and 62% and 52% of eyes during year 5 received no injections in the two arms, respectively.30 Similarly, in the RESTORE extension study, patients in the prior RBZ groups were able to maintain the initial BCVA gains achieved at month 12 to months 24 and 36 with individualized RBZ treatment.33 Significant improvements in BCVA and OCT outcomes were observed with a mean of 3.9 and 2.9 injections in the year 2 and year 3 in the RBZ arm of the RESTORE extension study. Based on these results, future clinical trials are being designed using the PRN dosing of RBZ in order to optimize the cost-effectiveness and reduce the treatment burden.37

In addition to improvements in BCVA and CST values, treatment with RBZ has also shown to improve the retinal function in patients with DME. In the LUCIDATE study (n = 33), treatment of patients with center-involving DME with monthly RBZ was associated with an improvement in contrast threshold, retinal sensitivity on microperimetry, and amplitudes and implicit times on electrophysiology.38 In a DRCR.net study (Protocol J) (n = 345), 14-week effects of RBZ was evaluated in eyes receiving focal/grid laser for DME and pan-retinal photocoagulation (PRP) for proliferative DR. Laser-treated eyes received 0.5 mg RBZ at baseline and week 4. The results of this study demonstrated that addition of 2 injections of RBZ improved BCVA outcomes in eyes receiving PRP.39 Another recently published RCT by Ferraz et al. (n = 30) demonstrated that RBZ in combination with PRP may be an effective treatment in eyes with nonhigh risk proliferative diabetic retinopathy with DME.40

Based on the extensive evaluations of RBZ, this agent is regarded as highly efficacious with an overall acceptable safety profile for use among patients with DME. A majority of modern clinical protocol designs testing novel agents for DME consist of RBZ as the active comparator.

Aflibercept

Aflibercept (AFL) (Eylea, Regeneron Pharmaceuticals, Inc., Tarrytown, NY, USA) is a 115 kDa recombinant fusion protein consisting of portions of human VEGFR-1 and VEGFR-2 fused to the Fc domain of human immunoglobulin G1 domain.41 AFL binds all isoforms of VEGF-A, VEGF-B, and placental growth factor.4142 The drug is currently approved for the treatment of neovascular age-related macular degeneration, macular edema associated with vein occlusion, and DME.

A phase 2 RCT, the DA VINCI study (n = 221), was a landmark trial evaluating intravitreal injections of AFL for DME.43 The study compared four different doses and dosing regimens of intravitreal AFL versus focal/grid laser therapy in eyes with center-involved DME. At the month 12 follow-up, the improvement in BCVA ranged from 9.7 to 13.1 letters across the four AFL arms compared to a change of −1.3 letters of the laser group (P < 0.0001). The DA VINCI study demonstrated that intravitreal AFL was very effective in improving the BCVA and CST values compared to laser. AFL was well tolerated, and most common ocular adverse events were typical of those associated with intravitreal injections. The proportion of eyes that gained ≥ 15 letters was significantly greater in the AFL arms than the laser arm (except 2 mg every 8 weeks arm) at month 12 (P < 0.0031). However, it is important to note that the study was not adequately powered to discern the differences with regard to efficacy among the AFL treatment groups.44

Results of two parallel phase 3, double-masked, active control RCT of AFL (VIVID-DME and VISTA-DME studies) led to the approval of this agent for the treatment of DME by the US FDA (n = 872).45 In these studies, eyes with DME were randomized to receive either 2 mg every 4 weeks or 2 mg every 8 weeks intravitreal AFL after 5 initial monthly doses, or focal/grid laser photocoagulation. At month 12, the mean change in BCVA (+12.5 vs. +0.2 letters; P < 0.0001) and CST values (−185.9 vs. −73.3 µm; P < 0.0001) were significantly greater with AFL compared to laser therapy. The efficacy was statistically similar in the 2q4 and 2q8 treatment arms. In addition, a significantly greater proportion of eyes treated with AFL demonstrated ≥2 step improvements in the ETDRS DR severity scale as compared to the laser group (33.8% vs. 14.3%; P < 0.01). The overall incidence of ocular and nonocular adverse events was similar across all treatment groups.45

The 100-week results of the VIVID and VISTA studies demonstrated sustained anatomical and functional benefits with intravitreal injections of AFL compared to the laser control arm. The mean BCVA gain was +11.5 and +11.4 letters in the AFL (2 mg every 4 weeks) arm in the VIVID and VISTA studies, respectively. On the other hand, BCVA improvement was only +0.9 and +0.7 letters in the laser arm in the two studies (P < 0.0001). Similar to RBZ, in addition to the reduction in DME, treatment with AFL resulted in significant long-term improvement in ETDRS DR severity compared to laser therapy (P < 0.0004).46

Bevacizumab

Bevacizumab (BCZ) (Avastin®, Genentech Inc., San Francisco, USA) is a recombinant humanized monoclonal antibody with a molecular mass of 149 kDa that effectively binds and inhibits all the isoforms of VEGF.47 Although BCZ has been approved for the treatment of metastatic malignancies such as colon cancer48 and ovarian cancer49 by the US FDA, it is being widely used off-label for the treatment of DME world over.

The safety and efficacy of BCZ were first evaluated in phase 2 multicenter RCT by a DRCR.net study (Protocol H) (n = 121).50 In this study, the patients were randomized to five arms focal laser at baseline, 1.25 mg BCZ at baseline and week 6, 2.5 mg BCZ at baseline and week 6, 1.25 mg BCZ at baseline and sham at week 6, or 1.25 mg BCZ at baseline and week 6 with a laser at week 3. Compared to the laser arm, patients receiving both 1.25 mg and 2.5 mg BCZ had a greater reduction in CST values at 3 weeks. No significant functional or anatomical differences were noted between the two doses of BCZ in short-term. While the injections were generally considered safe, a concern of cerebrovascular and cardiovascular adverse events was highlighted by the authors.50 Following the DRCR.net study, a number of prospective trials were conducted to evaluate the role of BCZ for the treatment of refractory DME (DME was considered refractory if it was unresponsive to previous macular laser therapy, the standard of care at that time). In an RCT by Ahmadieh et al. (n = 115), a significant reduction in CST was observed in the BCZ arm compared to placebo (−95.7 vs. +34.9 µm; P = 0.012). Addition of triamcinolone did not provide significant additive effect to BCZ.51 Similar results were echoed by other similar trials comparing BCZ to either laser or triamcinolone for the treatment of DME.525354

The bevacizumab or laser therapy (BOLT) study was a prospective RCT evaluating the efficacy of BCZ in patients with center-involving DME (n = 80).55 The study randomized patients into two groups. In one group, the patients received macular laser therapy at baseline, and retreatment at weeks 16, 32, and 48 based on ETDRS guidelines. In the second group, patients received BCZ at baseline, weeks 6 and 12, and further retreatment was guided by an OCT-based protocol (with a maximum of 9 injections in 12 months). At month 12, patients in the BCZ arm gained significantly higher ETDRS letters on BCVA testing compared to the laser arm (+8 vs. −0.5 letters; P = 0.0002). 31.0% patients gained ≥10 letters in the BCZ arm compared to 7.9% in the laser arm (P = 0.01). Decrease in the CST was −130 ± 122 µm in the BCZ arm compared to −68 ± 171 µm in the laser arm (P = 0.06). At month 12, no significant systemic adverse event data was noted in the BCZ arm.55 The month 24 study results demonstrate that the gain in BCVA in the BCZ arm was significantly higher compared to the laser arm (+9 vs. +2.5 letters; P = 0.005). The median number of BCZ injections required during the 24 month period was 13. The BOLT study supported the longer-term use of BCZ for treatment of DME.56

BCZ is the most commonly used anti-VEGF agent for the treatment of DME worldwide and is preferred due to its affordability. However, the use of BCZ is still off-label. In a first of its kind, the DRCR.net conducted a large, prospective RCT that compared the three anti-VEGF agents, BCZ, RBZ, and AFL for the treatment of DME (Protocol T).57 This study was sponsored by the National Eye Institute, and is considered as the most impactful study for DME in the modern era.

In the Protocol T of DRCR.net, 660 patients with center-involved DME were enrolled at 89 clinical sites. Patients were randomized to receive 2 mg AFL (n = 224), 1.25 mg BCZ (n = 218), or 0.3 mg RBZ (n = 218) every 4 weeks according to the protocol-specified criteria. The study results up to month 12 have been published thus far. At month 12, there was a significant improvement in the BCVA compared to baseline in all the three arms (mean change of + 13.3 in AFL, +9.7 in BCZ and +11.2 in RBZ arms). Although the BCVA gain appeared to be higher in the AFL arm, it was not considered to be clinically meaningful in the overall analysis. On the other hand, when adjusted for the baseline BCVA, AFL might have performed relatively better than BCZ and RBZ. When the initial BCVA loss was mild (20/40–20/32), there were no significant differences among the study groups (P > 0.50). However, when the initial BCVA score was <69 letters (approximately ≤20/50), BCVA improvement was approximately 4 lines (18.9 letters) with AFL, 3 lines (14.2 letters) with RBZ and 2.5 lines (11.8 letters) with BCZ (P = 0.0003 for AFL vs. RBZ and P = 0.0001 for AFL vs. BCZ). The reduction in CST was greater with AFL and RBZ compared to BCZ. In addition, a smaller percentage of the patients on AFL required laser treatment for persistent DME compared to BCZ and RBZ.57

CORTICOSTEROIDS FOR DIABETIC MACULAR EDEMA

Inflammation has an important role in the pathogenesis of DME.12 The use of intravitreal corticosteroids for the treatment of DME has been extensively evaluated, due to their ability to inhibit proinflammatory cells and leukostasis,58 inhibit the expression of prostaglandins, proinflammatory cytokines, and VEGF,59 and to enhance the barrier function of vascular tight junctions.6061 While anti-VEGF agents appear to be superior over other available therapies for DME, intraocular steroids may be considered in a certain subset of patients with pseudophakia, persistent suboptimal response to anti-VEGF agents, or systemic vascular comorbidities that preclude the use of anti-VEGF agents.

Triamcinolone acetonide

The clinical efficacy and safety of intravitreal triamcinolone acetonide injections have been evaluated in a number of trials.6263 However, its use is limited by a high number of ocular adverse events such as progression of cataract and elevation of intraocular pressure (IOP). As described in the preceding sections, Protocol B of the DRCR.net studies demonstrated the superiority of focal/grid laser therapy over intravitreal triamcinolone.15 However, at 3 years, approximately 33% receiving 4 mg triamcinolone showed a rise of >10 mm Hg IOP and the 3-year cumulative probability of cataract surgery was 83% in the 4 mg group compared to 31% in the laser treatment arm. Similarly, in the Protocol I of the DRCR.net, the results showed that the BCVA among patients in the TA + laser group was 1.5 letters worse compared to the sham + laser group. Higher number of elevated IOP and cataract surgeries was reported in the TA group.28

Due to its poor adverse effect profile, intravitreal triamcinolone is no longer the preferred therapy for treatment of DME. Therefore, recent study protocols for DME avoid the use of triamcinolone.

Dexamethasone implant

Sustained-release corticosteroids have been developed to reduce the need for frequent intraocular injections. The dexamethasone implant consists of micronized dexamethasone in a biodegradable copolymer of polylactic-co-glycolic acid which releases the potent corticosteroid into the vitreous over a period of approximately 6 months.64 The implant (Ozurdex®, Allergan, Inc., Irvine, California, USA) is placed in the vitreous cavity in an office procedure using a single-use applicator system with a 23-gauge needle.

The dexamethasone implant has been evaluated in phase 2 studies for its safety and efficacy in the treatment of persistent or diffuse DME.6566 In the dexamethasone drug delivery system Phase II Study (n = 171), patients with DME ≥3 months duration were randomized to receive either 700 or 350 µg dexamethasone implant or observation. The study results demonstrated that 33.3% patients had an improvement of ≥10 letters in BCVA compared to 12.3% in the observation group (P = 0.007) at day 90. There was a significant improvement in CST values and fluorescein leakage compared to the observation group.65 The Ozurdex PLACID study evaluated the role of dexamethasone implant in combination with laser therapy (at month 1) compared to laser therapy with sham for diffuse DME (n = 253). The study protocol allowed the patients to receive up to 3 additional laser treatments and 1 additional dexamethasone implant or sham. In this study, the mean improvement in BCVA was greater in the dexamethasone + laser group compared to the sham + laser at month 12 (+7.9 vs. +2.3 letters; P < 0.013). Dexamethasone group demonstrated a larger reduction in the area of diffuse leakage on fluorescein angiography.66

The US FDA approval of the Dexamethasone Implant for DME was based on the study results of two large, sham-controlled, masked, prospective RCT, the MEAD study (n = 1048).67 This study demonstrated the long-term efficacy of dexamethasone implant in the treatment of DME without the need for monthly injections. Patients were randomized 1:1:1 to receive 0.7 mg or 0.35 mg dexamethasone implant, or sham procedure. Retreatment was performed if the patients were eligible based protocol-defined criteria no more often than every 6 months. At the end of 3 years, 22.2% patients receiving 0.7 mg implant achieved BCVA gain of ≥15 letters from baseline (primary efficacy endpoint for US FDA) with a mean of 4.1 treatments compared to 12.0% with a mean of 3.3 sham procedures (P ≤ 0.018). The mean reduction of CST was −111.6 µm with the 0.7 mg dose implant compared to −41.9 µm with sham (P < 0.001). Compared to other steroids, dexamethasone implant appeared to have a lower rate of adverse effects. The incidence of cataract was 67.9% in the 0.7 mg group and was higher with longer duration of exposure to the implant. Cataract surgery was performed in 59.2% of the phakic eyes. 27.7% patients had an elevation of ≥10 mm Hg IOP from baseline in the 0.7 mg group that was adequately controlled with topical medications. The overall incidence of adverse events adjusted for treatment exposure time was similar among the treatment groups.67

In an attempt to compare the efficacy of dexamethasone implant with intravitreal anti-VEGF agents, the BEVORDEX study was conducted in patients diagnosed with center-involving DME (n = 61).68 Head-to-head comparison between the two agents was performed by randomizing patients to two arms one arm receiving BCZ every 4 weeks and the other receiving dexamethasone implant every 16 weeks in a PRN regimen. Although similar rates of BCVA improvement were observed (improvement of ≥10 letters in 41% with the implant versus 40% with BCZ; P = 0.83) compared to BCZ, more implant-treated eyes lost vision due to the development of cataract. BCZ-treated eyes received a mean of 8.6 injections compared to 2.7 injections in the dexamethasone implant group.68 Another study has evaluated the safety and efficacy of dexamethasone implant with RBZ in patients with DME (n = 363). Patients were randomized to receive either 0.7 mg implant at baseline, month 5 and month 10, or 0.5 mg RBZ at baseline, and PRN every 4 weeks thereafter. Thus far, only partial study results are available.69

Due to an increasing number of vitrectomy procedures among patients with DR, there is a need for pharmacologic therapy with a long duration of action in such vitrectomized eyes. In an open-label, prospective multicenter study, the Ozurdex CHAMPLAIN study, treatment with single intravitreal injection of dexamethasone implant resulted in reduction of CST values (−65 µm at week 26; P = 0.004) and improvement of BCVA (+3.0 letters at week 26; P = 0.046) in vitrectomized eyes with DME.70 In another prospective RCT, dexamethasone implant showed favorable results in preventing and treating DME for 6 months in pseudophakic patients following a single injection administered at the time of cataract surgery (n = 16).71

In summary, patients treated with dexamethasone implant may achieve statistically significant and clinically meaningful visual improvements over long-term. These data support the use of dexamethasone implant in the management of patients with DME.

Fluocinolone acetonide implant

Fluocinolone acetonide implant (Retisert®, Bausch and Lomb, Rochester, New York, USA) was a nonbiodegradable implant containing 0.59 mg of the drug. Following surgical implantation, the implant is designed to release fluocinolone for over 2 years. A phase 3 prospective RCT evaluating Retisert for the treatment of refractory DME (n = 196), randomized patients 2:1 to receive 0.59 mg implant or standard of care. The study reported improved visual and anatomic outcomes. However, high rates of cataract progression (91% incidence of cataract surgery among phakic eyes) and elevated IOP (61.4% patients with IOP ≥30 mm Hg).72 There were no registration trials for Retisert to be indicated for DME in the US. Given the incidences of cataract and ocular hypertension, Retisert may not be best for use among patients with DME.

A next-generation fluocinolone acetonide insert, Iluvien® (Alimera Sciences, Alpharetta, GA, USA), has been developed as a nonbiodegradable, sustained-release device containing 0.19 mg drug. Safety and efficacy of Iluvien in DME has been tested in a phase 2 RCT, the FAMOUS study (n = 37). Patients with persistent DME (despite ≥1 focal/grid laser) received either low-dose (0.2 µg/day) or high-dose (0.5 µg/day) insert. The mean change in the BCVA was 7.5 and 5.1 letters with the high-dose and low-dose insert, respectively. There was no risk of elevated IOP with the low-dose insert in the study population. The results demonstrated that fluocinolone insert provides excellent sustained-release of the drug for control of DME over 1 year.73 36-month data of the study showed that both high-dose, as well as low-dose inserts provided comparable and stable peak levels of fluocinolone in the aqueous (≥1 ng/ml).74

Iluvien has been recently approved by the US FDA for the treatment of DME in patients who have not experienced a rise in IOP with steroids. The approval was based on two large, parallel, double-masked, sham injection-controlled, phase 3 RCTs, the FAME A, and FAME B studies (n = 953).75 Patients were allowed to receive retreatment with study drug or sham after 1 year. Significantly, higher number of patients gained ≥15 letters BCVA in both high-dose and low-dose group compared to sham (28.7% and 28.6% vs. 16.2%; P = 0.0002) at month 24. The mean improvement in BCVA was 5.4 and 4.4 letters in the high- and low-dose groups, respectively, which was significantly higher than the sham group at month 24 (1.7 letters; P > 0.016). The incidence of glaucoma was 7.6% in the high-dose and 3.7% in the low-dose group.75 Beneficial effects were observed at month 36 with 28.7% and 27.8% patients demonstrating ≥15 letter gain in the high- and low-dose groups, respectively.76 Careful subgroup analysis revealed that more number of patients with chronic DME (≥3 year duration) gained ≥15 letters compared to nonchronic DME (34.0% vs. 22.3% in 0.2 µg/day group).77 Further, subgroup comparison of patients receiving 0.2 µg/day insert prior to cataract surgery with those undergoing cataract surgery after implantation showed that the BCVA results were similar in both the subgroups. However, a higher number of patients who underwent cataract extraction after implantation experienced ≥15 letter BCVA gain compared to those who had cataract surgery before the implant (35.1% vs. 29.3%). This post-hoc analysis supports the use of fluocinolone insert in the management of DME in phakic as well as pseudophakic patients.78

Thus, the next-generation fluocinolone acetonide insert appears to be a promising long-term therapeutic strategy for DME patients with potential advantages of lack of repeat injections, cost effectiveness, and multiple clinic visits.

FUTURE DIRECTIONS

Given the complex nature of the disease and the large scale proportions of visual debility, there is a large unmet need for newer, more efficacious treatment options for DME. Novel agents are being explored for their safety and efficacy in DME. One such study, the READ-4 study, is being launched comparing RBZ and tocilizumab (TCZ), an interleukin-6 inhibitor, for the treatment of DME. The study aims to randomize patients to receive RBZ, TCZ or a combination with primary endpoint analysis at month 6. The enrollment for the study is expected to begin in the first quarter of 2016.79 Another agent, ASP8232, belonging to a novel class of vascular adhesion protein-1 inhibitors, is being evaluated in phase 2 multicenter RCT (the VIDI study) for the treatment of DME. Administered orally, the safety and efficacy of ASP8232 + sham are being compared to ASP8232 + RBZ and placebo + RBZ. With an enrollment target of 84 patients, the primary completion date is estimated to be April 2016.80 A novel class of drugs, insulin-like growth factor (IGF-1) inhibitors, is also being evaluated for its safety and efficacy in patients with DME. Teprotumumab (RV001) is an IGF-1 inhibitor administered as an intravenous infusion and is being evaluated in an open-label phase 1 study across three centers in the United States.81 A number of other novel agents such as abicipar pegol (designed ankyrin repeat protein),82 ALG-1001 (an anti-integrin oligopeptide),83 among others are being assessed in well-designed multicenter RCTs.

In the next few years, with level 1 evidence coming from large well-designed clinical trials, a number of novel agents may be introduced in the clinics for the management of DME. It is hoped that the outcome of patients with DME will continue to improve as newer pharmacotherapeutic agents are identified.

CONCLUSIONS

The previous standard of care for patients with DME has focused on the prevention of further deterioration in vision using a macular laser, with very few patients experiencing any subsequent gain in the BCVA. Anti-VEGF therapy and steroids offer substantially improved outcomes and have set new standards of care for patients with DME. In addition, therapy with VEGF inhibitors may also prevent or slow the worsening of retinopathy status and the need for PRP in a subset of patients with DME. Given their long-term availability of drugs and relatively acceptable risks of cataract and ocular hypertension, sustained-release devices of corticosteroids have now become a therapeutic choice for only a selected group of patients.

In the future, development of novel pharmacologic agents and combination therapies may help to address the concerns of frequent injections of anti-VEGF agents and decrease the burden of therapies.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.