Diabetic retinopathy clinical research retina network protocols in the management of diabetic macular edema :Mohamed Ashraf, Michael Gilbert, Abdulrahman Rageh, Ahmed Souka, Jennifer K Sun, Egyptian Retina Journal

REVIEW ARTICLE
Year : 2021 | Volume : 8 | Issue : 2 | Page : 47--56

Diabetic retinopathy clinical research retina network protocols in the management of diabetic macular edema

Mohamed Ashraf¹, Michael Gilbert², Abdulrahman Rageh², Ahmed Souka³, Jennifer K Sun⁴,
¹ Beetham Eye Institute, Joslin Diabetes Centre, Beetham Eye Institute, Boston, USA; Department of Ophthalmology, Alexandria Faculty of Medicine, Alexandria, Egypt; Department of Ophthalmology, Harvard Medical School, Boston, Massachusetts, USA
² Beetham Eye Institute, Joslin Diabetes Centre, Beetham Eye Institute, Boston, USA
³ Department of Ophthalmology, Alexandria Faculty of Medicine, Alexandria, Egypt
⁴ Beetham Eye Institute, Joslin Diabetes Centre, Beetham Eye Institute; Department of Ophthalmology, Harvard Medical School, Boston, Massachusetts, USA

Correspondence Address:
Dr. Jennifer K Sun
Beetham Eye Institute, Joslin Diabetes Center, One Joslin Place, Boston, Massachusetts 02215
USA

Abstract

The Diabetic Retinopathy Clinical Research Network (DRCR) retina network was formed in 2002 through a United States National Eye Institute and National Institute of Diabetes and Digestive and Kidney Diseases-sponsored cooperative agreement with the objective of creating a collaborative network for multicenter clinical trials focusing on diabetic retinopathy (DR) and its associated complications. The DRCR Retina Network has initiated and completed 30 multicenter studies in over 350 clinical sites. The goals of the DRCR was to design, implement, and report clinical studies that would answer important questions related to clinical practice and management of DR as well as DME. Diabetic macular edema (DME) is the leading cause of vision loss in the working age population and until the turn of the century the treatment options were limited to macular laser. This review aims to summarize the major DME studies completed by the DRCR The current review covers the major clinical trials that have helped help establish the current standard of care in the management of DME including protocols A, B, I, T, U and V.

How to cite this article:
Ashraf M, Gilbert M, Rageh A, Souka A, Sun JK. Diabetic retinopathy clinical research retina network protocols in the management of diabetic macular edema.Egypt Retina J 2021;8:47-56

How to cite this URL:
Ashraf M, Gilbert M, Rageh A, Souka A, Sun JK. Diabetic retinopathy clinical research retina network protocols in the management of diabetic macular edema. Egypt Retina J [serial online] 2021 [cited 2023 Jun 2 ];8:47-56
Available from: https://www.egyptretinaj.com/text.asp?2021/8/2/47/355268

Full Text

The Diabetic Retinopathy Clinical Research Network (DRCR Retina Network, or DRCR.net) was formed in 2002 through a United States National Eye Institute and National Institute of Diabetes and Digestive and Kidney Diseases-sponsored cooperative agreement with the objective of creating a collaborative network for multicenter clinical trials focusing on diabetic retinopathy (DR) and its associated complications.[1] Up until the turn of the century, the treatment options for the management of diabetic macular edema (DME) were limited to macular laser as demonstrated by the Early Treatment DR Study (ETDRS) and intensive systemic control of glycemia, blood pressure, and cholesterol as evaluated by the Diabetes Control and Complications Trial and the United Kingdom Prospective Diabetes Study.[2],[3],[4] The goal of the new network was to design, implement, and report clinical studies that would answer important questions related to clinical practice and management of DR. The Network's approach was innovative in encouraging involvement of both community and academic-based practices from its inception. In addition, there was an emphasis on developing collaborations with industry support that could supplement federal funding while maintaining rigorous academic independence. Since its inception, the DRCR Retina Network has initiated and completed 30 multicenter studies in over 350 clinical sites.[1] This review aims to summarize the major DME studies completed by the DRCR.net [Table 1].{Table 1}

Macular Laser

A pilot study of laser photocoagulation for diabetic macular edema (Protocol A)

The ETDRS demonstrated the efficacy of focal/grid macular laser in preventing vision loss in eyes with clinically significant macular edema.[4],[5] Although effective, the focal technique required a careful assessment with fluorescein angiography to identify and treat leaking microaneurysms. An easier, modified macular grid (MMG) protocol was proposed to ameliorate the challenges of targeting focal laser to specific microaneurysms. Thus, the first study initiated by the DR clinical research network (DRCR.net) aimed to compare this MMG technique with modified ETDRS (mETDRS) technique in eyes with previously untreated DME at or involving the macular center.[6] These two techniques differ in density and location of the laser burns. MMG burns are distributed all over the macula in both areas with and without edema and microaneurysms were not directly targeted [Table 2]. This contrasts to the mETDRS technique which treated only areas of retinal thickening and leaking microaneurysms. Two modifications of the classic ETDRS technique were included in both MMG and mETDRS laser guidelines; laser burns were less intense requiring only graying of the retina and smaller in size (50 μg). Furthermore, while microaneurysms are targeted in the mETDRS technique, a change in color of the microaneurysm was not required.{Table 2}

The primary outcome of the study was change in central retinal thickening on optical coherence tomography (OCT) at 12 months of follow-up with change in visual acuity being the secondary outcome. The study included 323 eyes of 263 patients with a mean baseline VA of 20/32 and mean OCT central subfield thickness (CST) of 340 ± 123 μg (OCT3 Zeiss Meditec, TD-OCT). The CST decreased significantly at 12 months compared to baseline in both the mETDRS and MMG groups. The reduction in retinal thickening was more prominent in the mETDRS group, with a decrease of 88 μg compared to 49 μg in the MMG group (P = 0.02). Laser was effective in reducing CST regardless of the baseline CST, with eyes having a baseline CST >450 μg showing the most decrease at 12 months. Furthermore, at 12 months, approximately 23% in the mETDRS group versus 17% in the MMG had a normal CST.

There was no significant change in mean VA at 12 months compared to baseline in either group. Of note while the mean VA was not significantly different between the treatment groups at 12 months, 27%–32% of eyes had ≥5-letter while 5%–7% had ≥15-letter improvement. In the cohort with a baseline VA of 20/40 or worse, 63%–52% had a ≥5-letter gain while 16%–29% had a ≥15-letter improvement. These findings demonstrated that vision gain was frequent in eyes treated with macular laser. Study results also suggested that laser was equally effective in improving vision in eyes with mild versus more pronounced retinal thickening.

Steroids

A randomized trial comparing intravitreal triamcinolone acetonide and laser photocoagulation for diabetic macular edema (Protocol B)

Prior to this study, macular focal/grid laser was the standard of care in the management of DME as demonstrated by the ETDRS.[4],[5] Nonetheless, early case reports and series suggested potential benefit from intravitreous injections of corticosteroid in eyes with DME. By 2005, a Preferences and Trends Survey from the American Society of Retina Specialists showed that 91% of retina specialists had used intravitreal triamcinolone (TA) at some point during that year in patients with DME who had been nonresponsive to laser. Therefore, it became important to evaluate the safety and efficacy of this therapy compared to the standard of care at the time (laser).

The main objective for Protocol B was to evaluate the efficacy and safety of 1-mg and 4-mg intravitreal TA compared to focal/grid laser for the treatment of DME.[7] Patients with a VA of 20/40 or worse and a CST of >250 (TD-OCT) were recruited. Patients in the TA groups were re-treated at 4-month intervals if they had residual edema on OCT and their VA was 20/32 or worse. At 2 years, the mean number of treatments was 2.9 in the laser group, 3.5 in the 1-mg TA group, and 3.1 in the 4-mg TA group.

At 4 months, the mean VA was better in the 4-mg TA versus laser group (difference of + 3.8 letters, P < 0.001) and in the 4-mg TA versus 1-mg TA group (difference of + 3.6 letters, P = 0.001). However, beginning at the 16-month visit until the 2-year follow-up, the laser group showed greater visual acuity gains (+1 letter) compared to both the TA groups (−2-letter 1-mg TA and −3-letter 4-mg TA, P = 0.02 laser versus 1-mg TA and P = 0.002 laser versus 4-mg TA group). Significantly more eyes receiving TA had a 15-letter or more loss in VA (20% versus 14% in the laser group, P < 0.05). The OCT changes mirrored the changes in the VA, with 4-mg TA resulting in significantly greater decrease in CST at 4 months compared to the other two treatments and at 2 years, the laser group demonstrating a significantly greater decrease compared to the two TA groups. At 2 years, 53%, 34%, and 38% of eyes had a CST <250 in the laser, 1-mg TA, and 4-mg TA groups, respectively.

At baseline, 79% of eyes were phakic. Subsequently, 51% in the 4-mg TA group and 23% in the 1-mg TA group received cataract surgery prior to the 2-year visit compared to only 13% in the laser group (P < 0.001). A subgroup analysis limited to eyes that were pseudophakic at baseline demonstrated similar visual acuity changes in the three groups (+2 in the laser, +2 in the 1-mg TA, and −1 in the 4-mg TA groups). Of note, significantly more eyes in the TA groups (40% in the 4-mg TA and 20% in the 1-mg TA) compared to the laser group (10%) had either a ≥10-mmHg IOP elevation compared to baseline, an IOP of ≥30 at any visit, or initiation of IOP-lowering medications.

The lesser efficacy of TA on VA compared to laser was unlikely due to lens changes alone given that laser had significantly greater improvements in OCT at 2 years and that a subgroup analysis of eyes that were either pseudophakic at baseline or showed little to no changes in lens status at 2 years did not demonstrate a benefit for TA versus laser.

Unpublished data from the ETDRS reveal that the natural history of eyes with DME and a baseline VA of 20/32 or worse is an average decrease of 6 letters compared to baseline, with 43% of eyes losing 10 or more letters.[7] Based on these data, both the TA groups and the laser group fared better than the natural history of untreated eyes with DME. Protocol B provided further support for the idea that macular laser can result in substantial vision gain in many eyes with DME. An analysis of a cohort of eyes with similar demographics to Protocol B demonstrated that eyes receiving laser had a + 4-letter visual gain with 29% improving 10 letters or more in the ETDRS and a +6-letter gain with 37% improving 10 letters on more in Protocol A.[7]

Intravitreal ranibizumab or triamcinolone acetonide in combination with laser photocoagulation for diabetic macular edema (Protocol I)

The advent of anti-vascular endothelial growth factor (anti-VEGF) therapy provided yet another potential treatment alternative for eyes with DME. Protocol I aimed at evaluating the efficacy of 0.5-mg ranibizumab or 4-mg TA combined with focal/grid laser compared with focal/grid laser alone for DME.[8] We will describe the ranibizumab arm in the anti-VEGF section, focusing here on results from the TA group. Although TA monotherapy had previously been evaluated in Protocol B, combining TA with focal/grid photocoagulation had not been evaluated.[7] A TA and laser combination arm was included with the rationale that if the retina was less edematous, this might enhance the effect of laser therapy and lead to a more rapid visual improvement compared to the slower benefits seen with laser over time.

The study recruited patients with center-involved macular edema (CST >250 μg) and a VA of 20/32 or worse. Eyes assigned to this group could receive injections as frequently as every 16 weeks for persistent DME and were required to receive laser within 3–10 days of injection (prompt laser).

Overall, TA combined with laser did not result in superior outcomes compared to laser alone at years 1 and 2 and was inferior to outcomes achieved with anti-VEGF therapy. While mean VA in the TA group showed an initial improvement until week 24, this was followed by a decrease in mean VA thereafter. At years 1 and 2, VA gains were not significantly different between the TA group and the laser group (+1.1-letter and −1.5-letter difference at 1 and 2 years, respectively).[8],[9] These findings may have been related to the development of cataracts or the negative impact of cataract surgery on macular edema in the TA group eyes. A subgroup analysis in TA-treated eyes that were pseudophakic at baseline demonstrated that the visual acuity results were substantially better than for phakic eyes. Pseudophakic eyes treated with TA achieved similar visual gains compared to the ranibizumab group and significantly greater gains compared to the laser group (+1.6 letters greater).[8],[9]

OCT changes showed a steady decrease in CST in year 1, which was comparable to the ranibizumab groups and was significantly greater than the laser group. This was followed by an increase in CST in the 2nd year of follow-up, and at the 2-year visit, there was no longer a significant difference between the laser and TA groups (P = 0.37).[8],[9] At 2 years, the percentage of eyes with a CST >250 was 59% in the laser group and 52% in the TA group, which was not statistically different.[8],[9] Approximately 50% of eyes in the TA group had an IOP elevation >10 mmHg from baseline, IOP >30, or initiation of IOP-lowering medications at 1 or more visits during 2 years of follow-up. In addition, 59% required cataract surgery during the 2 years of follow-up.

A follow-up study looked at the 5-year outcomes in eyes randomized to TA at baseline and then switched to ranibizumab (57%) subsequently.[10] Whether phakic or pseudophakic, eyes receiving delayed ranibizumab did not achieve the same improvement in VA as those treated with ranibizumab from baseline. Furthermore, while eyes switched to ranibizumab demonstrated anatomic and visual improvements, these changes appeared slower and less robust compared to the changes observed in the first 6 months of treatment in the ranibizumab group.

Anti-Vascular Endothelial Growth Factor

Injection protocol

The DRCR Retina Network anti-VEGF retreatment algorithm consists of the following steps. Six monthly injections are administered unless vision is 20/20 or better and OCT CST is <305 for women and <320 for men on Heidelberg Spectralis, in which case injections can be withheld at the 4th or 5th month. Starting at the 6-month visit, anti-VEGF is withheld if VA or OCT CST has neither improved or worsened compared to the last 2 injections or if the DME has resolved on OCT. Improvement was defined as OCT CST ≥10% change or BCVA ≥5-letter change (approximately 1 line on a Snellen chart). Anti-VEGF is resumed if VA worsens (≥5 letters) in the setting of persistent DME or if OCT CST worsens (≥10% change).

There are slight differences in the injection protocols used in Protocols I and T.[11] In Protocol T, anti-VEGF was required for worsening of DME or worsening of VA in the presence of DME, whereas it was at investigator discretion in these cases in Protocol I. In Protocol I, focal/grid laser was administered at baseline to all eyes in the prompt group or to eyes meeting certain criteria after 24 weeks in the deferred laser group. In Protocol T, laser was deferred in all eyes for at least 24 weeks. Laser was administered using the mETDRS protocol and re-treatment was only allowed every 4 months. Whereas laser was required in eyes with persistent DME despite anti-VEGF therapy and treatable microaneurysms in Protocols I and T, laser was optional in Protocol V.

Intravitreal ranibizumab or triamcinolone acetonide in combination with laser photocoagulation for diabetic macular edema (Protocol I)

Since VEGF is a key mediator of vascular permeability, it was unknown if combined therapy with intravitreal anti-VEGF and macular laser could provide superior visual acuity improvement in eyes with DME. As previously described in this article, Protocol I evaluated the efficacy of 0.5-mg ranibizumab compared to TA + prompt laser and mETDRS laser alone.[8] It also aimed to answer the question as to when the ideal time was to administer focal/grid laser in patients receiving anti-VEGF (prompt versus deferred at 24 weeks).

At 1 year, both the ranibizumab-treated groups achieved significantly greater mean visual gains (+9 letters in the prompt and deferred groups) compared to both the laser (+3 letters) and TA groups (+4 letters), and these observations were sustained at 2 years.[8],[9] Visual acuity improvements achieved by both the ranibizumab groups at year 1 were sustained at 5 years despite a decreasing frequency of injections (8 and 9 in year 1, 2 and 3 in year 2, 1 and 2 in year 3, 0 and 1 in year 4, and 0 in year 5 in the prompt and deferred groups, respectively).[12] At 5 years, only 44% of eyes in the deferred laser group received focal/grid laser. Furthermore, at year 5, the percentage of eyes gaining >10 letters (+11% more, P = 0.04) and >15 letters (+10% more, P = 0.03) was significantly higher in the deferred laser compared to the prompt laser group.[12] At 5 years, more VA gain was achieved in the prompt laser group versus the deferred group (+9.8 vs. +7.2), although this trend was not statistically significant. This difference was mainly driven by the group with a baseline VA of 20/50 or worse and when evaluated separately showed significantly greater positive changes in VA in the prompt group compared to the deferred laser group (P < 0.001). In contrast, there were no significant differences in OCT changes between the two groups at 5 years (−8-μg difference between both the groups, P = 0.48) with approximately 35% of eyes having a CST >250 μg.[12]

Of note, in the deferred laser + ranibizumab group, only 44% received grid/focal laser and received significantly more injections (a median of 4 more over 5 years), and it was suggested that this difference might be a contributing factor to the greater visual gains. Another possibility is that laser may have had a damaging effect on the macula with a reduction in VA gains despite subsequent thinning.

A comparative effectiveness study of intravitreal aflibercept, bevacizumab, and ranibizumab for diabetic macular edema (Protocol T)

Prior to this study, three anti-VEGF agents were commercially available and used frequently for treatment of DME; ranibizumab, aflibercept, and bevacizumab. However, only ranibizumab and aflibercept were approved by the United States Food and Drug Administration for the treatment of DME. Given substantial differences in availability and cost, bevacizumab was compared to aflibercept and ranibizumab in order to evaluate the relative efficacy and safety of the three drugs.

This study compared the relative efficacy of intravitreous 2.0-mg aflibercept, 1.25-mg bevacizumab, and 0.3-mg ranibizumab for treatment of DME. The study reported that treatment with all three agents resulted in substantial vision gains. However, whereas eyes with better baseline VA (20/32–20/40) showed similar outcomes between the three medications, there were significant differences in visual outcomes between medications noted in eyes with baseline VA of 20/50 or worse.

Eyes with VA 20/50 or worse

At 1 year in the group with baseline VA 20/50 or worse, the mean improvement was 18.9 letters in the aflibercept group, 11.8 in the bevacizumab group, and 14.2 letters in the ranibizumab.[13] While the gains in the aflibercept group were significantly greater than the bevacizumab and ranibizumab groups (P < 0.001 and P = 0.003, respectively), there was no significant difference between the ranibizumab and bevacizumab groups (P = 0.21). This was achieved with a mean of 10 injections in the aflibercept group, 11 injections in the bevacizumab group, and 10 in the ranibizumab group. Aflibercept also had a greater effect on OCT changes with 70% of eyes having a CST <250 μg compared to 39% (P < 0.001 vs. aflibercept) in the bevacizumab and 56% in the ranibizumab groups (P = 0.02 vs. aflibercept).

In the second year of follow-up, aflibercept had superior 2-year VA outcomes compared with bevacizumab (P = 0.02) but was no longer superior to ranibizumab (P = 0.18).[14] With regard to changes in CST, in the 2nd year, there were no significant changes in the CST in the aflibercept or ranibizumab group compared to year 1. However, in the bevacizumab group, the CST decreased by an additional 42 μg, and at 2 years, the difference in CST change between the ranibizumab and bevacizumab groups was no longer significant (P < 0.001 at year 1 and P = 0.19 at year 2). In a post hoc analysis, the decrease in CST was mainly driven by the group receiving focal laser, suggesting a possible role in anatomic thinning of the retina in those eyes without a corresponding visual improvement.[15] The percentage of eyes with CST <250 was 75%, 46%, and 66% in the aflibercept, bevacizumab, and ranibizumab groups, respectively. The median numbers of injections in the 2nd year were 5, 6, and 6 in the aflibercept, bevacizumab, and ranibizumab groups, respectively, with more eyes in the bevacizumab group requiring focal/grid laser (64%) compared to the aflibercept (41%) and ranibizumab groups (52%).

A cost-effectiveness analysis based on average United States Medicare rates in 2015 (aflibercept: $1950, bevacizumab: $50, and ranibizumab: $1200) suggested that the cost of aflibercept and ranibizumab would need to decrease by 62% and 84%, respectively, to reach a cost-effectiveness threshold similar to bevacizumab during a projected 10-year period and concluded that these medications “are not cost-effective relative to bevacizumab for treatment of DME unless their prices decrease substantially.”[16]

Group with VA 20/32–20/40

At 1 year, in the group with baseline VA 20/32–20/40, the mean visual gains were 8.0 letters in the aflibercept group, 7.5 letters in the bevacizumab group, and 8.3 letters in the ranibizumab group. There were no significant differences between any of the groups. This was achieved with a median of 9 injections in each group.[13] At 2 years, there was still no statistically significant difference in visual outcomes between the three groups with a mean improvement of 7.8, 6.8, and 8.6 letters in the aflibercept, bevacizumab, and ranibizumab groups, respectively. However, while VA changes were not significant between any of the groups, OCT showed significantly more improvement in the aflibercept versus bevacizumab and ranibizumab versus bevacizumab groups at both year 1 and year 2 (P < 0.001 for both). There was no difference found in OCT change between the aflibercept and ranibizumab groups. However, it was noted that in eyes with better initial VA (20/32–20/40) and thicker CST (>400 μg), visual outcomes were worse for the bevacizumab group compared to the ranibizumab and aflibercept groups.[17] Over the 2 years, the cumulative number of injections was similar in all the three groups, with significantly less injections being received in year 2 compared to year.

Five-year extension study

A follow-up visit was conducted at 5 years for participants in Protocol T, 3 years after the end of the study. Thus, from years 2–5, patients were treated through routine care at the discretion of their retina specialist (AAO Retina Subspecialty Day, San Francisco, 2019). Treatment data from eye care visits during years 2–5 were retrospectively acquired. Unlike the maintenance of VA gains seen in the Protocol I cohort, results from this extension study suggest that VA declined after patients ended study participation in year 2. The cause of this decline cannot be determined from the study data, but this finding suggests that methods to optimize long-term VA in eyes with DME are still needed in routine clinical care.

Treatment for central-involved diabetic macular edema in eyes with very good visual acuity (Protocol V)

Since prior DME studies only included patients with a visual acuity of 20/32 or worse, it was unknown whether treatment should be started for patients with better vision and ci-DME. In addition, it was also unknown if deferral of treatment would result in visual loss.

This study included patients with baseline VA of 20/25 or better and ci-DME.[18] The study randomized participants into three groups: 2.0-mg aflibercept using the injection retreatment algorithm as previously described, focal/grid laser photocoagulation, or observation. Patients in the laser or observation group were required to receive aflibercept if they demonstrated a decrease of 10 letters on any visit (approximately >2 lines) or a 5–9-letter decrease (1–2 lines) on 2 consecutive visits.

The percentage of patients with at least a 5-letter decrease in VA was not statistically different between the three groups (16%–19%) with a mean change in VA of 0.9 letters in the aflibercept, 0.1 in the laser group, and − 0.4 letters in the observation group. At 2 years, the percentage of eyes with a VA of 20/20 or better was higher in the aflibercept group (77%) compared to the laser (71%) and observation groups (66%), but this difference was only significantly different between the aflibercept and observation groups (P = 0.03).

By 2 years, aflibercept was initiated in 25% and 34% of eyes in the laser and observation groups, respectively, with eyes initiating treatment requiring a median number of 7–9 injections. This is similar to the mean number of 8 injections required in the group initiated with aflibercept at baseline.

A post hoc analysis was conducted for Protocol V looking at eyes in the observation group that needed aflibercept during the course of the study.[19] Risk factors for needing aflibercept included a thicker CST at baseline (>300-μg Zeiss Stratus Equivalent), a DR severity level of moderately-severe NPDR or worse, and a history of anti-VEGF treatment in the nonstudy eye within 4 months of randomization. These eyes were twice as likely to receive aflibercept compared to eyes that did not have these risk factors.

Short-term evaluation of combination corticosteroid + anti-vascular endothelial growth factor treatment for persistent central-involved diabetic macular edema following anti-vascular endothelial growth factor therapy (Protocol U)

Data from Protocol T demonstrated that despite timely anti-VEGF therapy, a substantial percentage of eyes had residual macular edema at 1 and 2 years. Hence, the purpose of this study was to see whether combining anti-VEGF treatment with intravitreal steroid may help improve visual outcomes in those eyes. This study compared the effects of continued ranibizumab versus supplementing treatment with an intravitreous dexamethasone implant in a cohort of patients with persistent DME despite previous anti-VEGF treatment.[20] Persistent DME was defined as patients with a VA of 20/32 or worse and a CST above threshold (≥305 in females and ≥320 in males) despite having received at least 3 anti-VEGF injections in the previous 20 weeks prior to enrollment.

Recruited patients were enrolled in a 12-week run-in phase in which they received 3 ranibizumab injections at baseline, 4 weeks, and 8 weeks. After the run-in phase, they were re-evaluated and only eyes meeting baseline VA and OCT thresholds were randomized to receive dexamethasone + ranibizumab or to continue ranibizumab alone. Of the 236 eyes enrolled in the run-in phase, only 129 were randomized (54.6%). This suggested that timely q4 week therapy or continued therapy beyond 12 weeks may be considered prior to switching given the anatomic and visual improvements. Of note, while the study initially attempted to recruit only pseudophakic patients, due to slow recruitment phakic eyes were subsequently enrolled and represented 60%–50% of the total cohort in the combination and ranibizumab groups, respectively.

At 24 weeks, there was no significant difference in the VA gains in the combination group versus the ranibizumab group (+2.7 letters vs. +3.0 letters, P = 0.73). The percentage of eyes gaining or losing 10 or more letters was similar between both the groups. A subgroup analysis noted that compared to the ranibizumab group, the combination group had a +3.1-letter gain in the pseudophakic eyes and a −3.0-letter loss in the phakic group. However, this was still not significant. While there were no significant differences with regard to the VA changes, the combination group showed significantly greater anatomical improvements versus the ranibizumab group with 52% versus 31% having a CST below threshold. Furthermore, the mean change in CST was greater in the combination group compared to the ranibizumab group (P < 0.001). However, there were significantly more complications in the combination group with 29% of eyes (23% had an increase of 10 mmHg or more) versus 0% of eyes in the ranibizumab group having an increased IOP or initiating IOP-lowering medications at any of their visits.

The study concluded that while steroids may demonstrate anatomical improvements in patients with persistent DME, this was not associated with visual improvements and comes at an increased risk of elevated IOP. Furthermore, as demonstrated by the run-in phase, timely anti-VEGF therapy or continuing anti-VEGF may be worth considering prior to switching.

Incomplete Responders to Anti-Vascular Endothelial Growth Factor

Incomplete anatomic responders

Post hoc analyses of network studies have provided insight into outcomes for eyes with persistent DME despite anti-VEGF therapy. Persistent DME was defined as those eyes with a CST of 250 μg or more at all completed visits through 24 weeks.[21],[22] Chronic persistent DME was defined as eyes that had not achieved a CST <250 μg and a 10% or more reduction on at least 2 consecutive visits subsequent to the 24-week visit at any time point.[21],[22]

In Protocol I, 39.5% of eyes had persistent edema at 24 weeks. Of those eyes, 40.1% had chronic persistent edema at 3 years. Visual acuity improved by + 13 letters in the group that did not have chronic persistent DME versus + 7 letters in the group with persistent DME, P = 0.02. While the VA gains were less in the group with chronic persistent edema, the median visual acuity at 3 years was 20/32 which is comparable to the VA in the group without edema 20/25. Furthermore, 42.5% of eyes in the chronic persistent edema group gained 10 letters or more from baseline.

In Protocol T, a similar analysis was conducted. Through 24 weeks, persistent DME was more frequent in eyes with bevacizumab (65.6%) compared to both the aflibercept (31.6%) and ranibizumab groups (41.5%) (P < 0.001).[22] Eyes in which edema persisted chronically through 2 years were also more common in the bevacizumab (68.2%) compared to the ranibizumab (54.5%) and aflibercept groups (44.2%). Despite the presence of chronic persistent edema, 62%, 51%, and 45% of eyes gained 10 or more letters in the aflibercept, bevacizumab, and ranibizumab groups, respectively, at 2 years which was comparable to the 63%, 55%, and 66% in eyes without chronic persistent edema gaining 10 or more letters. This suggested that even eyes with residual edema at 24 weeks may potentially show anatomical improvements with continued therapy. Furthermore, a significant proportion of those eyes showed visual improvements comparable to eyes that did not have chronic persistent edema, with very few eyes losing vision.

Incomplete vision responders

While the previous analysis looked at worsening of vision and nonresponse from an anatomical point of view, another post hoc analysis evaluated visual improvement in patients showing poor visual response after 12 weeks.[11] While there was some association between VA gains at 12 weeks and final visual acuity, this relationship was weak for aflibercept, bevacizumab, and ranibizumab (R2 = 0.38, 0.29, and 0.26 for the 2-year change, respectively). In eyes having <5-letter gains at 12 weeks, 42%, 31%, and 47% had gained 10 or more letters from baseline at 2 years in the aflibercept, bevacizumab, and ranibizumab groups, respectively. This was associated with a median VA of 20/32, 20/32, and 20/25 in the three groups, respectively. In addition, 12-week CST response was not associated with 2-year VA changes.[11] This is consistent with previous DRCR.net studies which have demonstrated a poor correlation between VA and CST. A post hoc analysis of Protocol A was conducted to explore the association of OCT thickness and VA in eyes with DME before and after macular laser. The study demonstrated that OCT CST had a modest correlation (r = 0.36–0.52) with VA not just at baseline but at all follow-up time points (3.5, 8, and 12 months) after laser.[23] There were eyes with a CST <225 who had a VA 20/30 of worse, and in addition, there were eyes with a CST >400 who had a VA of 20/20 or better. It was estimated that for every 100-μg decrease in CST, visual acuity improves by 4.4 letters. While 27% of the variation in visual acuity could be predicted only by a CST, incorporating additional parameters to the linear regression model (age, A1c, and severity of fluorescein leakage) only improved this rate to 36%. In addition, the correlation between the change in visual acuity and the corresponding decrease in CST at 3.5 months was modest (0.44). The study also identified paradoxical changes in two cohorts of patients; one showed improving VA with increasing CST (7%–17%) while another demonstrated worsening VA with decreasing in CST (18%–26%).

These findings were reproduced in a post hoc analysis of Protocol T.[24] The study demonstrated a poor correlation between VA and CST at baseline (r = 0.36), 12 weeks (r = 0.24), 52 weeks (r = 0.31), and 104 weeks (r = 0.23). In addition, there was a poor correlation between change in CST thickness and change in VA (r = 0.33–0.36). It was estimated that for every 100-μg decrease in thickness, there was a 2.5–2.7-letter improvement in VA. However, only 12%–14% of the total variation in VA could be explained by CST changes. Similar to Protocol A, paradoxical changes were noted at 2 years with 7.4%–9.1% of eyes demonstrating worsening of VA with a decrease in CST while 5.1%–5.7% showed an improvement in VA with an increase in CST.

Predictors of response

Post hoc analyses for both Protocols I and T looked at 37 baseline systemic and local factors associated with VA and CST changes.[25],[26] In both studies, older age and worse baseline DR severity were identified as predictors of poor VA response, while in Protocol T, higher baseline A1c was identified, and in Protocol I, the presence of surface wrinkling was associated with worse VA gains.[25],[26]

For every 10 years of increased age, VA improvement decreased by 2.1 letters in Protocol T and 2.2 letters in Protocol I. In Protocol T, eyes with no prior PRP and DR severity less than severe NPDR had approximately 3-letter improvement compared to eyes with prior PRP while it was 4 letters in Protocol I.[25],[26] In Protocol T for every 1% increase in A1c levels, VA improvement was reduced by 1 letter.[25] In Protocol I, the absence of surface wrinkling was associated with an additional mean improvement of 4 letters.[11]

With regard to anatomic changes in Protocol T, African Americans had greater decreases in CST compared to white patients. In addition, eyes with baseline subretinal fluid (SRF) had significantly greater decreases in CST compared to those without SRF.[25] This was not noted in Protocol I which described greater deceases in CST in the presence of hard exudates within the 6-mm ETDRS grid.[11]

Diabetic Macular Edema and Proliferative Diabetic Retinopathy

The presence of PDR is important in the management of eyes with PDR. Both Protocol I and Protocol T have demonstrated that eyes with PDR who have received PRP achieve less visual gains compared to eyes with no PRP and DR less than severe NPDR.[25],[26] The treatment of PDR using PRP may itself induce DME. If DME and PDR are co-existent, this may affect treatment decisions as to whether PRP versus anti-VEGF should be used, since anti-VEGF is clearly first-line therapy for most eyes with visual impairment from DME.

Protocol S was a randomized control trial that aimed to compare the visual outcomes in patients with PDR receiving either PRP or 0.5-mg ranibizumab.[27] Eyes with baseline center-involved DME (22%) were allowed to receive ranibizumab from baseline if they were randomized to the PRP group. In eyes without baseline DME, significantly more eyes in the PRP group developed vision impairing DME compared to the ranibizumab group at 2 years (Kaplan–Meier estimates; 29% vs. 11%) and at 5 years (Kaplan–Meier estimates; 38% vs. 22%). Furthermore at 2 years, the ranibizumab group gained 2.8 letters compared to 0.2 letters in the PRP group (P < 0.001 for noninferiority).[27],[28] When stratified based on the presence of baseline DME, 2-year visual gains were greater but not significantly different in the ranibizumab group compared to the PRP group in the group with baseline DME (+7.9 vs. +1.9) compared to the group without baseline DME (+1.8 vs. −0.5). The 5-year results mirrored the 2-year results and demonstrated that baseline DME did not affect final visual outcomes in either group.[28] The 2-year results suggested that in eyes with baseline DME in whom anti-VEGF therapy is already planned, PRP may be unnecessary provided that the patient is compliant with follow-up and treatment recommendations. Furthermore, a post hoc analysis suggested that eyes with elevated A1c, high risk PDR (vs. nonhigh-risk PDR), and the presence of cystoid abnormalities within 500 μg of the macula (not meeting ci-DME treatment thresholds) were more likely to develop ci-DME when receiving PRP and should be observed carefully after commencing therapy.[29] From a cost-effectiveness standpoint, it was suggested that ranibizumab may be a cost-effective alternative to PRP only in patients with both ci-DME and vision PDR but not in those with PDR alone.[30]

Conclusion

In summary, DRCR.net protocols have provided results to guide the management and follow-up of patients with DME. In patients with excellent VA (20/25 or better) and center-involved macular edema, treatment can generally be deferred unless visual acuity worsens. In contrast, for most patients with vision of 20/32 or worse from DME, intravitreal anti-VEGF should be considered first-line therapy. In patients with a baseline VA of 20/32–20/40, anti-VEGF can be started, and regardless of whether aflibercept, bevacizumab, or ranibizumab is used, outcomes are likely to be similar. For patients with a VA of 20/50 or worse, although aflibercept provides superior anatomic and visual outcomes at 1 year compared to ranibizumab and bevacizumab, bevacizumab is effective for many eyes, and may be preferable from a cost-effectiveness standpoint. Although persistent DME is a frequent occurrence after anti-VEGF therapy, visual outcomes are generally good even in eyes with persistent edema. When considering switching therapeutic methods, it is recommended to wait at least until 24 weeks of q4 timely anti-VEGF therapy; however, in case of stability in the presence of persistent DME, stopping therapy and observing is a viable alternative. No clear role exists for steroids as primary therapy for DME or as secondary therapy in incomplete responders.

Ongoing DRCR.net studies will provide additional data on the treatment of eyes with DME. Protocol AC is exploring whether the initiating therapy with bevacizumab followed by switching to aflibercept in incomplete responders is a viable alternative to immediate aflibercept therapy for all eyes. Protocol AE is evaluating the safety and efficacy of photobiomodulation therapy for eyes with good vision despite center-involved DME. These and future studies will continue to inform the treatment of DME and other diabetic eye complications to best preserve vision in patients with diabetes.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1	Sun JK, Jampol LM. The diabetic retinopathy clinical research network (DRCR.net) and its contributions to the treatment of diabetic retinopathy. Ophthalmic Res 2019;62:225-30.
2	Diabetes Control and Complications Trial Research Group; Nathan DM, Genuth S, Lachin J, Cleary P, Crofford O, et al. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977-86.
3	Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998;352:837-53.
4	Photocoagulation for diabetic macular edema. Early Treatment Diabetic Retinopathy Study report number 1. Early Treatment Diabetic Retinopathy Study research group. Arch Ophthalmol 1985;103:1796-806.
5	Treatment techniques and clinical guidelines for photocoagulation of diabetic macular edema. Early Treatment Diabetic Retinopathy Study Report Number 2. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology 1987;94:761-74.
6	Writing Committee for the Diabetic Retinopathy Clinical Research Network; Fong DS, Strauber SF, Aiello LP, Beck RW, Callanan DG, et al. Comparison of the modified Early Treatment Diabetic Retinopathy Study and mild macular grid laser photocoagulation strategies for diabetic macular edema. Arch Ophthalmol 2007;125:469-80.
7	Diabetic Retinopathy Clinical Research Network. A randomized trial comparing intravitreal triamcinolone acetonide and focal/grid photocoagulation for diabetic macular edema. Ophthalmology 2008;115:1447-9.e1-10.
8	Diabetic Retinopathy Clinical Research Network; Elman MJ, Aiello LP, Beck RW, Bressler NM, Bressler SB, et al. Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology 2010;117:1064-77.e35.
9	Elman MJ, Bressler NM, Qin H, Beck RW, Ferris FL 3^rd, Friedman SM, et al. Expanded 2-year follow-up of ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology 2011;118:609-14.
10	Bressler SB, Glassman AR, Almukhtar T, Bressler NM, Ferris FL, Googe JM Jr., et al. Five-year outcomes of ranibizumab with prompt or deferred laser versus laser or triamcinolone plus deferred ranibizumab for diabetic macular edema. Am J Ophthalmol 2016;164:57-68.
11	Bressler NM, Beaulieu WT, Maguire MG, Glassman AR, Blinder KJ, Bressler SB, et al. Early response to anti-vascular endothelial growth factor and two-year outcomes among eyes with diabetic macular edema in protocol T. Am J Ophthalmol 2018;195:93-100.
12	Elman MJ, Ayala A, Bressler NM, Browning D, Flaxel CJ, Glassman AR, et al. Intravitreal Ranibizumab for diabetic macular edema with prompt versus deferred laser treatment: 5-year randomized trial results. Ophthalmology 2015;122:375-81.
13	Diabetic Retinopathy Clinical Research Network; Wells JA, Glassman AR, Ayala AR, Jampol LM, Aiello LP, et al. Aflibercept, bevacizumab, or ranibizumab for diabetic macular edema. N Engl J Med 2015;372:1193-203.
14	Wells JA, Glassman AR, Ayala AR, Jampol LM, Bressler NM, Bressler SB, et al. Aflibercept, Bevacizumab, or Ranibizumab for diabetic macular edema: Two-year results from a comparative effectiveness randomized clinical trial. Ophthalmology 2016;123:1351-9.
15	Jampol LM, Glassman AR, Bressler NM, Wells JA, Ayala AR; Diabetic Retinopathy Clinical Research Network. Anti-Vascular Endothelial Growth Factor Comparative Effectiveness Trial for Diabetic Macular Edema: Additional Efficacy Post Hoc Analyses of a Randomized Clinical Trial. JAMA Ophthalmol. 2016;134:10.1001/jamaophthalmol.2016.3698. doi: 10.1001/jamaophthalmol.2016.3698. PMID: 27711918; PMCID: PMC5567802.
16	Ross EL, Hutton DW, Stein JD, Bressler NM, Jampol LM, Glassman AR, et al. Cost-effectiveness of Aflibercept, Bevacizumab, and Ranibizumab for diabetic macular edema treatment: Analysis from the diabetic retinopathy clinical research network comparative effectiveness trial. JAMA Ophthalmol 2016;134:888-96.
17	Wells JA, Glassman AR, Jampol LM, Aiello LP, Antoszyk AN, Baker CW, et al. Association of baseline visual acuity and retinal thickness with 1-year efficacy of Aflibercept, Bevacizumab, and Ranibizumab for diabetic macular edema. JAMA Ophthalmol 2016;134:127-34.
18	Baker CW, Glassman AR, Beaulieu WT, Antoszyk AN, Browning DJ, Chalam KV, et al. Effect of initial management with Aflibercept vs. laser photocoagulation vs. observation on vision loss among patients with diabetic macular edema involving the center of the macula and good visual acuity: A randomized clinical trial. JAMA 2019;321:1880-94.
19	Glassman AR, Baker CW, Beaulieu WT, Bressler NM, Punjabi OS, Stockdale CR, et al. Assessment of the DRCR retina network approach to management with initial observation for eyes with center-involved diabetic macular edema and good visual acuity: A secondary analysis of a randomized clinical trial. JAMA Ophthalmol 2020;138:341-9.
20	Maturi RK, Glassman AR, Liu D, Beck RW, Bhavsar AR, Bressler NM, et al. Effect of adding dexamethasone to continued ranibizumab treatment in patients with persistent diabetic macular edema: A DRCR network phase 2 randomized clinical trial. JAMA Ophthalmol 2018;136:29-38.
21	Bressler SB, Ayala AR, Bressler NM, Melia M, Qin H, Ferris FL 3^rd, et al. Persistent macular thickening after ranibizumab treatment for diabetic macular edema with vision impairment. JAMA Ophthalmol 2016;134:278-85.
22	Bressler NM, Beaulieu WT, Glassman AR, Blinder KJ, Bressler SB, Jampol LM, et al. Persistent macular thickening following Intravitreous Aflibercept, Bevacizumab, or Ranibizumab for central-involved diabetic macular edema with vision impairment: A secondary analysis of a randomized clinical trial. JAMA Ophthalmol 2018;136:257-69.
23	Diabetic Retinopathy Clinical Research Network; Browning DJ, Glassman AR, Aiello LP, Beck RW, Brown DM, et al. Relationship between optical coherence tomography-measured central retinal thickness and visual acuity in diabetic macular edema. Ophthalmology 2007;114:525-36.
24	Bressler NM, Odia I, Maguire M, Glassman AR, Jampol LM, MacCumber MW, et al. Association between change in visual acuity and change in central subfield thickness during treatment of diabetic macular edema in participants randomized to Aflibercept, Bevacizumab, or Ranibizumab: A post hoc analysis of the protocol t randomized clinical trial. JAMA Ophthalmol 2019;137:977-85.
25	Bressler SB, Odia I, Maguire MG, Dhoot DS, Glassman AR, Jampol LM, et al. Factors associated with visual acuity and central subfield thickness changes when treating diabetic macular edema with anti-vascular endothelial growth factor therapy: An exploratory analysis of the protocol T randomized clinical trial. JAMA Ophthalmol 2019;137:382-9.
26	Bressler SB, Qin H, Beck RW, Chalam KV, Kim JE, Melia M, et al. Factors associated with changes in visual acuity and central subfield thickness at 1 year after treatment for diabetic macular edema with ranibizumab. Arch Ophthalmol 2012;130:1153-61.
27	Writing Committee for the Diabetic Retinopathy Clinical Research Network; Gross JG, Glassman AR, Jampol LM, Inusah S, Aiello LP, et al. Panretinal photocoagulation vs. intravitreous Ranibizumab for proliferative diabetic retinopathy: A randomized clinical trial. JAMA 2015;314:2137-46.
28	Gross JG, Glassman AR, Liu D, Sun JK, Antoszyk AN, Baker CW, et al. Five-year outcomes of panretinal photocoagulation vs. intravitreous Ranibizumab for proliferative diabetic retinopathy: A randomized clinical trial. JAMA Ophthalmol 2018;136:1138-48.
29	Bressler SB, Beaulieu WT, Glassman AR, Gross JG, Melia M, Chen E, et al. Panretinal photocoagulation versus Ranibizumab for proliferative diabetic retinopathy: Factors associated with vision and edema outcomes. Ophthalmology 2018;125:1776-83.
30	Hutton DW, Stein JD, Glassman AR, Bressler NM, Jampol LM, Sun JK, et al. Five-year cost-effectiveness of intravitreous Ranibizumab therapy vs. panretinal photocoagulation for treating proliferative diabetic retinopathy: A secondary analysis of a randomized clinical trial. JAMA Ophthalmol 2019;137:1424-32.