As we have no doubt seen in our clinics, many patients presenting with diabetic macular edema (DME) have good visual acuity (VA): 40% of patients presenting for the laser treatment trials in the Early Treatment Diabetic Retinopathy Study (ETDRS) had VA of 20/20 or better; 45% had visual acuities of 20/25 to 20/40.¹ Intravitreal anti-VEGF therapy, often guided by optical coherence tomography (OCT) appearance of the macula to detect center-involving diabetic macular edema (CI-DME; see Figure 1), has been shown to improve vision and allow for regression of diabetic retinopathy severity.^2-6 A question that retina specialists have had for decades is whether treatment of DME in patients with good vision is beneficial. To answer that, the DRCR Retina Network Protocol V study evaluated whether anti-VEGF injections were a necessary strategy in patients with good vision who have CI-DME.

Ninety-one clinical sites in the United States and Canada participated in this multicenter study between November 2013 and September 2016. Random assignment of 702 eyes (among 702 patients) with CI-DME and VA of 20/25 or better was arranged to initial management with prompt administration of 2.0 mg of intravitreous aflibercept (n=226), prompt laser photocoagulation (n=240) or observation (n=236).

In the aflibercept group, injections were administered up to every 4 weeks. In the laser photocoagulation group, patients were treated at baseline with retreatment considered at 13 weeks. In the observation group, no treatment was given at baseline. Patients in the laser and observation groups were monitored at weeks 8 and 16 and then at 16-week intervals. Those in the laser and observation groups were switched to aflibercept (“rescue”) if a decrease of 2 or more lines at any visit or 1 line at 2 consecutive visits was reported.

At 2 years, the rates of VA loss of 5 or more ETDRS letters (≥1 line of vision) did not differ significantly between groups (16% of the aflibercept group, 17% of the laser group, and 19% of the observation group); the average VA was 20/20, as it was at baseline, with all 3 management strategies. Three-quarters of the eyes in the laser group and two-thirds of the eyes in the observation group did not need to receive anti-VEGF injections over the 2 years of follow-up.^7-9 That deserves emphasis: two-thirds of the observation group never experienced a 5-letter or more decline in VA over the 2-year period.

Observations regarding diabetic retinopathy severity are also important, of course; severity may be influenced by local treatment as well as systemic control. Interestingly, approximately 10% of those in the observation group showed a 2-step improvement in diabetic retinopathy severity level, which was not statistically dissimilar to that seen in the other groups (14% of those in the prompt aflibercept therapy and 12% of those in the laser photocoagulation groups showed 2-step improvements during the study).

The DRCR Retina Network has delivered level 1 evidence to assist physicians in the management of the common scenario of the patient with DME and good vision with Protocol V. How should we use this information to help our patients?

We are fortunate to have commentary on the effects of the results of Protocol V on our practices by Pravin Dugel, MD, managing partner, Retinal Consultants of Arizona; Retinal Research Institute and clinical professor, Roski Eye Institute, Keck School of Medicine, University of Southern California, and David Boyer MD, Retina-Vitreous Associates Medical Group in Los Angeles, California.

Real-World Patient Compliance

David Boyer, MD
Retina-Vitreous Associates Medical Group, Los Angeles, California

Protocol V was a randomized phase 3 trial investigating 3 different treatment paradigms for the treatment of CI-DME in patients with good vision. Patients’ vision had to have 20/25 or better at 2 consecutive visits within 28 days. The patients were assigned 1:1:1 as follows:

• Prompt focal/grid photocoagulation + deferred intravitreal anti-VEGF (aflibercept)
• Observation + deferred intravitreal anti-VEGF (aflibercept)
• Prompt intravitreal anti-VEGF (aflibercept)

The primary endpoint was the percentage of eyes that lost at least 5 letters of VA at 2 years compared with baseline VA (mean of the 2 VA letter scores within 1 to 28 days required for eligibility). Anti-VEGF was needed in 34% of the observation subgroup and 25% of the prompt laser subgroup.

Although this study was well powered, with 702 eyes being studied, it is important to realize that patients in randomized studies tend to be compliant, maintaining good blood sugar control and returning for their follow-up examinations. The patients in this study were followed and returned as scheduled, but still almost 30% of patients required anti-VEGF. We don’t know what would happen if the patients did not return.

We also need to explore the treatment burden of examinations and the expense of treatments in these 3 groups. The study does give us information that we do not have to begin treatment in these patients as soon as we see them.

Extrapolating Results to the Clinic

Pravin U. Dugel, MD
Retinal Consultants of Arizona; Roski Eye Institute, Keck School of Medicine, University of Southern California

The DRCR Retina Network Protocol V is a seminal study for both its study topic and its execution. All the landmark studies thus far, whether industry or DRCR Retina Network sponsored, have enlisted patients with CI-DME and poor vision (78 ETDRS letters or Snellen equivalent of 20/32 or worse). However, the majority of patients with DME, whether center involved or not, have good vision. Protocol V certainly makes a dent in this knowledge gap. Moreover, like most other DRCR Retina Network studies, the design and analyses of Protocol V are excellent.

How should Protocol V impact my patients? The most pertinent question is, were the patients enrolled in Protocol V representative of my clinic patients? The answer in this and all clinical trials, by definition, is, “not entirely.” By necessity, there is a self-selection of patients who enroll in a clinical trial: they tend to be more motivated, healthier, and more compliant than typical clinic patients.

On average, patients had well controlled diabetes with a mean hemoglobin A_1c of 7.6, and enrolled eyes had earlier stages of diabetic retinopathy, with moderate or milder diabetic retinopathy severity in >60% of eyes, and mild DME. At baseline, mean central subfield thickness was 311 mm, markedly lower than means in the high 400s to 500s reported in most phase 3 DME trials and below the enrollment threshold for men using the Heidelberg Spectralis OCT in Protocol V itself. Therefore, and most importantly, whether the Protocol V results can be extrapolated to our typical clinic patients who have more severe disease is not known. To do so may be inappropriate at best and dangerous at worst.

Another disconnect may be the measure of visual function. Clinical trials rely on VA (typically ETDRS vision). However, rarely do we live life looking at a bright light at the end of a dark tunnel. Typical patients in our clinic often complain of visual dysfunction independent of VA, such as distortion, loss of contrast sensitivity, and blind spots, all of which can be debilitating despite excellent VA. Akin to the diabetes severity disconnect, this is also an equally important disconnect. Therefore, as before, it is simply not known whether the Protocol V data can be extrapolated.

In conclusion, history will record Protocol V as the first seminal study to target our most common DME patient population. It is an excellent study. However, we should exercise great caution in extrapolating its findings to all our clinic patients. This study is an important first piece of a large puzzle that will define how we personalize care for our patients with DME. We await the other equally important pieces to this puzzle. RP

REFERENCES

1. Early Treatment Diabetic Retinopathy Study design and baseline patient characteristics. ETDRS report number 7. Ophthalmology. 1991;98(5 Suppl):741-756.
2. Elman MJ, Aiello LP, Beck RW, et al; Diabetic Retinopathy Clinical Research Network. Randomized trial evaluating ranibizumab plus prompt or deferred laser or triamcinolone plus prompt laser for diabetic macular edema. Ophthalmology. 2010;117(6):1064-1077.
3. Michaelides M, Kaines A, Hamilton RD, et al. A prospective randomized trial of intravitreal bevacizumab or laser therapy in the management of diabetic macular edema (BOLT study) 12-month data: report 2. Ophthalmology. 2010;117(6):1078-1086.
4. 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(4):615-625.
5. 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(4):789-801.
6. Ip MS, Domalpally A, Hopkins JJ, et al. Long-term effects of ranibizumab on diabetic retinopathy severity and progression. Arch Ophthalmol. 2012;130(9):1145-1152.
7. Baker CW, Glassman AR, Beaulieu WT, 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 VA: a randomized clinical trial. JAMA. 2019;321(19):1880-1894.
8. Peto T, Chakravarthy U. New findings from diabetic retinopathy clinical research retina network protocol V confirm a role for focal laser photocoagulation or observation for eyes with center-involved diabetic macular edema and good VA: new is not always best. JAMA Ophthalmol. 2019;137(7):838-839.
9. Chew EY. Patients with good vision and diabetic macular edema involving the center of the macula: to treat or not to treat? JAMA. 2019;321(19):1873-1875.

Editor’s note: This article is part of a special edition of Retinal Physician that was supported by REGENXBIO.