Laird Harrison

August 13, 2017

BOSTON — Real-world patients with diabetic macular edema treated with anti–vascular endothelial growth factor (anti-VEGF) do not do as well as similar patients in clinical trials, according to an analysis of medical records from nearly 6000 patients seen at ophthalmology practices in the United States.

"The patients had about eight injections, and they gained six letters at 1 year, so from that perspective, it's good news," said Thomas A. Ciulla, MD, MBA, from the Indiana School of Medicine in Indianapolis, and an employee of Spark Therapeutics in Philadelphia, Pennsylvania.

"From the perspective of matching a randomized clinical trial, it's bad news," he said here at the American Society of Retina Specialists (ASRS) 2017 Annual Meeting.

From the perspective of matching a randomized clinical trial, it's bad news. Dr Thomas A. Ciulla

Three anti-VEGF treatments — ranibizumab (Lucentis, Genentech), bevacizumab (Avastin, Genentech), and aflibercept (Eylea, Regeneron Pharmaceuticals) — have achieved dramatic results in clinical trials.

For example, in the Diabetic Retinopathy Clinical Research Network (DRCR) trial, patients receiving aflibercept 2.0 mg improved a mean of 13.3 letters after 1 year of treatment.

As a result, anti-VEGF agents have become the treatment of choice for diabetic macular edema. In a 2016 ASRS survey of retina specialists, 95% said anti-VEGF was their first choice for these patients, with 60% choosing bevacizumab, 10% ranibizumab, and 25% aflibercept.

But patients and physicians face more challenges outside of the rarefied atmosphere of these trials, Dr Ciulla told Medscape Medical News.

The injections are uncomfortable, and many patients are elderly and often have other health problems that make it hard for them to get to appointments, he explained. And increasingly, physicians are being judged, and reimbursed, on the basis of the outcomes in their own patients, he pointed out.

As previously reported by Medscape Medical News, studies of these drugs in neovascular age-related macular degeneration have shown that patients in the real world have fared much worse than patients in clinical trials.

To see how the results in patients with diabetic macular edema compared with those from clinical trials, Dr Ciulla and colleagues looked at data from the Vestrum Health Retina Research Dataset of electronic medical records from 800,000 patients at hundreds of retina practices across the United States.

The researchers selected 5872 patients who received at least 3 monthly anti-VEGF injections between January 2011 and April 2016 and for whom follow-up data were available up to October 2016. The patients' mean initial age was 62 years, and their mean baseline visual acuity was 58 letters.

The researchers divided the eyes into 3 groups according to which drug was used. Then they subdivided the eyes into 3 cohorts, depending on length of follow-up (6, 12, and 24 months), with each cohort being mutually exclusive.

The patients in each group received a similar mean number of macular and pan-retinal laser treatments.

Although all the groups improved, the improvement was less than that seen in the DRCR trial.

Table 1. Change in Letter Score by Treatment Duration


Treatment 6-Month Cohort 12-Month Cohort 24-Month Cohort
Bevacizumab +6.1 +6.0 +6.5
Ranibizumab +4.9 +6.1 +6.2
Aflibercept +6.9 +7.4 NA

One reason that the patients in the real world did not get the same benefit as those in clinical trials may be that they got fewer injections, said Dr Ciulla.

In the DRCR study, patients got a median of nine or 10 injections over the course of 12 months. In this real-world study, the 12-month cohort got a mean of 8.6 injections. (The 6-month cohort received a mean of 5.2 injections, and the 24-month cohort received 14.8).

In addition, real-world patients are more likely to have other characteristics that would exclude them from clinical trials, such as proliferative diabetic retinopathy, ischemia, prior laser treatment, extremes in baseline visual acuity, high HbA1C levels, or comorbidities such as hypertension or renal insufficiency, Dr Ciulla said.

These results also differed from those in the DRCR trial, in that all three anti-VEGF drugs produced about the same amount of improvement, he reported. (There were not enough patients taking aflibercept for 24 months to analyze those data.) In the DRCR study, aflibercept was more effective than the other two drugs in patients who started with visual acuity of about 20/50 or worse.

One explanation for the difference could be that patients in the real world did not receive as intensive a treatment, said Dr Ciulla.

In keeping with other clinical trials, however, the patients who started off with the worst visual acuity improved the most.

Table 2. 12-Month Change in Letter Score by Baseline Visual Acuity


Baseline Visual Acuity Bevacizumab Ranibizumab Aflibercept
20/201 or worse +37.3 +31.3 +36.6
20/71 to 20/200 +8.7 +10.2 +11.3
20/41 to 20/70 +3.0 +4.5 +4.3
20/40 or better −2.3 −1.9 −2.0

"Patients with good vision always do less well than patients with poor vision because there's a ceiling effect," said Dr Ciulla. "Patients who have really bad vision at the start have more to gain."

But this does not mean that patients with relatively good vision will not benefit from treatment, he added. "If highly motivated patients are followed closely, and if they start taking care of themselves, they should improve."

I think it's encouraging. This shows that most patients will tend to get better. Dr Carl D. Regillo

In the discussion period after the presentation, one person wanted to know at what point clinicians should turn to other treatments such as steroids or laser to treat patients with diabetic macular edema. "In the real world, we have more arrows in our quiver," he said.

Dr Ciulla responded that clinicians should not give up on anti-VEGF treatments too early. "They actually work quite well at 12 months," he said.

Other panelists said they eventually use laser or steroids if they do not see enough improvement in their patients.

Moderator Carl D. Regillo, MD, from the Wills Eye Center in Philadelphia, Pennsylvania, saw the results as mostly good news.

"I think it's encouraging," he told Medscape Medical News. "This shows that most patients will tend to get better."

Dr Ciulla is an employee of Sparks Therapeutics. Dr Regillo has worked on clinical trials supported by Genentech and Regeneron.

American Society of Retina Specialists (ASRS) 2017 Annual Meeting. Presented August 13, 2017.

Follow Medscape Ophthalmology on Twitter @MedscapeEye and Laird Harrison @LairdH


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.