A Historical Look at Early Diabetes Research

How key studies shaped later developments and still guide research today

BY SUSAN L. WORLEY, CONTRIBUTING EDITOR

Far from being closed chapters in the history of diabetes research, the Early Diabetic Retinopathy Study (ETDRS; 1979-1990), The Diabetes Control and Complications Trial (DCCT; 1983-1993) and the subsequent Epidemiology of Diabetes Interventions and Complications follow-up study (EDIC; 1993-present) produced findings that significantly inform clinical practice today and continue to influence and inspire new clinical research. Indeed, recent editorials and peer-reviewed articles that reflect on the findings of these trials and their impact on subsequent developments in the field have shown that they continue to yield valuable lessons.

The highlights of these landmark trials lie in the thought-provoking and, in many cases, paradigm-shifting answers to the fundamental questions they posed. Most of these questions evolved gradually, in response to discoveries that emerged from years of earlier research, says Frederick Ferris, MD, director of the Division of Epidemiology and Clinical Applications at the National Eye Institute (NEI), a division of the National Institutes of Health (NIH). This was particularly true for the three most important questions addressed by the ETDRS. When Dr. Ferris first joined the NEI in 1973, for example, a trial examining the benefits of laser treatment and xenon arc photocoagulation for diabetic retinopathy was already planned — one that built on years of prior experimentation, sometimes with relatively rudimentary methods of photocoagulation.

“One early instrument was positioned on the roof of a building,” says Dr. Ferris, “where it focused the sun on the peripheral retina and made burns, much as you might use a magnifying glass to burn a leaf.”

Over time, experimentation taught researchers that lasers could be used to treat abnormal blood vessels in the peripheral retina and help prevent visual loss, but it also generated questions. When was the best time to introduce laser treatment — before a patient developed proliferative disease, or after proliferative retinopathy had already developed? And could photocoagulation be used to treat macular edema? Meanwhile, the anecdotal observation that patients with diabetes who took aspirin regularly, for conditions such as rheumatoid arthritis, seemed less likely to develop retinopathy inspired a third question: Could aspirin possibly prevent the progression of diabetic retinopathy?

FIGURE 1. Frederick Ferris, MD, director of the Division of Epidemiology and Clinical Applications at the National Eye Institute, discusses how past studies (ETDRS, DCCT and EDIC) still impact and guide studies and treatment approaches today.

Approximately 6 years after the launch of the ETDRS, a multicenter randomized clinical trial involving 3,711 patients with nonproliferative or early proliferative diabetic retinopathy and varying degrees of macular edema, the answers to all three questions seemed clear.

“The answer to the first question,” says Dr. Ferris, who served as co-chair of the ETDRS, “was that it didn’t seem to matter very much whether laser treatment was used early or was deferred. Very few eyes in either group went blind during this study — even after 5 to 7 years. But for reasons that we still don’t completely understand, there was one exception to this finding: Patients with type 2 diabetes did seem to benefit from earlier treatment.”

And argon laser treatment of proliferative diabetic retinopathy wasn’t just effective; EDTRS unequivocally confirmed that it was capable of reducing the risk of blindness by approximately 90%. Moreover, investigators discovered during the trial that patients with severe nonproliferative retinopathy — particularly when both eyes were affected — could benefit from ‘scatter’ laser treatment, or panretinal photocoagulation.

A very different type of laser treatment, employing significantly smaller burns (about 50 microns, rather than 500) that were focused on microaneurysms and other areas of leakage in the macula was used to answer the second question posed by the study. The ETDRS demonstrated that this type of laser treatment effectively reduced macular edema and helped preserve central vision. Notably, it was during this study that ‘clinically significant macular edema’ — a term still recognized as a legacy of the EDTRS — was coined for practical reasons. It evolved to describe eyes at risk for visual loss that could benefit from laser treatment.

Today, during what is now thought of as the anti-VEGF era, laser treatment is used far less frequently, and some clinicians question its value. However, many experts, including Dr. Ferris and colleague David S. Boyer, MD, co-founder of the Retina-Vitreous Associates in Los Angeles, believe there’s still a role for laser in clinical practice.

“In studies that have shown anti-VEGF therapy to be a remarkably effective treatment,” says Dr. Boyer, “the primary criteria for enrollment was center-involving macular edema. However, for the patient who presents with pericentral involvement, or with involvement in the center, with leaks away from the foveal center and the capillary-free zone, laser treatment can be beneficial. Moreover, although treatment with anti-VEGF therapy alone is sufficient in most cases, there are times when supplemental laser photocoagulation therapy can help reduce edema that may be present outside the foveal area.”

The third formal question posed by the ETDRS presented a dilemma for investigators who designed the trial.

“We felt that we couldn’t subject thousands of patients to the high doses of aspirin typically taken by patients with RA,” says Dr. Ferris, “so we arrived at a sort of halfway approach. Instead of taking the equivalent of a baby aspirin, the patients received 650 mg. The 80-mg dose would have been enough to have an effect on platelets, but 650 mg provided some anti-inflammatory effect, and at the time, it was believed that inflammation worsened diabetic retinopathy.”

Ultimately, the ETDRS demonstrated that aspirin had no beneficial effect on the progression of retinopathy. However, equally important was the finding that aspirin wasn’t harmful — a discovery that has benefitted patients for decades and likely saved many lives.

“The observation that there was no difference in the frequency of hemorrhage or related visual loss meant that there was no rationale for discontinuing aspirin taken by patients at risk for complications related to heart disease or stroke,” says Dr. Ferris.

Not long after the launch of the ETDRS, a more controversial clinical study — the DCCT — was under way. The subject of significant debate, the DCCT was designed to investigate the validity of the glucose hypothesis, by determining the potential macrovascular as well as microvascular benefits of tight control of glucose.

“This study answered a question that had been tormenting physicians since the introduction of insulin,” says Dr. Ferris. “While some believed that tight control was critically important to preventing secondary complications of diabetes, others believed that blood sugar should be lowered only to the point of eliminating symptoms, and there was disagreement about the difference in the risks associated with these approaches.”

The potentially serious consequences of excessive control of glucose, including seizure and adverse events requiring hospitalization, prevented a comparison of these approaches until the advent of newer, safer methods of monitoring glucose, such as the HbA1c assay, made a controlled study feasible. The findings from the DCCT were revolutionary and had far-reaching implications.

“The results directly translate to the recommendations that we now make to patients on a daily basis,” says Dr. Boyer. “Every ophthalmologist now has an essential obligation to explain to patients with diabetes the importance of tight control of glucose, as well as the management of hypertension and lipids.”

It was during the 4-year EDIC follow-up study that investigators first encountered the remarkable phenomenon of metabolic memory, which remains an important subject of research today. The EDIC demonstrated that for patients in the DCCT who were able keep their A1C under tight control during intensive treatment, the benefits were surprisingly durable.

FIGURE 2. Severe nonproliferative diabetic retinopathy with DME (California ultra-widefield imaging by Optos).
Image courtesy of Seenu Hariprasad, MD

“When patients who had been in the DCCT’s conventional treatment group ended up requiring more laser than those who had been under tight control, it became apparent that a kind of ‘memory’ was operating,” says Dr. Boyer, who credits the EDIC with underscoring the enormous value of tight metabolic control of diabetes, and likewise, the significant consequences of failing to achieve control.

Dr. Ferris notes that a major legacy of all of these trials, perhaps as important as the questions they answered, was a new and immensely valuable understanding of the natural history of diabetes. For the first time, clinicians and researchers had the ability to identify modifiable risk factors for microvascular complications of the disease, and a foundation for developing indispensible clinical research tools.

“After rating retinopathy in yearly photos over a period of 5 to 6 years or longer, we learned to accurately identify eyes that were progressing,” says Dr. Ferris. “Today, we regularly use the progression of retinopathy as an outcome variable when testing new drugs in clinical trials.”

Although with hindsight investigators might conduct these trials somewhat differently today, Dr. Ferris and Dr. Boyer agree that they were extraordinary for their time.

“It’s important to consider when these studies were conducted — to imagine a time when OCT was not available,” says Dr. Boyer. “Today, investigators conducting a trial such as the ETDRS might, for example, divide patients into groups with extra-foveal leaks, central-involving leaks, and diffuse leaks. But back then, researchers were working with what was available. You have to crawl before you walk, and then run, and we were crawling at that time. For the first time in history, these trials gave us a handle on the microvascular and macrovascular complications of diabetes.” ❖