Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Study: Continuing to Build on 40 Years of Diabetes Research

Over a hundred years ago, the discovery of insulin—one of the most notable scientific advances of the 20th century—ushered in a new era in chronic disease management. Specifically, with the clinical availability of insulin, an incredible paradigm shift emerged that changed type 1 diabetes from a rapidly progressive disease associated with significant morbidity and mortality to a more manageable chronic condition. With the shift to a chronic condition, questions then emerged as to how to best manage the disease and prevent the development of devastating complications. It was critical to test the hypothesis that managing the hyperglycemia associated with diabetes would delay the development and progression of the complications and improve the health outcomes of people with type 1 diabetes. Fortunately, definitive clinical trial evidence proving the “glucose hypothesis” of diabetes complications was indeed obtained and the data presented for the first time in June 1993 at the 53rd American Diabetes Association Annual Meeting and Scientific Sessions in Las Vegas, NV (1). At this meeting, many of us were in awe as we heard the presentation of the initial results of the Diabetes Control and Complications Trial (DCCT) (supported by the National Institute of Diabetes and Digestive and Kidney Diseases [NIDDK] of the National Institutes of Health), demonstrating that the risk of microvascular complications over an average of 6.5 years of follow-up was reduced by 26–63% with intensive versus conventional therapy.

After the initial results of the DCCT were presented, the first principal results publication appeared in the New England Journal of Medicine in September 1993 and remains one of the most cited publications in the diabetes literature (1,2). After the close of the DCCT, the NIDDK launched the Epidemiology of Diabetes Interventions and Complications (EDIC) study, for which the original cohort has been longitudinally assessed for the development of significant microvascular disease and the development of cardiovascular and other macrovascular diseases. It is truly incredible how this landmark study has continued to present novel results 40 years since the feasibility study was first launched and 30 years after the initial scientific findings. Thus, in this issue of Diabetes Care are five new reports from DCCT/EDIC that continue to show how stable long-term investment in research provides significant public health benefits. These reports are the latest in DCCT/EDIC’s long history of providing new knowledge on diabetes complications and the effects of intensive glycemic management. Here, DCCT/EDIC researchers specifically provide more in-depth understanding of diabetes-associated retinal structure abnormalities, bone microarchitecture, neuropathic pain, and plasma biomarkers of brain injury and kidney tubular function (3–7).

From data obtained during EDIC, it is well known that intensive as opposed to conventional therapy during the DCCT resulted in a significant decrease in the cumulative incidence of individuals showing retinopathy progression and progression to more severe levels of retinopathy that require intervention (8). In the study from Blodi et al. (3) in this issue of Diabetes Care, the authors provided a further evaluation of the effects of glycemic control on retinopathic changes in the DCCT/EDIC cohort. Specifically, they investigated changes in retinal structure (both quantitative and qualitative) using optical coherence tomography (OCT), which visualizes intraretinal structures at micron resolution, and evaluated associations with systemic and other risk factors in participants of the EDIC study cohort. Interestingly, they reported that participants originally in the conventional treatment group were more likely to have structural macular abnormalities, as reflected by more disorganization of retinal inner layers (DRIL) and more intraretinal fluid or intraretinal cysts. Of great clinical significance, the authors noted that DRIL, as a morphological abnormality, was associated with a history of macular edema, a history of ocular surgery, and worse visual acuity. The authors reported that early intensive glycemic management decreased risk of DRIL in DCCT/EDIC participants. The significance of this study is that it continues to emphasize the importance of metabolic memory by demonstrating that intensive glycemic management in DCCT/EDIC participants had benefits on the retina beyond features detected by fundus photographs and ophthalmoscopy routinely obtained in clinical practice. This important observation contributes to our understanding of retinopathy development in type 1 diabetes.

Previous reports on DCCT/EDIC have demonstrated significant declines in memory and psychomotor and mental efficiency and concluded that cognitive function declines with aging in type 1 diabetes (9). In this issue, Karger et al. (4) took the next logical step to understand mechanisms of brain injury by evaluating the associations between plasma biomarkers of brain injury and brain MRI and cognitive measures. The authors stated that in previously reported smaller studies, plasma biomarkers of brain injury revealed potential associations between the biomarker and brain structural abnormalities and/or cognitive function. Thus, in 373 participants in the EDIC study, the authors assessed plasma biomarkers (i.e., amyloid-β-40, amyloid-β-42, neurofilament light chain [NfL], phosphorylated Tau-181 [pTau-181], and glial fibrillary acidic protein [GFAP]) along with cognitive measures and MRI assessments. The authors reported the most notable biomarker results from their study were with NfL, a cytoskeletal filament protein in neurons released secondary to neuropathological processes (4). They found that NfL was the only biomarker significantly associated with accelerated brain aging in their models and that it significantly declined in individuals with type 1 diabetes with cognitive impairment. In addition, amyloid-β-42 measures were not associated with cognitive symptoms or MRI measures, and elevated NfL and pTau-181, but not amyloid-β-42, may reflect accelerated neurodegeneration without a prototypical Alzheimer disease pattern. Thus, this study provides important insights into the pathophysiology of cognitive dysfunction in older individuals with type 1 diabetes. Such novel observations clearly would not have been possible without DCCT/EDIC’s focus on closely following the DCCT/EDIC cohort over decades.

The long-term benefits of intensive glycemic management on nephroprotection in diabetes has also been clearly demonstrated in DCCT/EDIC (10). The promise and potential for using biomarkers to identify underlying renal dysfunction and clinical disease is that they may allow for earlier implementation of effective protective strategies. Therefore, Limonte et al. (5) sought to advance our understanding of renal dysfunction in long-term type 1 diabetes by examining tubular biomarkers reflecting kidney tubular injury and dysfunction in EDIC. Again, the significance of this report lies in the ability to leverage EDIC’s longevity and assess biomarkers over an extended period of time, allowing for an extensive assessment of temporal relationships with meaningful clinical renal outcomes. As stated by the authors, most studies to date that have assessed tubular biomarkers with kidney outcomes were measured at a single point in time. In a case-cohort study, the authors measured biomarkers across seven time points spanning ∼30 years and reported associations with diabetic kidney disease development. They observed that levels of two serum /plasma biomarkers, KIM-1 and sTNFR1, rising together over time better predicted the development of macroalbuminuria and poor kidney function than did measuring individual biomarkers alone. The novel information from serial measures from long-term follow-up of EDIC participants suggests that tubular biomarkers were more strongly associated with these clinically relevant outcomes over medium-term follow-up. The authors concluded that serial monitoring of biomarkers of kidney tubular injury and dysfunction may be useful for detecting early indicators of ongoing, progressive kidney damage in diabetic kidney disease.

An understudied complication related to type 1 diabetes is its effect on skeletal health, and it has been documented that type 1 diabetes is associated with increased risk of fractures (11). In this regard, EDIC has also furthered our understanding of bone health, demonstrating that type 1 diabetes is associated with lower bone mineral density (BMD) and that diabetes-related factors such as poorer glycemic control, advanced glycated end product (AGE) accumulation, and kidney disease are independent risk factors for lower hip BMD (11). A new report from EDIC researchers on this issue provides more key data on the adverse effect of type 1 diabetes on skeletal health. Specifically, Sinha Gregory et al. (6) describe a cross-sectional study embedded in EDIC where they sought to determine whether type 1 diabetes and its complications are associated with specific bone geometry or deficits in bone microarchitecture. They evaluated 183 individuals with type 1 diabetes from the EDIC cohort and 94 individuals without diabetes, obtaining high-resolution peripheral quantitative computed tomography (HR-pQCT) scans of the distal radius and distal and diaphyseal tibia. The authors report that, compared with control participants, EDIC participants with type 1 diabetes had lower total BMD at the distal radius and distal tibia and larger total area at distal radius, distal tibia, and diaphyseal tibia. Individuals with type 1 diabetes in EDIC had poorer radius trabecular and cortical microarchitecture, but estimated failure load was similar. They concluded that suboptimal glycemic control, AGE accumulation, and microvascular complications, particularly reduced renal function, were associated with deficits in bone microarchitecture as well as lower BMD in the EDIC participants. The clinical significance of this report suggests that maintenance of adequate glycemic control resulting in reduced AGE accumulation and prevention of microvascular complications such as renal disease may reduce fracture risk by preventing loss of bone density and preserving healthy bone microarchitecture. This study continues to confirm the benefits of early glycemic control and importantly expands our understanding of the less-studied interactions between type 1 diabetes and bone health.

Diabetic peripheral neuropathy (DPN) remains a challenging complication to manage clinically given the risk for severe pain, lower limb amputations, and impact on daily function and quality of life. In addition, data obtained in DCCT/EDIC demonstrated that DPN is significantly related to glycemic control (12). Specifically, prior research from EDIC participants noted that a higher mean HbA_1c was the most significant risk factor for DPN, followed by older age, longer duration of type 1 diabetes, greater height, macroalbuminuria, higher mean pulse rate, β-blocker use, and sustained albuminuria (12). In the final article in this series in Diabetes Care, Braffett et al. (7) sought to characterize the features and burden of neuropathic pain in type 1 diabetes. For this report, the authors evaluated the incidence of first occurrence, annual prevalence, remission, and risk factors for neuropathic pain during EDIC. Specifically, they administered the Michigan Neuropathy Screening Instrument annually between 1994 and 2020 and defined neuropathic pain with clinical signs of diabetic peripheral neuropathy by self-reported neuropathic pain plus an examination score >2, while neuropathic pain without clinical signs of DPN was defined by self-reported neuropathic pain and an examination score ≤2. The authors report that by year 26, the cumulative incidence of neuropathic pain was 57% regardless of clinical presence of diabetic peripheral neuropathy. Interestingly, they also noted there may be frequent remission, as the prevalence of neuropathic pain at 26 years was 20%. Remission was associated with several modifiable and nonmodifiable risk factors, but interestingly use of pain medication did not affect annualized remission rates. These findings highlight the high incidence of neuropathic pain in individuals with type 1 diabetes, even in the absence of accompanying clinical signs of DPN. This report further demonstrated that intensive glucose control in those with type 1 diabetes, applied early as done in the DCCT, lowers the risk of later developing neuropathic pain—findings that have not been available from any other cohort. These novel results may be key to improving clinical care of people with type 1 diabetes and to development of new management strategies for this common and burdensome disease complication.

A major strength of all five of the studies presented in this issue is the leveraging of the extremely well-characterized cohort of participants with type 1 diabetes in EDIC. All of these studies provided a better understanding of the already well-known but sometimes poorly understood clinical complications associated with diabetes. Some of the studies sought to further evaluate the effect of metabolic memory on development of their specific complication of interest, whereas the others leveraged the cohort to further characterize complications in the population with long-term type 1 diabetes. EDIC continues to provide unique opportunities to improve the lives of those with type 1 diabetes. The long-standing EDIC cohort allows investigators to propose novel hypotheses and answer meaningful research and clinical questions that cannot be addressed by shorter studies.

Collectively, these five reports add to the enduring legacy of the DCCT/EDIC studies, which continue to inform the diabetes community and provide new findings on the benefits of early glycemic control. These reports highlight how EDIC’s unique and extremely well-characterized cohort of participants with type 1 diabetes continues to contribute new knowledge on the natural history of this disease and its complications. For over four decades, the EDIC study has provided an extraordinarily detailed assessment of microvascular and macrovascular complications as well as a comprehensive assessment of long-term diabetes-related risk factors, including HbA_1c and other markers of glycemic control.

After 40 years, DCCT/EDIC remains an incredibly unique resource that still inspires and informs the current direction of diabetes research. The hundreds of published reports from DCCT/EDIC—and the resulting findings that have become accepted clinical standards of care—have been incredibly impressive contributions to public health. They also have stood as testaments to the benefits of sustained, long-term funding for medical research, demonstrating how investments in research today can yield dividends for years to come. These landmark findings, though, would not have been possible without the DCCT/EDIC investigators, study staff, and volunteer research participants. NIDDK applauds their dedication in contributing to this unique and uniquely successful research effort. Because of their vision, DCCT/EDIC’s legacy continues to grow, and we continue to celebrate the release of new findings that aim to improve the lives of those with type 1 diabetes.