Diabetes trials: is an ounce of prevention enough?

A review of recently registered interventional trials in diabetes has found that only 10% of such trials have a preventive, rather than therapeutic, purpose [1]. Will this adequately address our need for effective and practical strategies to prevent diabetes? The number of individuals estimated to have diabetes in 2012 exceeded 371 million, and at least US$465 billion was spent on diabetes-related healthcare during the prior year [101]. The number of individuals with diabetes is increasing in all countries and is projected to reach 552 million in the year 2030 [101], thus a critical need for preventive interventions exists.

Type 2 diabetes comprises the majority of diabetes cases, and it is clear from previous trials that a variety of preventive strategies can reduce the incidence of type 2 diabetes in at-risk individuals.Lifestyle modification including reduction in caloric intake and increased physical activity is unquestionably of benefit: the lifestyle intervention studied in the US Diabetes Prevention Program (DPP) reduced the risk of diabetes by an impressive 58% at 3 years and 34% at 10 years [2]. Acarbose appears to reduce the risk of diabetes by ~25%, and the DPP also showed that metformin reduced the risk of diabetes by 31% at 3 years and 18% at 10 years [2,3]. The thiazolidinedione (TZD) class of drugs has arguably proven to be the most effective in minimizing the risk of progression to type 2 diabetes, with trials of troglitazone, rosiglitazone or pioglitazone therapy demonstrating reductions in risk ranging from 50 to 75% [4–7]. Furthermore, weight-loss interventions such asorlistat given to obese subjects with impaired glucose tolerance reduced diabetes risk by 40–52%, and bariatric surgery has been shown to significantly reduce the long-term incidence of diabetes in obese subjects [8,9].

Given these effective options for intervention, why the continued onward march of the global population toward hyperglycemia? This fund of knowledge does not appear to have substantially reduced the number of individuals projected to develop diabetes in coming years. Interventions in youth may be most likely to meaningfully reduce future rates of diabetes; however, the predictors of future diabetes risk and responses to various interventions likely differ between adolescents and adults. With fewer than 4% of recently registered interventional diabetes trials intended for subjects aged <18 years [1], our understanding of preventive efforts in the young warrants further study.

Also problematic is the apparent disconnect between the results seen in preventive clinical trials and what can be readily accomplished in clinical care. Although lifestyle modification resulting in weight loss is undoubtedly an effective means of diabetes prevention, further study is needed to understand how long-term maintenance of lifestyle change can be achieved in clinical practice. Programs that effectively implement DPP-type lifestyle modification in the community have been described [8,10,11]; however, many of these programs were located in the USA and some have incorporated e-mail communication or educational websites to distribute and reinforce information. These types of interventions are likely not applicable to the majority of at-risk individuals worldwide, who currently live in low- or middle-income countries [101]. Thus, the study of preventive strategies suited to the needs of diverse populations and clinical settings is needed, particularly in low-income areas where the use of expensive medications or bariatric surgery to prevent diabetes is not practical.

The optimal role of medications in the prevention of diabetes also requires further study. For example, it remains unclear if antihyperglycemic medications have an additive preventive effect when combined with lifestyle modification. In the Indian DPP, the addition of metformin to intensive lifestyle intervention provided no preventive benefit beyond lifestyle change alone, and this combination was not studied at all in the US-based DPP [2,12]. In addition, newer classes of antihyperglycemic medications deserve study as potential preventive options – this is particularly important given the ongoing, as yet unsubstantiated use of incretin-based therapies for weight loss purposes. Certainly, the high rates of dropouts due to adverse side effects in trials of TZDs, acarbose, or orlistat support exploration of other therapeutic options [3–8]. Furthermore, the TZDs appear nearly unusable for preventive purposes given safety concerns related to the drug class. It is possible that more recently developed drugs may prove better tolerated and easier to utilize, have different associated costs or simply be more effective than older drugs, but this cannot be assumed without evaluation in appropriately designed clinical trials.

As previously noted, bariatric surgery offers a realistic option for diabetes prevention. Surgical outcomes trials have demonstrated diabetes risk reductions of 78% in obese individuals and an even greater 87% risk reduction in obese patients with impaired fasting glucose [9]. This preventive benefit has recently been found to extend to patients who did not even meet BMI criteria for bariatric surgery [13]. This finding requires additional study and if confirmed might significantly expand surgical eligibility criteria and alter guidelines for preventive care.

Trials, to date, have left many important questions unanswered. What do medications, when used successfully for diabetes prevention, actually accomplish in a physiologic sense? If a diagnosis of diabetes is delayed via use of a medication, is this simply a masking or delaying of diabetes diagnosis or has the intervention meaningfully affected the underlying disease process? The impact of regression to normoglycemia or a delayed disease diagnosis in reducing the risk of future diabetes-related complications remains poorly understood and requires further study [14]. Trials which enroll diverse populations and incorporate phenotype and/or genotype analyses may also facilitate more accurate risk stratification and prediction of responsiveness to interventions. A more robust knowledge base will permit a more refined assessment of the risks, benefits and costs associated with preventive interventions.

Although the study of treatments for established diabetes is essential, we must assess whether the percentage of study devoted to preventive efforts is adequate. It can be difficult, time-consuming and costly to conduct preventive trials; however, this may prove a wise investment given the diabetes care-related expenses projected to lie ahead. Analysis of information captured in electronic health records or disease registries may streamline these efforts in some regions, but strong support from scientific societies, government agencies and other interested organizations is needed.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.