The European Association for the Study of Diabetes (EASD) meeting traditionally opens with the Claude Bernard Lecture, which “recognizes contributions to the advancement of knowledge in the field of diabetes mellitus and related metabolic diseases.”
This year, the award went to Dr Jaakko Tuomilehto. Among his many accomplishments, Tuomilehto led the Finnish Diabetes Prevention Study (DPS), which demonstrated a 58% reduction in diabetes incidence with lifestyle intervention.
[1]
This finding was confirmed in other studies from around the world,
[2,3,4,5]including the US-based Diabetes Prevention Program (DPP).
[5] It was no surprise, then, that Tuomilehto’s lecture was entitled “Prevention of Type 2 Diabetes: A Dream That Came True.”
Prerequisites for Disease Prevention
Tuomilehto set the stage by identifying general prerequisites for disease prevention. One prerequisite is that there must be a window of opportunity between recognition of risk and onset of the disease.
The at-risk window of opportunity for diabetes prevention is the 10- to 12-year period of beta-cell decline.
The United Kingdom Prospective Diabetes Study estimated beta-cell function to be already reduced by approximately 50% at the time of diabetes diagnosis, with a subsequent linear decline over the ensuing 6 years.
[6]
Extrapolating backward, this means that beta-cell dysfunction begins 10-12 years before the onset of diabetes. In other words, the at-risk window of opportunity for diabetes prevention is this 10- to 12-year period of beta-cell decline.
According to Tuomilehto, another prerequisite for disease prevention is that the at-risk period be identifiable. Prediabetes is that period, but perhaps not as currently defined.
Before the use of the term “prediabetes,” diabetologists identified impaired glucose tolerance or impaired fasting glucose as hyperglycemia above normal but not sufficiently high to warrant a diagnosis of diabetes.
[7]
They recognized that these conditions do not perfectly overlap and that the risk for diabetes was highest when both were present. For that reason, the DPS and DPP required both impaired glucose tolerance and impaired fasting glucose for enrollment in the trials.
A1c Not Sufficiently Sensitive
Note that at the time of the DPS, A1c was used for diabetes management but not diagnosis, so there was no A1c-based definition for elevated but subdiagnostic hyperglycemia.
Since 2010, we have had such a definition, but Tuomilehto argued that A1c is not sufficiently sensitive to use for diagnosing diabetes and, by extension, defining the at-risk condition.
He pointed out that in post hoc analyses, the DPS concluded that, had an HbA1c-based definition been used for determining diabetes incidence in the DPS, the findings would not have been statistically significant.
[8]
In short, he argued that prediabetes should be defined by the presence of both impaired glucose tolerance and impaired fasting glucose, but not either one alone, and certainly not by A1c.
A Proven Intervention
Perhaps the most important prerequisite for disease prevention is that there must be a proven intervention that reduces or eliminates risk. Thanks to the DPS, the DPP, and other efforts, we know that weight loss and a healthy diet can substantially reduce the incidence of diabetes and that the reduction extends many years beyond the end of the active intervention.
[9,10,11]
One of the more interesting findings in the DPS was the relationship between diabetes incidence and attainment of intermediate measures. Five goals were set for the intervention group in the DPS: weight loss of 5% or more, fat intake less than 30% of energy consumed, saturated fat less than 10% of energy consumed, fiber intake of at least 15 g per 1000 kcal, and at least 30 minutes of exercise per day. No participants in the intervention or control groups who achieved four or all five goals developed diabetes.
[12]
Tuomilehto concluded that his dream of preventing diabetes has, to some extent, come true. He and others have shown that lifestyle changes can reduce the risk for diabetes, and the word is getting out. Although the reasons are unclear, there is some evidence that the incidence of diabetes is declining.
[13]
The Nightmare That Is Obesity
Nevertheless, obesity that leads to diabetes remains a massive problem worldwide, and unfortunately, many people simply cannot or will not engage in lifestyle changes, such as those in the DPS or DPP. Thus, there is a role for adjunct or alternative strategies, such as pharmacologic agents.
Lorcaserin is a weight loss drug with proven efficacy in clinical trials. It was conditionally approved by the US Food and Drug Administration, contingent on demonstrated cardiovascular safety in a postmarketing study.
[14]
The results of that study, the Cardiovascular and Metabolic Effects of Lorcaserin in Overweight and Obese Patients-Thrombolysis in Myocardial Infarction 61 (CAMELLIA-TIMI 61) trial,
[15] were reported at EASD.
The CAMELLIA-TIMI 61 Trial
CAMELLIA-TIMI 61 was a randomized, double-blind, placebo-controlled, multinational trial in which 12,000 patients received lorcaserin 10 mg twice daily or placebo (6000 participants in each group) on top of a standardized weight management program that included dietary and exercise information. Median follow-up was 3.3 years.
The primary safety outcome was a composite of cardiovascular death, myocardial infarction, or stroke, and the primary efficacy outcome was the safety composite plus hospitalization for unstable angina, heart failure, or coronary revascularization.
The primary metabolic efficacy outcome was time to incident diabetes among those with prediabetes at baseline. Secondary metabolic endpoints were incident diabetes in all those without diabetes, achievement of normoglycemia in those with prediabetes, and change in A1c in those with diabetes.
Greater Weight Loss and Risk Reduction
The weight loss and cardiovascular results, published simultaneously,
[16] were presented first. As expected, weight loss was greater among those in the treatment group, and a significantly greater number of participants in the treatment group experienced weight loss of 5% or more (39% vs 17%;
P < .001) or 10% or more (15% vs 5%;
P < .001).
No differences in cardiovascular safety or efficacy composite outcomes were found, nor were any components of the outcomes significantly different. Overall, the proportions reporting serious adverse events were equal, but headache and nausea were reported significantly more frequently in the treatment group, which drove a significantly higher total rate of adverse events possibly caused by the trial agent (7.22% vs 3.67%).
The metabolic results, also published simultaneously,
[17] were more interesting. The weight loss effects of lorcaserin were similar among patients with diabetes, prediabetes, and normoglycemia, but only a modest reduction in A1c was observed in those with diabetes and negligible changes in those with prediabetes or normoglycemia. However, patients with prediabetes in the treatment group had a 19% lower risk for incident diabetes.
When all participants without diabetes (ie, prediabetes and normoglycemia combined) were considered, the risk reduction was 23%. Nearly 60% of patients with prediabetes in the lorcaserin group achieved normoglycemia, a rate that was 26% greater than with placebo.
As an exploratory outcome, risk for persistent microalbuminuria was reduced in the treatment group, although the absolute rates themselves were fairly low (7.8% vs 10.0%). All these results were achieved with a slight but rare increase in hypoglycemia requiring hospitalization (0.4% vs 0.1%; P = .054).
A ‘Down the Line’ Option
Dr Naveed Sattar provided independent commentary. He complimented the study team on a well-conducted trial and was pleased that lorcaserin did not show any increased risk for cardiovascular events. The metabolic results, however, were viewed less favorably.
Despite the weight loss results, if the goal is to prevent incident diabetes, there are better options. Specifically, metformin reduces diabetes risk by 31%,
[5]orlistat by 37%,
[18] and acarbose by 25%
[19] (vs 19% with lorcaserin). All of these are available in generic formulations and thus have considerably lower costs.
Liraglutide offers a larger weight loss effect and better metabolic effects than lorcaserin,
[20] and sodium-glucose cotransporter-2 inhibitors provide nearly equivalent weight loss; superior metabolic effects; and of course, cardiovascular benefit.
[21] In light of existing agents, therefore, Sattar suggested that lorcaserin should be considered a “down the line” option.
In summary, prevention of diabetes remains a hot topic of research and will continue as such for as long as the worldwide epidemic continues. We can look forward to more excellent presentations at next year’s EASD meeting in Barcelona, Spain.
