Understanding the Phenotypes of Prediabetes to Prevent Type 2 Diabetes

Hans-Ulrich Häring, MD, PhD


October 01, 2015

This feature requires the newest version of Flash. You can download it here.

Good afternoon. My name is Dr Hans Häring, and I am a professor of medicine at the University of Tübingen, Germany. I have been working with my team for the past 20 years on the prevention of type 2 diabetes. [In the TÜbingen Family study], we have collected a group of 3000 individuals who are at high risk of developing T2D in the future, and we have phenotyped them.

When someone is in a prediabetic state, there is a chance to do prevention by lifestyle intervention. However, it has become clear over the years that there is a substantial group of people who do not respond to interventions; they are known as "lifestyle intervention nonresponders." This is an important problem, because obviously if you introduce lifestyle intervention programs and you are not successful, it is frustrating to both the patient and the doctor.

In a lecture [at the European Association for the Study of Diabetes (EASD) 2015 Annual Meeting], I summarized the data that we have collected over the years.[1] The key message is that there are many different subphenotypes of prediabetes. In the prediabetic state, people go from having normal glucose tolerance to impaired glucose tolerance, to finally ending up with type 2 diabetes.

However, there are two phenomena underlying this progression:

  1. Insulin resistance increases over the years, and we are interested in the mechanisms underlying this elevation.

  2. When insulin resistance increases, only a certain percentage of people can compensate for this increase by pancreatic hypersecretion (beta cells). The other individuals cannot compensate for insulin resistance, and they are the ones who go on to develop type 2 diabetes.

TCF7L2 Genetic Variant and Inability for Compensatory Insulin Hypersecretion

Now, what determines the ability to compensate with increasing insulin secretion? There were many studies done in the past 10 years to understand the genetic contribution to this phenomenon, and we have tested the genetic variations that were described by others. What we found in our population is that most of these genetic variations indeed induce an abnormality of insulin secretion. However, overall, the quantitative effect of all the known diabetes genes today—even if you look at the additive effect—remains quite small. I am talking about a 10%-15% reduction of insulin secretion.

So there is a lot of room to explain the inability to upregulate insulin secretion in response to insulin resistance by other mechanisms, either genetic—which are not yet identified, and there is a lot of research going on in this area—or by other mechanisms that are not primarily determined by genetics.

Among the genetic variations, there is one exception, which is probably practically very important; that is the so-called TCF7L2 (transcription factor 7-like 2), the strongest type 2 diabetes gene. We have seen that people with this gene cannot respond with upregulation of insulin. So this gene variant probably causes an inability to induce compensatory insulin secretion. In about 10% of the prediabetic population, this gene is prevalent in homozygous form, and we think that these 10% are a specific subgroup that can be treated by both early glucose-lowering agents and lifestyle interventions.[2] The German Diabetes Research Center is testing this hypothesis in a large, multicenter study, and we hope that by the end of next year, we will know whether this is really a genetic variation that leads to practical consequences in prevention.

25% of MUHO Individuals With Fatty Liver Do Not Respond to Exercise

On the other side on the question of what pushes people towards insulin resistance over the years, the obesity subphenotypes clearly play major roles. We were able to show that about 25% of the obese people in our population are in fact very insulin-sensitive. They are not at risk of developing diabetes, so we call it "metabolically healthy obesity" (MHO); in contrast, three quarters of the population have another type of obesity, the kind that is metabolically unhealthy (MUHO). This is a different fat distribution pattern, but in particular, it's the presence of a fatty liver. Only metabolically unhealthy obese people develop a fatty liver, which is a major driver of insulin resistance and is therefore a key target in the prevention of type 2 diabetes.

However, there are subforms of fatty liver. About one half of people with fatty livers have a form of fatty liver disease, which pushes insulin resistance. It is extremely important to treat these people, because fatty liver is an obstacle in successfully performing lifestyle intervention programs.[3]

It is difficult to decrease the liver fat, particularly in people who do not respond to exercise. Our study shows that 25% of people don't benefit from exercise. They do not increase the fitness of their skeletal muscles, they do not improve their insulin sensitivity, and they cannot reduce liver fat efficiently. So exercise nonresponders are a major problem in the prediabetic situation, because it is difficult for them to target the key problem—the fatty liver. Most likely, these people require specific pharmacologic treatment to allow them to succeed in lifestyle intervention programs.

The Key Role of Brain Insulin Resistance

Ten years ago, the brain appeared in the picture of the causes of prediabetes and diabetes. Insulin action in the human brain seems to be much more important than what we have thought for many years. Now it is clear that the insulin increase after a meal gives a signal to the brain, which leads to altered behavior. For instance, if you see something that you want to eat before you have the postprandial insulin increase and the signal to the brain, you are ready to want to eat this food. If the insulin signal appeared in the brain after the meal, you should not want to eat the same thing again. However, there are people who are insulin-resistant in the brain, and for them, these feedback loops that control behavior are obviously not fully functioning.

Therefore, brain insulin resistance is probably either a contributor to obesity or a consequence of obesity. We don't know that yet. Which came first, the chicken or the egg? Brain insulin resistance is also a major problem in lifestyle intervention, because people with brain insulin resistance cannot effectively reduce their visceral fat through a lifestyle intervention.

Recent data, which we have provided, show that brain insulin resistance might be an event that occurs very early in life. It might even start in utero, in pregnancies that have high glucose and insulin levels, such as gestational diabetes.[4] If this is true, then efficient diabetes prevention has to start probably during pregnancy.


The take-home messages are:

  1. We have to do lifestyle intervention according to the phenotype and the subphenotype.

  2. The major problem is in people with a fatty liver. For many of them, the only solution for efficient prevention of type 2 diabetes will be to use drugs to treat the fatty liver.

  3. Because brain insulin resistance starts so early in life, we have to focus very much on gestation and on very early prevention.


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.