Central effects of endogenous GLP-1 and the GLP-1 analog liraglutide on brain satiety and reward circuits and feeding behavior in diabetes.

Comparable to abnormal central nervous system (CNS) reward and satiety
responses contributing to drug addiction, it has been hypothesized that
alimentary obesity, the key factor in the global type 2 diabetes (T2DM)
pandemic, may result from changes in CNS reward and satiety responses to the
consummation of food in susceptible populations. However, the mechanisms
underlying these altered CNS responses to food stimuli are not clear. Treatment
with a GLP-1 receptor agonist (GLP-1RA), such as the GLP-1 analog liraglutide,
enhances nutrient-induced insulin secretion but also reduces food intake and
body weight in individuals with both obesity and T2DM. The latter effect has
been attributed to the fact that GLP-1 delays gastric emptying, but it has also
been suggested that GLP-1 plays a physiological role in the CNS regulation of
satiety and reward. We hypothesize that GLP-1 receptor activation affects
central reward and satiety circuits in the context of food stimuli, by which it
regulates appetite control, and that these processes may be altered in obesity
and T2DM. Accordingly, treatment with the GLP-1 analog liraglutide may restore
these signals resulting in weight loss.
Furthermore it is known that gastric bypass surgery in obese individuals, the
meal-related endogenous GLP-1 secretion is increased. Also altered CNS
responses have been observed after gastric bypass. We hypothesize that these
increased postprandial endogenous GLP-1 levels also affect activation in the
central reward and satiety circuits on food-stimuli, thereby regulating
appetite control

Hypotheses
1) Food-related endogenous GLP-1 is involved in the activation of CNS circuits
regulating satiety and reward, and this process is altered in individuals with
obesity and T2DM, relative to lean controls
2) The increased endogenous GLP-1 secretion after gastric bypass surgery leads
to alterations in activation of CNS circuits regulating satiety and reward, and
this contributes to the weight loss seen after gastric bypass surgery.
3) Treatment with the GLP-1 analog liraglutide may restore these signals in
obese patients with T2DM and these actions are instrumental in the observed
GLP-1(RA)-induced weight loss.

The objective of the study is to gain insight in the role of endogenous GLP-1
on food-stimuli related CNS satiety and reward responses and the alterations
between lean and obese-T2DM individuals. Also, to gain information about
treatment with a GLP-1 analog, liraglutide, may restore these signals in the
CNS in the context of food stimuli, resulting in weighloss. Furthermore we want
to investigate the involvement of the increased meal-related endogenous GLP-1
levels after gastric bypass surgery in these food-stimuli related CNS satiety
and reward responses

Study 1A:
A randomized, cross-sectional, single blind study. We will measure food stimuli
(i.e. visual food cues and liquid sips) related activity of CNS reward and
satiety circuits, using blood-oxygen level dependent (BOLD) functional Magnetic
Resonance Imaging (fMRI), and the role of endogenous (meal-stimulated) GLP-1 in
the CNS effects, using a GLP-1 antagonist infusion and saline infusion as
placebo.

Study 1B:
In this study, 10 obese, normo-glycaemic individuals, who are scheduled for
gastric bypass surgery, will be exposed to the same protocol as in study 1A.
This protocol will now be performed twice, namely 4 weeks before and 4 weeks
after the surgical procedure. The dosage of the GLP-1 antagonist (exendin 9-39)
as well as the MRI protocol will be identical to protocol 1A.
Participants will be asked to repeat the same measurements approximately 4
months after the surgery (with again the same protocol as in study 1A), to
determine the long-term effects of gastric bypass surgery.

Study 2
In a randomized, controlled, cross-over study, 20 T2DM patients will be treated
with liraglutide (at a titrated target dose of 1.8 mg daily) and insulin
glargine ( in a titrated dosage, based on self monitored fasting blood glucose
concentration) in a random order, for a period of 12 weeks each. A 12-week
washout period between the treatment episodes will be observed to minimize
carry-over effects. We intend to measure the changes in food-stimuli (i.e.
visual food cues, palatable food cues and a liquid meal) related activity in
CNS reward and satiety circuits in response to treatment with liraglutide
versus insulin in this cross over study. We will compare the effects of
liraglutide with insulin. fMRI will be performed at the begin and after 12
weeks of treatment and after 10 days of treatment, before weight changes have
yet occured.

We will investigate and compare all participants with respect to food(-related)
neuronal activity in central reward and satiety circuits by blood oxygen
level-dependent (BOLD) fMRI. Neuronal activity will be expressed in signal
change from baseline (%).

In study 1A we will perform these measurements at 2 different occations, with
a) during a saline infusion and b) during GLP-1 antagonist infusion (exendin
9-39). The order of the infusion will be randomized. Repeated blood sampling
will be performed before, during and after meal-consumption, for determination
of relevant biomarkers, metabolites and (incretin)hormones.

In studie 1B we will perform the exact same study protocol as in study 1A, but
twice (once before and once after gastric bypass surgery).
The dosage and route of administration of the GLP-1 antagonist, as well as the
MRI protocol will be identical to study 1A.
When participants also participate to the optional part of this study (aiming
to investigate the long term effects), the same protocol as study 1A will be
performed again 4 months after surgery.

In study 2 we will perform the same fMRI protocol, but without the infusion of
the GLP-1 antagonist. The T2DM participants will be treated with subcutaneous
injection of liraglutide for 12 weeks and 12 weeks insulin glargine.

The rate of gastric emptying will be estimated indirectly from the kinetics of
absorption of orally administred paracetamol.
After the fMRI sessions in study 2, participants will be presented with a
varied choice buffet, after
approximately 3* hours of ingestion of the liquid test meal (i.e. to mimic a
lunch, as compared to the timing of the liquid test meal, which represents
breakfast). Participants will not be aware that their choices and food intake
are being monitored.

Metabolic rate will be measured in fasting status and postprandial, using
indirect calorimetrie. This is a non-invasive method.

Microfasculare function will be assessed by capillary video microscopy.
Cardiovascular autonomic function will be measured with a non-invasive Nexfin
monitor, using standardized test, including Valsalva manoeuvre, deep and quiet
breathing tests and active orthostatic test, in order to study the hart-rate
response, heart-rate variation en blood pressure response and hereby the
cardiovascular autonomic function.

We are well aware of the possible demand that may be imposed on the
participants in this study. After the screening visit, the healthy participants
will travel 2 times, and the T2DM participants 7 times. For the T2DM
participants, this will be spread in a timespan of 10 months.
The duration of the visits are approximately 5-6h. A total amount of blood will
be withdrawn in study 1A of 159mL (in 1-2 months), study 1B of 301mL (in 3
months) (or 443mL in 5 months, if participating to the optional part of the
study) and in study 1+2 (only T2DM) of 528 mL (in 9-10 months). All possible
measures will be taken to minimize the discomfort for the participants.
Liraglutide is registrated already being used for the treatment of T2DM. The
side effects described are nausea and vomiting. To minimize the side effects,
we included a titration fase, starting with 0.6mg/day, increasing to 1.2mg
after 1 week, and 1.8mg after to weeks. After 10 days side effects will be
monitored by the investigator. There will also be phone contacts to inform and
monitor about the tolerence of the treatment.
Insulin glargine is a longworking insulin and is already registred in the
Netherlands for the treatment of diabetes. Risks for hypoglycemia are increased
with treatment with insulin. In this study the starting dosage of 10 units/day
will be used and considering the glycemic control of the participants in this
study, hypoglycaemic events are not expected with the start dosage of 10IU/day.
A previously used titration schedule for insulin dosage will be used, using the
self monitored fasting blood glucose as guideline for adjustment of the
insuline dosage, in steps of 2 units. This schedule has previously been used in
our group without showing adverse events. Therefore, expected safety risk will
be minimal. Nevertheless, there will be an regular contact with all
participants. In addition, the research physician is daily available for
questions etc.
We will try to make this study as bearable as possible for our participants.
All tests will be done by one researcher.