Effect of the GLP-1 receptor agonist exenatide on impaired awareness of hypoglycemia in type 1 diabetes

Iatrogenic hypoglycemia is the most frequent, acute, complication of insulin
therapy in people with type 1 diabetes. These patients experience on average
2-3 hypoglycemic events per week and one severe event requiring external
assistance every year. Accurate and timely recognition of the typical symptoms
of falling glucose levels are of pivotal importance to prevent severe
hypoglycemia. However, 25-40% of patients with type 1 diabetes have lost the
capacity to timely detect hypoglycemia, a condition referred to as impaired
awareness of hypoglycemia (IAH) that specifically increases the risk for
severe, potentially hazardous, hypoglycemia up to sixfold. IAH is usually the
end-result of a process of habituation to recurrent hypoglycemia and meticulous
avoidance of hypoglycemia for 2-3 weeks can reverse this process, thus
ameliorating symptomatic awareness of hypoglycemia. However, the often
coexistent worsening of glycaemic control precludes such an approach as a
viable treatment option.
Many patients with IAH suffer from high glucose variability, in which both
hypoglycemic events and hyperglycemic excursions occur at high frequency.
Indeed, hyperglycemic excursions often develop after recovery from
hypoglycemia, usually because of ingesting (too many) carbohydrates in the
context of hyperglucagonemia. In turn, correction of such a hyperglycemic
excursion too aggressively, especially in the context of failing
counterregulatory hormone responses and impaired awareness, creates a high risk
for hypoglycemia.
More stability in day-to-day glucose control with reduced hypoglycemic exposure
may be important to resolve IAH. Paradoxically however, improvement in
awareness of hypoglycemia may also stimulate hyperglycemic excursions after
hypoglycemia, because the stronger hunger response (as a component of improved
awareness) may enhance the ingestion of carbohydrates, thus worsening overall
glycaemic control. Glucagon-like-peptide-1 receptor agonists (GLP-1RAs),
developed for the treatment of type 2 diabetes, may reduce glucose variability,
because these agents reduce postprandial glucose excursions without increasing
the risk of hypoglycemia. These agents also reduce body weight, because they
delay gastric emptying, cause early satiety and reduce food intake. Recently,
treatment with GLP-1RAs in people with type 1 diabetes improved postprandial
and overall glucose control and reduced insulin requirements and glucose
variability. These agents have not been tested in patients with type 1 diabetes
and IAH. We posit that treatment with GLP-1ra (added to insulin therapy) will
reduce glucose variability and the incidence of hypoglycemia, so that IAH can
be improved. Moreover, since such treatment will limit food intake in the
recovery period after hypoglycemia, we posit that overall glycemic control will
not deteriorate.

Primary objective: to investigate the effect of treatment with the GLP-1ra
exenatide on the awareness of and counterregulatory hormone responses to
hypoglycemia in people with type 1 diabetes and impaired awareness of
hypoglycemia.

Secondary objective: to investigate the effect of treatment with the GLP-1ra
exenatide on the glycemic recovery from hypoglycemia and the post-hypoglycemic
glucose excursion in people with type 1 diabetes and impaired awareness of
hypoglycemia.

Randomized double-blind placebo-controlled cross-over trial
2 periods of 6 weeks treatment with exenatide versus placebo
5 days blinded continuous glucose sensor monitoring in final week of each
treatment period
Hyperinsulinemic hypoglycemic glucose clamp at the end of each treatment period
to quantitate symptomatic awareness of hypoglycemia

6 weeks treatment with the glucagon-like peptide receptor agonist exenatide or
placebo according to the following regimen:
weeks 1-2: exenatide BID 5 microgram sc
weeks 3-6: exenatide BID 10 microgram sc

The extent of the burden of participating to this study include the following:
- Total time spent: 16 weeks: 2 periods of 6 weeks and a 'wash-out' of 4 weeks.
Prior to randomization, participants will visit the clinic for a screening
examination. During both treatment periods, two telephone visits and one
regular visit (for inserting the glucose sensor) will be scheduled.
- During treatment periods: subcutaneous injection of study medication
(exenatide or placebo) twice daily. A special designed injection pen,
comparable to conventional insulin pens, is used for these injections.
- Final week of treatment periods: continuous glucose sensor measurement for 5
days
- Final day of treatment periods: 5-hour experiment involving hyperinsulinemic
hypoglycemic glucose clamp (8AM-1PM). This experiment involves stepwise
lowering of plasma glucose levels to 2.5 mol/l with intravenous infusions of
insulin and glucose 20% according to a titration protocol. Hypoglycemia
normally elicits typical symptoms, including sweating, feeling hungry,
palpitations, trembling and tingling, and - to a lesser extent - concentration
disturbances, tiredness, difficulty speaking and visual disturbances. Such
hypoglycaemia does not cause severe cognitive dysfunction, coma or epileptic
insults. This hypoglycemic phase does not last longer than 45 minutes. The
recruited patients' condition of impaired awareness means that they will
probably remain largely unaware of these typical symptoms, although we
hypothesize that exenatide may enhance these symptoms.

Potential risks and adverse events associated with participation:
1. during the 6-week treatment periods:
- Nausea and vomiting. These are well-known adverse effects associated with the
use of GLP-1 receptor agonists that may affect up to 20% of patients, but much
less when lower starting doses are used.
- Hypoglycaemia. The risk of hypoglycaemia may be increased when exenatide is
started as prior studies have shown that insulin requirements decrease; to
mitigate this risk, the dose of insulin will be decreased by 20% in the first
week of each treatment period and further adjusted on the basis of frequent
glucose self-measurements.
- Hyperglycaemia. This is a risk largely associated to diabetes itself, but
could potentially increase in the placebo treatment arm as a consequence of the
20% lowering of insulin (see above).
- Local skin reactions or inflammation. This can be a consequence of
subcutaneous injection per se or (local) allergy, the risks of which are low.
- Rare adverse reactions: anafylactic reactions and acute pancreatitis
2. Continuous glucose sensor monitoring:
- Local skin reactions or inflammation. This can be a consequence of
subcutaneous insertion of the sensor, but this risk is very low.
3. Hyperinsulinemic hypoglycaemic glucose clamp
- Symptoms of hypoglycemia. These are anticipated as per protocol (see above)
- Deeper hypoglycemia than targeted. This risk is minimal because of the
simultaneous infusion of glucose and glucose measurements every 5 minutes.
- Hematoma as a result of venous cannulations
- Flebitis as a consequence of glucose 20% infusion. This occurs occasionally
but is usually self-limiting
- Disruption in glycemic control after the experiment. Glucose levels may show
more fluctuations, but this is usually transient and can be mitigated by
intensifying glucose self control.

It is essential for patients with type 1 diabetes that they retain the capacity
to timely and accurately detect hypoglycemia. Patients who will be recruited to
this study have lost this capacity and are consequently at high risk of
developing severe hypoglycemia. In this project, it will be investigated
whether this loss of hypoglycemic awareness can be treatment with exenatide. In
this light, we believe that the burden of 16 weeks of experimental treatment
and the two experimental days involving hypoglycemic glucose clamps are
reasonable.