Metformin didn’t work – what’s next? Adding a second treatment in uncontrolled type 2 diabetes

This study compared the effects of adding either basal insulin or a glucagon-like peptide-1 (GLP-1) receptor agonist to the treatment regimens of patients with uncontrolled type 2 diabetes.

Type 2 diabetes is a progressive disease. While patients may be able to control their disease with diet and lifestyle changes at first, eventually most will need to add on some form of glucose-lowering drug therapy. Metformin (Glucophage) is generally the first choice, however, eventually other treatments will need to be added. A sequence of treatments has not been widely accepted for type 2 diabetes, though the method of administration is often a factor, as injections (such as for insulin therapy) are considered to be more difficult than other forms of therapy.

GLP-1 receptor agonists, an injectable therapy, decrease blood glucose levels by increasing insulin production and slowing the emptying of the stomach. Both GLP-1 receptor agonists and basal insulin (long-acting insulin delivered once a day) have been shown to effectively lower blood glucose levels in type 2 diabetics. However, the majority of studies have only included patients with relatively well-controlled diabetes, with HbA1c (an average measure of blood glucose over 3 months) levels of around 8%. It is not clear which treatment, basal insulin or GLP-1 receptor agonists, is more useful for patients with poorly controlled blood glucose, particularly since insulin doses can be adjusted based on blood glucose levels, while GLP-1 receptor agonists are delivered at fixed doses.

The current study compared the effectiveness of insulin glargine (Lantus; a form of basal insulin) and liraglutide (Victoza; a GLP-1 receptor agonist) in patients with poorly controlled type 2 diabetes. Patients had HbA1c levels between 7.5% and 12%, and all had been treated with metformin for at least 3 months. 978 patients were randomly assigned to receive either insulin glargine (489 patients) or liraglutide (489 patients). Patients were treated for 24 weeks. HbA1c was measured at the start of the study, at week 12, and at week 24. During the three days prior to these clinic visits, patients measured their blood glucose levels seven times over the course of the day: before and two hours after each meal, and prior to bedtime.

At the end of 24 weeks, 48.4% of the insulin glargine patients and 45.9% of liraglutide patients had achieved HbA1c levels of less than 7%. However, the average fasting glucose level (first thing in the morning) was significantly lower in the insulin glargine patients (6.2 mmol/l) compared to the liraglutide patients (7.9 mmol/l), as was the average daily glucose level (8.1 mmol/l for insulin glargine; 8.6 mmol/l for liraglutide).

65.9% of the liraglutide patients and 50.2% of the insulin glargine patients reported an adverse (negative) effect from their treatment. The liraglutide patients experienced a higher incidence of gastrointestinal side effects, including nausea, diarrhea, and vomiting. Hypoglycemia (dangerously low blood glucose) was reported by 45% of the insulin glargine group, compared to 18% of the liraglutide group. Body weight increased by 2 kg in the insulin glargine group, and decreased by 3 kg in the liraglutide group. 

This study concluded that the addition of an injectable therapy, either basal insulin or a GLP-1 receptor agonist, can improve blood glucose control in patients with type 2 diabetes not controlled by oral therapies such as metformin.

This study was funded by Sanofi, Sanofi also market insulin glargine under the trade name Lantus.