A recent meta-analysis has suggested that combining long-acting insulin with a GLP-1 (glucagon-like peptide 1) agonist may be the “most effective” drug treatment for type 2 diabetes, meaning the best blood glucose (BG) control with the fewest side effects.
GLP-1 is a hormone secreted by the intestines, and an agonist is a compound that stimulates another one. (An antagonist does the opposite.) Examples of GLP-1 agonists approved in the United States include exenatide and liraglutide.
A meta-analysis is a study in which the authors combine data from previously published research. By so doing, they increase the number of subjects, which can make the statistical results more meaningful. For example, if there were 100 studies that each had only 10 patients, each individual study couldn’t prove much because of the small number of people tested. But combining the studies would result in 1000 patients studied.
In the current study, the combined studies included 4,300 participants. The patients began with A1c’s between 8 and 13, and the target was an A1c level below 7. The proportion achieving the target was higher in the group on the combination treatment, but the results are presented in comparison with the other treatments, so it’s not clear how many actually achieved that goal.
The lead author, Dr. Ravi Retnakaran of the University of Toronto, kindly sent me a copy of the study, which noted that some of the GLP-1 drugs tested in the studies included in the meta-analysis were ones that aren’t yet available in the United States, although one, albiglutide received US approval last spring. Approval of another, lixisenatide has been delayed.
Individual studies had suggested that this combination of long-acting (basal) insulin and a GLP-1 agonist was more effective than “standard treatment,” but the meta-analysis confirmed the individual results and added strength to their conclusions.
Some people criticize meta-analyses because each individual study often uses different patient populations or different study durations or different endpoints. Nevertheless, such studies can provide evidence and suggest ideas for future research.
This combination of a basal insulin and a GLP-1 agonist makes a lot of sense. When diet and exercise and first-line drugs like metformin fail to produce satisfactory BG control, type 2 patients are often prescribed a basal insulin. This usually brings BG levels down, but some patients gain weight on insulin, and because most are overweight in the first place, this is not good.
Furthermore, when the treatment is too aggressive, many patients suffer from hypoglycemia (low BG levels), and this is especially a problem in elderly patients and in those with some cognitive impairment and no caregiver to monitor the insulin injections.
When the basal insulin fails to control BG levels sufficiently, a mealtime (bolus) insulin is often added to the treatment. This requires calculating the level of carbohydrate eaten at each meal and injecting accordingly, which is an additional burden, especially to patients with limited cognitive abilities.
The GLP-1 agonists, on the other hand, have effects that are complementary to those of insulin. Although they do stimulate insulin release, they do so only when BG levels are elevated, so they don’t cause hypoglycemia as oral insulin-stimulating drugs like the sulfonylureas do.
They also slow down gastric emptying, which means that as with low-glycemic-index foods, the meal doesn’t hit the intestine all of a sudden, requiring a huge release of insulin all at once. Some people with type 2 are able to produce enough insulin when the demand is not too great.
They can result in weight loss, the opposite of the effect of insulin in some people.
Finally, and I think very importantly, the GLP-1 agonists inhibit the release of glucagon. Glucagon is a hormone that does pretty much the opposite of what insulin does. Insulin lowers BG by stimulating its uptake by muscle and fat, and glucagon increases BG by releasing glucose from the liver. Insulin is secreted by the beta cells in the pancreas when BG levels are high, and glucagon is secreted by the alpha cells in the pancreas when BG levels are low.
Normally, after a meal, the insulin stimulated by the carbohydrate in the meal stops the liver from releasing glucose. But because of the insulin resistance found in type 2, the insulin doesn’t stop the liver’s release of glucose in people with type 2, so we have to deal with not only the glucose from the meal but the glucose released from the liver, and our BGs go up too much after meals.
Hence by combining these two drugs with complementary effects, one may get the benefits of both and limit the side effects. If the GLP-1 agonist is helping to control BG, then one doesn’t need to inject as much insulin, for example.
A basal insulin is said to control primarily fasting and premeal BG levels. This is true, but in a sense it can also help to control postmeal levels. If your premeal BG is 75 mg/dL and you go up 30 points from a meal, then your postmeal level will be only 105, an acceptable number within the normal range. But if your premeal level is 120, then those 30 points would send you to 150, not astronomical, but higher than normal.
The GLP-1 agonists control primarily postmeal BG levels. So again, their results are complementary.
Hence this sounds like a sensible approach to BG control in type 2 patients who can’t control with other drugs or with diet and exercise. But there are several limitations.
The main limitation is cost. The GLP-1 agonists are new and very expensive. The basal insulins are also expensive, although generic versions are in the works. “Both insulin agonists and insulin analogues are among the most expensive in diabetes care,” wrote endocrinologist Dr. John Buse of the University of North Carolina in a linked commentary. People with good insurance or high incomes would have no problem affording the combination of the two, but for others it could be a real barrier.
Various companies are working on combination drugs, but as they will be new, they will also likely be expensive. One has been recommended by European authorities, and a final decision is expected next month.
Another limitation is the fact that both these drugs have to be injected. I don’t think injections are a big deal, but some people are afraid of needles. Injections do require additional equipment (syringes and needles), and the lifetime of injected drugs is usually shorter than the lifetime of tablets.
Some people who can’t control their BG without strong drugs may be on the low-fat high-carbohydrate diets recommended for years by the American Diabetes Association. In that case, it would make sense to try a lower-carb, higher-fat diet before moving to stronger drugs.
I’ve been on a low-carb diet for almost 20 years, and although I’ve begun injecting insulin because I was not happy about having a hemoglobin A1c level of about 6, I’ve actually lost weight while injecting insulin, which means that using insulin doesn’t invariably lead to weight gain, especially when you’re not on a high-carb diet.
However, if for whatever reason, a low-carb diet is not possible for you and if you’re able to afford these drugs, it sounds as if the combination might be worth a try.