The second long-acting insulin analogue developed by Novo Nordisk, insulin degludec is in many ways similar to its predecessor, insulin detemir. It derives its prolonged duration of action from a fatty-acid tail and it is associated with very high plasma insulin levels as a result of albumin binding. However, whereas insulin detemir can hardly be called long-acting in the true sense, degludec does indeed have a very long duration of action of up to 42 hours, leading to relatively stable steady-state plasma insulin levels after several days of use. Unfortunately, as with all analogues, the clinical benefits of these improved pharmacokinetics have been rather limited. Insulin degludec may be associated with slightly lower nocturnal hypoglycaemia rates than insulin detemir or insulin glargine but independent meta-analyses are still lacking, and the FDA has postponed marketing approval pending a trial looking into the cardiovascular safety of the drug.
Figure 1. Composition of insulin degludec compared to human insulin.Like its sister-molecule insulin detemir, insulin degludec derives its prolonged action from the attachment of a fatty acid tail, in this case hexadecandioyl, to the 29th position of the B-chain through a glutamic acid linking molecule, after removal of Threonine at the B30 position (Figure 1).
The exact mechanism of prolongation of action has not been conclusively established. The company states that the formation of multihexamers at the site of injection is responsible for a slow release of the insulin into the circulation. The observation that insulin degludec achieves much higher plasma levels than NPH insulin at steady state (up to 4000 pmol/l, as compared to peak NPH insulin levels of 150-200 pmol/l), coupled with the fairly rapid appearance of degludec in the plasma following injection suggests that - as with insulin detemir - the fatty acid tail leads to considerable binding of degludec to proteins both in the subcutaneous compartment and in the plasma which forms a depot for the slow release of the insulin. Thus, the postulated predominant mechanism has not been independently confirmed.
Unlike Insulin glargine, which cannot be mixed, insulin degludec can be used in a fixed combination with aspart known as IDegAsp.
Pharmacokinetics and pharmacodynamics
After a single injection of insulin degludec, peak insulin levels are reached after 8 to 12 hours . Due to its prolonged terminal half-life of 25 hours, steady state plasma levels can be realised with repeated once-daily injections; under these circumstances the peak plasma insulin levels are about 30% higher than the nadir (lowest level just before the next injection).
The pharmacodynamics of degludec in patients with type 2 diabetes follow a similar pattern, with fairly constant glucodynamic effects over 24 hours, with a very slight peak action at around 12 hours after injection. Insulin degludec has a linear dose-response curve (i.e. a doubling of the dose leads to a doubling of the glucodynamic effect).
As with detemir, in study glucose clamp settings, degludec has been associated with less variability in action than insulin glargine, but this has not translated in less variability in fasting plasma glucose.
In type 1 diabetes, several company sponsored trials have compared the use of insulin degludec to insulin glargine. The general pattern that arises from these trials, which was also reported in a company sponsored meta-analysis, is that degludec gives similar glycaemic control with equal rates of overall hypoglycaemia but less nocturnal hypoglycaemic episodes than insulin glargine.
There was no difference in the occurrence of severe hypoglycaemia. Of note, an analysis demanded by the FDA found that the difference in nocturnal hypoglycaemia disappeared when the time period used to define nocturnal was shifted by two hours, suggesting that this difference may be partly attributable to differences in the timing of insulin administration in these trials (degludec was invariably administered with the evening meal, whereas glargine was usually administered at bedtime). In a separate trial, the long duration of action of degludec did seem to allow for somewhat more variable timing of injection during the day.
In type 2 diabetes, use of insulin degludec as a once daily basal insulin resulted in comparable glycaemic control as insulin glargine. In this group, the already very low hypoglycaemia rates (less than 2 episodes per patient-year) were similar between groups, with slightly less nocturnal hypoglycaemia with insulin degludec.
When used in a basal bolus regimen in those with longer duration of type 2 diabetes, again HbA1c was similar compared to insulin glargine; however, overall and nocturnal hypoglycaemia rates were slightly lower (a difference of about 2 episodes per patient-year).
Due to its long duration of action the use of insulin degludec three times a week rather than once a day has been investigated, and was indeed shown to be feasible in patients with type 2 diabetes. This could result in social benefit, for example when other people are needed to administer the injection, but confers no other benefit in terms of glycaemic control or hypoglycaemia rates.
The phase II and phase III trials of insulin degludec assessed cardiovascular safety of degludec in comparison to comparator drugs. The company analysis presented to the FDA found no increased risk for cardiovascular events. However, analyses requested by the FDA, which used different definitions for cardiovascular events, found a non-significantly increased risk of Major Adverse Cardiovascular Events, with hazard ratio's of about 1.4 to 1.6. Since the registration dataset was underpowered to draw any firm conclusions, the FDA requested the company to gather additional cardiovascular data from a dedicated cardiovascular outcomes trial.
When looking at all the long-acting insulin analogues currently available, the bulk of the evidence suggests that their advantages are relatively modest when compared to conventional NPH insulin. This is particularly true when used as first-line basal insulins in type 2 diabetes, since their main advantage (lower hypoglycaemia rates) is limited to relatively few individuals due to the low baseline hypoglycaemia rate in this patient group.
Given the fact that the benefits of insulin detemir over NPH insulin are in some doubt, and the benefits of insulin glargine compared to NPH insulin are modest, it would have been relevant to have compared insulin degludec directly to NPH insulin. As it stands, we can conclude that it may be a bit better than NPH insulin, since it seems slightly better than insulin glargine, but whether this difference is important enough to warrant the far higher costs might be questioned.