Potential Advantages of a Basal–Bolus Regimen Using Insulin Glulisine as Prandial Insulin

Several interventional studies have demonstrated that achieving near-normal glycaemic control by means of intensive insulin therapy is the best strategy to avoid and slow the progression of chronic complications in type 1 and type 2 diabetes.^1–3 Interestingly, the benefits of intensive treatment seem to extend over time, at least in people with type 1 diabetes, as shown in the Epidemiology of Diabetes and Interventions and Complications (EDIC) trial.⁴ However, intensive insulin therapy using multiple daily injections (MDIs) or continuous subcutaneous insulin infusion (CSI

Several interventional studies have demonstrated that achieving near-normal glycaemic control by means of intensive insulin therapy is the best strategy to avoid and slow the progression of chronic complications in type 1 and type 2 diabetes.^1–3 Interestingly, the benefits of intensive treatment seem to extend over time, at least in people with type 1 diabetes, as shown in the Epidemiology of Diabetes and Interventions and Complications (EDIC) trial.⁴ However, intensive insulin therapy using multiple daily injections (MDIs) or continuous subcutaneous insulin infusion (CSII) increased the risk of severe hypoglycaemia about three-fold.1 One of the probable explanations behind this observation was the use of unphysiological insulin formulations, because both regular human insulin (RHI) and isophane insulin (NPH) are far from ideal as insulin replacements.

In the last 15 years, new insulin formulations have been coming into the market. First, short-acting insulin analogues (SAIAs) were developed to reproduce the physiological prandial insulin response, which is rapid, powerful and of short duration.⁵ More recently, long-acting insulin analogues (LAIAs) were introduced to replace basal insulin secretion, which is peakless, sustained and necessary to avoid excessive hepatic glucose production.⁵ Insulin glulisine is the most recently marketed SAIA – beside insulin lispro and insulin aspart – and is available to be used as prandial insulin in adults with type 1 and type 2 diabetes. In this article, the potential advantages of insulin glulisine will be discussed, as well as its role as prandial insulin in a basal–bolus regimen.

Insulin Glulisine – A New Formulation
The SAIA glulisine (rDNA origin, Sanofi-Aventis, Inc.) is produced by recombinant DNA (rDNA) technology using a non-pathogenic strain of Escherichia coli (K12). Insulin glulisine is a modified human insulin in which the amino acid asparagine at position B3 is substituted with lysine, and the amino acid lysine at position B29 with glutamic acid (see Figure 1). These changes allow insulin glulisine to dissociate rapidly after subcutaneous (SC) injection, which results in a fast absorption from tissue into circulation. Additionally, insulin glulisine uses polysorbate 20 as a stabilising agent instead of zinc, which is used with insulin lispro and insulin aspart. This unique formulation is associated with a reduced hexamer and dimer formation in comparison with other SAIAs, favouring a more rapid absorption.⁶ These characteristics of insulin glulisine result in a pharmacokinetic (PK) profile that closely mimics the normal insulin response after meals.

Compared with RHI insulin, glulisine has equivalent bioefficacy but faster absorption and lower mean residence time.⁷ In this study, insulin glulisine was also compared with insulin lispro, demonstrating similar PK and pharmacodynamic (PD) profiles.⁷ Insulin glulisine is well absorbed irrespective of the injection site, although a slightly more rapid absorption was demonstrated when administered into the abdominal area.⁸ The short-acting profile of insulin glulisine is also maintained across different ethnic groups, paediatric patients and subjects with altered renal function.^9–11

The potential advantages of insulin glulisine in obese subjects deserve a separate commentary. It is well known that increased SC thickness alters insulin absorption, shifting the time–action profile curves to the right.¹² In a phase I, randomised, euglycaemic clamp study, a group of non-diabetic obese subjects received a single injection of insulin glulisine, insulin lispro or RHI (0.3U/kg SC).¹³ As expected, time to onset of glucose infusion was shorter, and maximal glucose insulin rates were greater for insulin glulisine and insulin lispro compared with RHI (see Figure 2). However, insulin glulisine differed from insulin lispro and RHI in that there was no significant correlation between skin thickness and body mass index (BMI) and the time to maximal concentration. In other words, it seems that insulin glulisine maintains its rapid-acting properties more consistently in obese individuals compared with insulin lispro and RHI.

Insulin Glulisine – Efficacy and Safety In Type 1 Diabetes
There are not many published studies comparing different SAIAs with RHI using the same LAIA in all arms. When insulin glargine is used in the evening as basal insulin by people with type 1 diabetes, no differences in reducing glycated haemoglobin (HbA1c) were found after 26 weeks between insulin lispro or insulin aspart, administered just before meals, and RHI injected 30 minutes before the meals, when all patients were instructed similarly in carbohydrate counting.¹⁴ However, in a recent published study with a larger number of patients but of shorter duration (only 12 weeks), insulin glulisine administered just before meals demonstrated a significantly greater reduction in HbA1c compared with RHI injected 30 minutes before meals, or insulin glulisine after – or 20 minutes after starting – the meal (-0.26% versus IRH -0.13% versus glulisine after the meal -0.11%; p<0.01).¹⁵

Although the clinical relevance of this finding is a matter of discussion, this study suggests that insulin glulisine may be injected either immediately before or at the end of a meal, increasing the flexibility of insulin administration.

In another multicentre, randomised, parallel-group study, the efficacy and safety of insulin glulisine was compared with insulin lispro in adults with type 1 diabetes, using insulin glargine once daily in the evening as basal insulin.¹⁶ In this study, both SAIAs, which were administered 0–15 minutes before meals, achieved similar reductions of HbA1c at 26 weeks (-0.14±0.04%), although higher total daily insulin doses were necessary in patients using insulin lispro. At the end of the study basal insulin dose remained unchanged with insulin glulisine but increased with insulin lispro (adjusted mean change +0.12IU in the insulin glulisine group versus +1.82IU of insulin glargine; p=0.0001), although prandial insulin needs were similarly reduced in both arms (adjusted mean change of -1.07IU in the insulin glulisine group and -0.81IU in the lispro group, not significant (NS)). There were no differences between the groups in symptomatic hypoglycaemia (overall, nocturnal and severe) or in weight changes.

The clinical relevance of the small divergences in basal and total daily doses found in this study remains to be established. An increasing proportion of people with type 1 diabetes use insulin pump therapy. CSII is considered as an alternative to MDI for those patients who are unable to achieve the glycaemic goals.¹⁷ To assess the safety of insulin glulisine in pumps, a 12-week multinational, parallel, randomised, controlled trial using insulin aspart as comparator was designed in adults with type 1 diabetes (n=59) under CSII therapy.¹⁸ Regarding safety issues, only seven patients (20.7%) in the insulin glulisine group reported at least one episode of unexplained hyperglycaemia (>350mg/dl) compared with 14 patients (40.0%) in the insulin aspart group, although this difference was not statistically significant. Additionally, four patients (13.8%) in the insulin glulisine group versus eight patients (26.7%) in the insulin aspart group experienced at least one catheter occlusion; however, this difference,again, was not statistically significant. Other evaluated variables, such as change of HbA1c from baseline, self-monitored blood glucose profiles, mean total insulin dose and frequency of hypoglycaemia, were comparable between groups. Therefore, insulin glulisine can be considered a safe alternative to other SAIAs for patients with type 1 diabetes on insulin pump therapy. Although the catheter change rate and time between catheter changes were comparable between groups, an emerging question from this study is whether insulin glulisine, formulated with polysorbate 20 as stabilising agent and in solution mainly as dimers and monomers, is more stable in pumps than insulin aspart (or insulin lispro), which in solution aggregates forming hexamers with zinc to achieve sufficient stability. Further studies in this field should clarify this question.