Insulin glulisine
glulisine, HMR1964, Lys(B3), Glu(B29) human insulin, B3-lysine B29-glutamic acid insulin
Insulin glulisine is a rapid-acting insulin analogue and the third of its class to reach the market, licensed in 2004. It carries two amino acid substitutions in the B chain and, unusually, is formulated without zinc, using polysorbate 20 as a stabiliser instead. It covers mealtime glucose and is used in pumps, and in practice performs essentially interchangeably with insulin lispro and insulin aspart.
Mechanism
Insulin glulisine differs from human insulin at two B-chain positions: asparagine at B3 is replaced by lysine, and lysine at B29 is replaced by glutamic acid. The two changes work in different ways. The B29 glutamate introduces a negative charge near the dimer-forming surface, destabilising self-association in the same general way as the aspartate substitution in insulin aspart. The B3 lysine substitution primarily improves chemical stability. Asparagine at B3 is prone to deamidation and to isoaspartate formation, and removing it makes the molecule more robust in a zinc-free environment.
That zinc-free formulation is what genuinely distinguishes glulisine. Other marketed insulin products use zinc to hold insulin in hexameric form for shelf stability. Glulisine is instead stabilised by polysorbate 20 and exists in the vial largely as monomers and dimers rather than hexamers, with the surfactant preventing aggregation and adsorption to surfaces. Because there is comparatively little hexamer to dissociate, absorption from the subcutaneous depot is fast from the outset: onset within 10 to 20 minutes, peak effect at around one hour, and total duration of three to five hours. Some early work suggested slightly faster absorption in people with obesity, where hexamer dissociation is a greater rate limitation, but the clinical significance of that has never been convincingly established. At the receptor, glulisine binds the insulin receptor with affinity comparable to human insulin, has broadly similar IGF-1 receptor affinity, and produces standard downstream signalling: receptor autophosphorylation, IRS-1/2 recruitment, PI3K-AKT activation, GLUT4 translocation, and suppression of hepatic glucose production.
What the research shows
Glulisine's registration programme did what it needed to: it demonstrated faster absorption than regular human insulin, lower postprandial glucose and a small HbA1c advantage in both type 1 and type 2 diabetes. The candour of the type 2 diabetes investigators in acknowledging that the clinical relevance of a 0.16 percentage point HbA1c difference remained to be established captures the honest position for the whole rapid-acting analogue class rather better than most summaries do. The type 1 diabetes trial added a genuinely useful practical finding (that giving glulisine immediately after a meal was as effective and safe as giving it before, and was the only arm without weight gain), which matters for people with unpredictable appetite, young children, and those with gastroparesis.
What the literature does not show is any advantage of glulisine over the other two rapid-acting analogues. Head-to-head and pooled analyses find lispro, aspart and glulisine essentially interchangeable in pharmacokinetics, HbA1c and hypoglycaemia, and Cochrane analyses do not distinguish them. Claims that the zinc-free formulation confers a clinically meaningful speed advantage, particularly in obesity, have not held up in outcome terms. Glulisine also has the thinnest evidence base of the three in pregnancy and in very young children, and it has never been the comparator arm of a major basal insulin outcome programme in the way aspart has. It carries no long-term complication or mortality data relative to regular human insulin, which is true of all rapid-acting analogues. In practice, selection between glulisine, lispro and aspart is a question of device, formulary and cost rather than pharmacology, and it is reasonable to say so plainly.
Evidence assessment
High-quality evidence
Approved by the FDA and in the European Union in 2004 on the basis of adequately powered randomised controlled trials in both type 1 diabetes (n=860) and type 2 diabetes (n=876). The registration evidence is regulator-grade. As with the other rapid-acting analogues, Cochrane systematic review evidence rates the clinical benefit over regular human insulin as minor, and the registration investigators themselves noted that the clinical relevance of the observed HbA1c difference remained to be established.
Tiers are applied consistently across the library and re-checked when new trials read out. Read the grading method.
Key studies
Insulin glulisine provides improved glycemic control in patients with type 2 diabetes Preclinical only
HbA1c fell slightly more with glulisine than with regular human insulin (−0.46 versus −0.30 percentage points), with lower postprandial glucose and comparable safety. The investigators themselves noted that the clinical relevance of a difference of this size remained to be established, an unusually candid qualification in a registration trial.
Optimized Basal-bolus insulin regimens in type 1 diabetes: insulin glulisine versus regular human insulin in combination with Basal insulin glargine Preclinical only
Premeal glulisine produced a greater HbA1c reduction than regular human insulin. Postmeal glulisine was comparably effective and safe, and was the only arm not associated with weight gain. Severe hypoglycaemia rates were similar across arms. The postmeal finding is practically useful for people with unpredictable food intake.
Short-acting insulin analogues versus regular human insulin for adults with type 1 diabetes mellitus Preclinical only
Only a minor benefit of short-acting insulin analogues over regular human insulin, with HbA1c differences around 0.15 percentage points and low certainty of evidence. No trial supplied data on long-term complications, mortality or quality of life. Applies equally to glulisine, lispro and aspart, which are not distinguishable from one another in pooled analyses.
Efficacy and Safety of Rapid-Acting Insulin Analogs in Special Populations with Type 1 Diabetes or Gestational Diabetes: Systematic Review and Meta-Analysis Preclinical only
Lispro, aspart and glulisine were at least as effective and safe as regular human insulin in these populations, with small effect sizes. Evidence specific to glulisine in pregnancy is notably thinner than for lispro and aspart, and glulisine does not have the dedicated randomised pregnancy trial that aspart has.
Safety
Hypoglycaemia is the dose-limiting adverse effect and, as with all rapid-acting analogues, occurs earlier after injection than with regular human insulin, so mistiming against a meal is a common precipitant. The short duration and absence of a persisting subcutaneous depot means that interruption of insulin pump delivery can produce diabetic ketoacidosis within a few hours. Weight gain, hypokalaemia, injection-site lipohypertrophy from inadequate site rotation and local injection reactions occur as with all insulins. Uncommon effects include systemic insulin allergy, lipoatrophy, peripheral oedema, transient refractive change when glucose falls rapidly, and clinically significant anti-insulin antibodies. Renal or hepatic impairment reduces insulin clearance and increases hypoglycaemia risk. Concurrent thiazolidinedione therapy increases fluid retention and heart-failure risk. Glulisine should not be mixed with any insulin other than isophane insulin, and even then only immediately before injection and never in a pump. Because the formulation is zinc-free and surfactant-stabilised, it behaves differently from other insulins in infusion sets and diluents, and manufacturer compatibility guidance is not interchangeable with that for other products. Recurrent hypoglycaemia produces hypoglycaemia unawareness with specific driving implications; UK drivers on insulin must comply with DVLA rules. Insulin is a high-alert medicine and mix-ups between rapid-acting and long-acting products are a documented cause of severe harm. Non-medical use by people without diabetes has caused deaths and permanent hypoglycaemic brain injury; rapid-acting analogues are the most dangerous in this respect because onset leaves little time to intervene.
Regulatory status
| Jurisdiction | Status |
|---|---|
| United Kingdom | Licensed by the MHRA for the treatment of diabetes mellitus in adults, adolescents and children aged 6 years and over. Prescription-only medicine (POM). Available as a rapid-acting insulin option within NICE guidance NG17 and NG28, without specific preference over insulin lispro or insulin aspart. |
| United States | FDA-approved in April 2004 for adults and children aged 4 years and above with diabetes mellitus for the improvement of glycaemic control. Regulated as a biological product since March 2020. Prescription-only. |
| WADA (sport) | Prohibited at all times, in and out of competition, under section S4 of the WADA Prohibited List covering insulins and insulin-mimetics. Athletes with diabetes require a Therapeutic Use Exemption. |
Questions
It is formulated without zinc, which is unusual among marketed insulins. Other insulins use zinc to hold the molecule as hexamers for shelf stability; glulisine is stabilised instead by polysorbate 20, a surfactant that prevents aggregation and surface adsorption. Because there is less hexamer to dissociate, absorption starts quickly. It also means glulisine's compatibility with diluents and infusion sets differs from other insulins and manufacturer guidance is not transferable.
Not to any degree that shows up in outcomes. Pharmacokinetic curves for the three rapid-acting analogues are close to superimposable in most studies, and pooled clinical analyses do not distinguish them for HbA1c or hypoglycaemia. Choosing between them is realistically a matter of device, formulary and cost.
The 860-participant type 1 diabetes trial included a postmeal arm and found it as effective and safe as premeal dosing, and it was the only arm without weight gain. That has genuine practical value where food intake is unpredictable. How and when any insulin is timed is a clinical decision for the prescribing team.
That substitution is about stability rather than speed. Asparagine at B3 is chemically vulnerable to deamidation and isoaspartate formation, which is a bigger problem in a zinc-free formulation where the protective hexamer structure is largely absent. Replacing it with lysine removes that degradation route. The kinetic acceleration comes mainly from the glutamate at B29.
The evidence for glulisine in pregnancy is thinner than for insulin aspart, which has a dedicated 322-participant randomised trial, or lispro, which has substantial observational experience. Meta-analyses of rapid-acting analogues in special populations include glulisine but with limited data specific to it. Insulin management in pregnancy should be directed by a specialist antenatal diabetes team.