Back grey_arrow_rt.gif
Insulin degludec, an ultra-long-acting basal insulin, once a day or three times a week versus insulin glargine once a day in patients with type 2 diabetes: a 16-week, randomised, open-label, phase 2 trial - pdf attached
  Download the PDF here
The Lancet, Early Online Publication, 10 March 2011
Ultra-long-acting insulins for a lifestyle-related pandemic - commentary
Yogish C Kudva aEmail Address, Ananda Basu a Endocrinology, Mayo Clinic College of Medicine, Rochester, MN 55905, USA
Type 2 diabetes mellitus, a chronic, common metabolic disorder of pandemic proportions, typically affects individuals in their 50s and is associated with devastating complications.1, 2 Long-duration randomised trials, with clinical event endpoints, show that glycaemic control is difficult to achieve and maintain and that the effect on macrovascular complications is small.3 Subsequently, an increasing number of therapeutic options have become available, but randomised trials that have sought to show an improvement in clinical macrovascular events (through tight glycaemic control by combination antihyperglycaemic therapy) have yielded results that are concerning and disparate. The ACCORD study showed an increase in mortality in patients with tight control,4 whereas ADVANCE5 showed an improvement in diabetes-associated endpoints, especially renal outcomes but without benefit to cardiovascular outcomes (as in the Veterans Affairs Diabetes Trial6). Therefore the US Food and Drug Administration has an additional requirement for newer antidiabetic agents: the evaluation of effect on cardiovascular endpoints.7
It is a challenge to achieve and maintain satisfactory glycaemic control in type 2 diabetes for various reasons: the need for multiple agents, adherence, costs, convenience of drug administration, efficacy, and a side-effect profile.8 Patients with diabetes mellitus spend more than 2 h a day on self care, which includes glucose monitoring, sorting medications, deciding about insulin dosing, and dietary decisions.9 Therefore a recent trend in many disorders, including type 2 diabetes, has been the introduction of long-acting drugs that could be administered less than daily to improve adherence while minimising side-effects. In this age of multitasking, which imposes an additional burden on human memory, dosing interval is especially important.
Insulin degludec is a unique form of insulin that is ultra-long-acting and the latest in the series of insulin analogues and insulin-receptor ligands developed to improve options, safety, and efficacy. Insulin degludec has sufficiently reproducible pharmacokinetics to enable dosing intervals of more than 24 h. Doses given three times a week might improve adherence, improve glycaemic control without an increase in hypoglycaemia, and cause less disruption to the patient's lifestyle.
In The Lancet, Bernard Zinman and colleagues10 report a 28-site study (USA, Canada, South Africa, and India) with insulin degludec. This phase 2 study was open label, in four groups of insulin-naive patients with inadequately controlled type 2 diabetes who were randomised for 16 weeks; all patients were taking metformin. The interventions were different formulations of insulin: degludec or glargine (a long-acting insulin analogue). A unique feature of this study is the head-to-head comparison of insulin degludec with insulin glargine, an unusual circumstance of one company funding a head-to-head trial with a competitor's drug. Insulin degludec was given once a day or three times a week; insulin glargine was given once a day. There were two different starting doses of the once daily insulin degludec.
In this proof-of-concept study, there was little difference in the primary endpoint of HbA1C between groups (HbA1C 7·2—7·5%). Other endpoints (difference in HbA1C, final mean fasting glucose, mean weekly insulin dose, rate of hypoglycaemia, adverse events, antibodies to insulin degludec, and cross-reactive antibodies) were also not different across groups. Mean HbA1C lowering with the addition of basal insulin alone to oral agents of 1·5% is greater than we commonly observe in routine clinical practice, and is probably related to the Hawthorne effect of increased self-awareness, increased surveillance, and the frequent visits demanded by trials. More than 80% of patients completed Zinman and colleagues' trial.
Future trials and clinical practice in diabetes will involve combination of several agents from multiple classes, chosen for efficacy, safety, and convenience. Drugs for weekly administration schedules will probably become available in the future, including long-acting exenatide and insulin degludec.11 As therapy for type 2 diabetes has increased in complexity, advances in several fields have enabled longer lifespan for those with the disease. This extension has resulted in the observation and reporting of cancers in patients with type 2 diabetes. Although the evidence is much debated, high circulating concentrations of insulin might increase the risk of neoplasia, an effect referred to as the insulin-supply hypothesis. Several observational studies and many editorials have addressed the complexity of the data, which reiterates the importance of keeping this issue on the radar.12 It is extremely worthwhile to remember that therapeutic lifestyle changes are inexpensive and favourable on a risk—benefit basis and need persistent re-emphasis, as is being done for the financial benefits accruing to patients from their employer and insurance by doing so. This option is not often thought about but needs resurrection, even as the numbers of long-acting drugs for type 2 diabetes increase.
We declare that we have no conflicts of interest.

Insulin degludec is a new basal insulin that forms soluble multihexamer assemblies after subcutaneous injection, resulting in an ultra-long action profile. This study aimed to assess efficacy and safety of insulin degludec injected once a day or three times a week compared with insulin glargine once a day in insulin-naive people with type 2 diabetes, who were inadequately controlled with oral antidiabetic drugs.
In this 16-week, randomised, open-label, parallel-group phase 2 trial, participants aged 18—75 years with type 2 diabetes and glycosylated haemoglobin (HbA1C) of 7·0—11·0% were enrolled and treated at 28 clinical sites in Canada, India, South Africa, and the USA. Participants were randomly allocated in a 1:1:1:1 ratio by computer-generated block randomisation to receive insulin degludec either once a day or three times a week or insulin glargine once a day, all in combination with metformin. Investigators were masked to data until database release. The primary outcome was HbA1C after 16 weeks of treatment. Analyses were done by intention to treat. This trial is registered with, number NCT00611884.
Of 367 patients screened, 245 were eligible for inclusion. 62 participants were randomly allocated to receive insulin degludec three times a week (starting dose 20 U per injection [1 U=9 nmol]), 60 to receive insulin degludec once a day (starting dose 10 U [1 U=6 nmol]; group A), 61 to receive insulin degludec once a day (starting dose 10 U [1 U=9 nmol]; group B), and 62 to receive insulin glargine (starting dose 10 U [1 U=6 nmol]) once a day. At study end, mean HbA1C levels were much the same across treatment groups, at 7·3% (SD 1·1), 7·4% (1·0), 7·5% (1·1), and 7·2% (0·9), respectively. Estimated mean HbA1C treatment differences from insulin degludec by comparison with insulin glargine were 0·08% (95% CI -0·23 to 0·40) for the three dose per week schedule, 0·17% (-0·15 to 0·48) for group A, and 0·28% (-0·04 to 0·59) for group B. Few participants had hypoglycaemia and the number of adverse events was much the same across groups, with no apparent treatment-specific pattern.
Insulin degludec provides comparable glycaemic control to insulin glargine without additional adverse events and might reduce dosing frequency due to its ultra-long action profile.
Prevalence of type 2 diabetes and diabetes-related health complications and mortality continue to increase.1—3 Despite availability of many therapies, many people with diabetes are unable to reach guideline-recommended rates of glycosylated haemoglobin (HbA1C).4—6
Insulin degludec is an ultra-long-acting insulin in clinical development. The ultra-long action profile of this insulin is mainly attributable to formation of soluble multihexamers at the injection site, from which monomers gradually separate and are absorbed into the circulation, resulting in a flat and stable pharmacokinetic profile at steady state.7 These features suggest that the risk of hypoglycaemia might be reduced and clinical effectiveness might be achievable with dosing three-times a week in people with type 2 diabetes who were previously insulin-naive, which could help with early initiation of and adherence to insulin treatment.
This clinical proof-of-concept trial aimed to assess efficacy and safety of insulin degludec once a day or three times a week compared with insulin glargine once a day, in combination with metformin, in insulin-naive people with type 2 diabetes who were inadequately controlled on oral antidiabetic drugs.
In this exploratory, clinical proof-of-concept trial in insulin-naive people with type 2 diabetes, insulin degludec provided once a day or three times a week as add-on to metformin did not differ from insulin glargine in terms of glycaemic control, with no apparent treatment-specific patterns or clustering of adverse events.
Physiological replacement of basal insulin in patients with type 1 and type 2 diabetes is challenging and remains an elusive goal.17 Presently, there are two basal insulin analogues with improved subcutaneous absorption pharmacokinetics in clinical practice (insulin glargine and insulin detemir).18 Although both drugs provide very similar glycaemic control to neutral protamine hagedorn insulin with fewer hypoglycaemic events, they remain less than ideal basal-replacement strategies.19 Hence, many patients are not treated optimally.20 Insulin degludec was developed as a basal insulin with pharmacokinetic properties resulting in an ultra-long action profile.7 On the basis of this protracted profile, new applications of this basal insulin such as alternate-day or three times a week injection become possible (panel).
Physiological replacement of insulin in diabetes is challenging and remains an elusive goal, as evidenced by the many patients who do not achieve recommended treatment targets. In basal insulin replacement, the presently used insulin analogues glargine and determir have some advantages over human neutral protamine hagedorn insulin. Insulin degludec has a half-life longer than 24 h7 and was developed to provide a more appropriate basal pharmacokinetic profile. In this proof-of-concept phase 2 study in patients with type 2 diabetes, insulin degludec provided much the same glycaemic control as did insulin glargine when injected once a day. Because of its ultra-long action profile, insulin degludec could also be injected three-times weekly and still provide comparable glycaemic control to insulin glargine once a day. This new basal insulin analogue might be a valuable addition to clinical practice.
In this trial, total weekly doses were much the same for three treatment groups but were higher in the insulin degludec once a day group B at the end of trial. The low rates of hypoglycaemia reported in this study might be partly attributed to the short duration of diabetes in this cohort. The higher mean reported rate of hypoglycaemic episodes in the three-dose per week insulin degludec group than in the other groups might be because larger doses of insulin had to be given at every injection to cover the weekly insulin requirements. However, the higher hypoglycaemic event rate in this group was not statistically significant compared with the other treatment groups and might be partly attributable to two participants having nine and six hypoglycaemic episodes each. Irrespectively, the proportion of participants who had one or more hypoglycaemic episodes was the same in the insulin degludec three times a week group and the insulin glargine group.
Ours is a novel clinical study reporting the use of a three dose per week insulin regimen in the treatment of type 2 diabetes, a regimen that was made possible because of insulin degludec's ultra-long action profile and unique mechanism of protraction. A three times a week, weekend-off, dosing regimen might appeal to some people with type 2 diabetes who are inadequately controlled on oral antidiabetic drug treatments, potentially helping with acceptance and early initiation of insulin therapy. The applicability and acceptance of such a dosing regimen needs to be assessed.
This proof-of-concept trial has several limitations inherent to phase 2 regulatory studies. Caution should be exercised in drawing of firm conclusions from the data. The open-label design that was used to accommodate the different insulin-injection systems could have affected efforts to attain blood glucose control and reporting of hypoglycaemia and adverse events.
In summary, findings from this trial show that insulin degludec can provide equivalent glycaemic control to insulin glargine without new or increased rates of adverse events in insulin-naive people with type 2 diabetes. The safety, efficacy, and optimum use of treatment regimens for insulin degludec need to be established.
Figure 1 shows the trial profile. Baseline characteristics at randomisation were much the same between treatment groups, apart from modest differences in race, sex ratio, and concentration of fasting plasma glucose (table 1).
Trial profile
Insulin degludec group A received a starting dose of 10 U (1 U=6 nmol) per day and group B received a starting dose of 10 U (1 U=9 nmol). IDeg 3TW=insulin degludec three times a week. IDeg OD(A)=insulin degludec (group A) once a day. IDeg OD(B)=insulin degludec (group B) once a day. IGlar=insulin glargine once a day. *Seven participants met a combination of criteria. †Withdrawal of consent, withdrawal because of personal reasons, withdrawal due to stress about giving injections on a regular basis, withdrawal for the safety of the participant, withdrawal due to meeting an exclusion criteria, withdrawal due to randomisation in error, or withdrawal because participant was lost to follow-up. ŗWeight gain (5·7 kg) was unacceptable to the participant. §Moderate case of pulmonary tuberculosis.
Mean HbA1C and fasting plasma glucose concentrations were much the same between treatment groups (figure 2). Mean HbA1C reductions from baseline were between -1·3% and -1·5% for all treatment groups (table 2), and reductions of HbA1C did not differ between treatment groups (table 3). At study end, mean fasting plasma glucose concentrations were much the same across treatment groups (table 2), and no differences were noted in the ANOVA analysis (table 3).
After 16 weeks, mean nine-point selfmonitored blood glucose profiles were lower in all treatment groups than they were at baseline, and the overall shape of the mean profiles was nearly the same for the four treatment groups (figure 2).
Insulin doses were adjusted during the trial to achieve specified fasting plasma glucose targets. The weekly starting doses for insulin degludec three times a week and once a day (group B) were higher than they were for the other two groups because of formulation differences (table 2). After 16 weeks, mean weekly insulin dose was very similar for all treatment groups apart from for insulin degludec once a day group B (table 2). In terms of mean daily doses after 16 weeks of treatment, the three times a week dose was 1·14 U/kg per injection (corresponding to 90 U per injection and 0·49 U/kg per day), and the once a day doses were 0·45 U/kg per injection (35 U per injection) for insulin degludec group A, 0·64 U/kg per injection (53 U per injection) for insulin degludec group B, and 0·48 U/kg per injection (38 U per injection) for insulin glargine.
Overall reported rates of hypoglycaemia were low in all treatment groups; 77—92% of participants did not have a hypoglycaemic episode (table 4). Rates of hypoglycaemia did not differ between groups as seen from the 95% CI (table 5, left column). However, the proportion of participants who had hypoglycaemia in insulin degludec group A was lower than was the proportion in the insulin glargine group and the insulin degludec three times a week group (odds ratio; table 5, right column). The rate of confirmed nocturnal hypoglycaemia was low in all treatment groups (table 4). Webappendix pp 2—3 lists hypoglycaemic episodes according to American Diabetes Association definitions.
Bodyweight was stable throughout the trial in every group (table 2). A significant difference in bodyweight was noted between the insulin degludec group B and insulin glargine groups (table 3). No obvious differences were noted between treatments in physical-examination findings, vital signs, standard laboratory analyses (haematology and biochemistry), fundoscopy, or ECGs.
Most (>97%) adverse events were mild or moderate in severity, and there was no apparent treatment-specific pattern. Only two serious adverse events were reported: aggravation of a pretrial coronary heart disease in the three times a week insulin degludec group and worsening of paroxysmal atrial fibrillation in the once a day insulin degludec group B; both events were deemed by the investigator to be unlikely to be related to trial product. Adverse events deemed by the investigators to have possible or probable relation to insulin were reported for six participants in the insulin degludec three times a week group (headache, dermatitis, pruritic rash, diarrhoea, stomach discomfort, and peripheral oedema), two participants in the insulin degludec group A (three events; dizziness, dysgeusia, and palpitations), five participants in the insulin degludec group B (headache, increased blood cholesterol, increased weight, pruritus, and muscle spasms), and two participants in the insulin glargine group (headache and increased blood cholesterol). There were few injection-site reactions. One participant in the insulin degludec group A had two itchy erythematous rashes on one observation day, but both were assessed by an independent dermatologist (Dermatolo-Venerology and Wound Healing Centre, Charlottenlund, Denmark) as being probably non-allergic on the basis of digital photographs and detailed history. One participant in the insulin degludec group B and two participants who were treated with insulin glargine had one non-allergic skin reaction each. One mild event of injection-site haematoma occurred in the once a day insulin degludec group B. No skin reactions occurred in the group injected with insulin degludec three times a week.
Concentrations of antibodies that were specific to insulin degludec and those cross-reacting between insulin degludec and human insulin remained close to zero during the trial (webappendix p 4). Most participants did not develop insulin degludec-specific or cross-reactive antibodies, and only eight participants had cross-reactive antibody concentrations of 10% B/T or more in the insulin degludec group A, three in group B, and two in the three times a week group. There was no apparent association between the development of cross-reacting antibodies and hypoglycaemia, HbA1C, or insulin dose in any of the treatment groups (data not shown).
Overall, quality of life assessments suggested marginal changes in patient-reported outcome scores at the end of the trial (webappendix p 5).
  icon paper stack View Older Articles   Back to Top