Clinical and cost–effectiveness of insulin degludec: from clinical trials to clinical practice

Insulin degludec is a new-generation basal insulin analog with a unique mode of protraction that differs from human insulin through the deletion of residue ThrB30 and the addition of an acyl covalent side-chain (hexadecandioyl), attached to LysB29 via a glutamic acid spacer. Postinjection, phenol in the pharmaceutical formulation disperses; this induces a conformational change in the insulin degludec dihexamers and promotes self-association into long multihexamer filaments in the subcutaneous space. Zinc ions slowly diffuse out of the complex, which allows individual insulin degludec monomers to be released into the blood stream at a slow and constant rate [19,20]. This mechanism generates a flat pharmacokinetic profile with a duration of action exceeding 42 h and low day-to-day and hour-to-hour intrapatient variability in blood glucose-lowering activity [21–23].

The extensive Phase III clinical trial program for insulin degludec included more than 11,000 patients with Type 1 or Type 2 diabetes, with participants from North America, South America, Europe, Africa and Asia. Studies were conducted with a treat-to-target design, as recommended in the European Medicines Agency and US FDA guidance [24,25]. The treat-to-target trial design balances the benefits of glycemic control with the associated side effects of a therapy (e.g., risk of hypoglycemia), so that a risk−benefit assessment can be made. Treat-to-target studies are considered best practice and the most ethical way to assess insulin therapies. In treat-to-target studies, the insulin dose is adjusted for each individual subject with the aim of achieving identical glycemic targets for both study drug and comparator. A result of the treat-to-target design is that HbA_1c will most likely be similar in each treatment group, as the primary aim of the study is to bring all patients to the same glycemic target [26]. The main difference between insulin therapies subject to this design will be seen in terms of safety parameters (e.g., rates of hypoglycemia). The treat-to-target design should result in more balanced outcomes than a trial design focused solely on reducing HbA_1c. In all of the Phase III trials versus insulin glargine, insulin degludec successfully lowered HbA_1c, demonstrating noninferiority [27–35].

A key benefit of insulin degludec that sets it apart from other basal insulins is the reduced restrictiveness of dose timing. Two 26-week, Phase III studies with insulin degludec, which assessed the potential for flexible dosing, have shown that it is possible to adjust the timing of administration without compromising efficacy or increasing the risk of hypoglycemia [32,33]. Although dosed once daily, insulin degludec's ultra-long duration of action allows for flexibility in the timing of administration, occasionally, when administration at the same time of day is not possible, provided there is a minimum of 8 h between injections [36]. For individuals with busy lifestyles or irregular schedules, this convenience could help in promoting greater adherence.

To assess rates of hypoglycemia, a prospectively planned meta-analysis was carried out using data from all seven treat-to-target Phase IIIa trials in which insulin degludec was compared with insulin glargine [27,29,30,32–35]. The use of a standardized definition of hypoglycemia across all trials facilitated this meta-analysis; furthermore, only confirmed episodes of hypoglycemia were included in the rates reported (i.e., those with a plasma glucose [PG] measurement of <3.1 mmol/l, irrespective of symptoms, or if the hypoglycemia was severe – requiring assistance from a third party). Nocturnal hypoglycemia was defined as occurring between 00:00 and 05:59 [37]. The estimated rate ratio (RR) of overall confirmed hypoglycemia was found to be significantly lower in Type 2 diabetes (RR: 0.83 [95% CI: 0.74; 0.94] and the rate of nocturnal confirmed hypoglycemia was numerically lower in Type 1 diabetes and significantly lower in Type 2 diabetes (RR: 0.68 [95% CI: 0.57; 0.82]) for insulin degludec compared with insulin glargine [37]. A statistically significant 86% reduction in the rate of severe hypoglycemia was also seen in previously insulin-naive patients with Type 2 diabetes receiving insulin degludec (RR 0.14 [95% CI: 0.03; 0.70] 0.003 vs 0.02 events/patient-year of exposure) [37]. A separate meta-analysis of Phase III trial data, focusing on hypoglycemia rates in elderly patients with Type 1 and Type 2 diabetes, also reported significant reductions in overall and nocturnal hypoglycemia with insulin degludec compared with insulin glargine. Rates of severe hypoglycemia were similar and too low for statistical analysis to be performed [38]. The clinical benefits of insulin degludec were maintained across ethnic groups, with a reduction in the risk of overall confirmed hypoglycemia also being observed in the BEGIN Once Asia study after stable glycemic control and insulin dose had been achieved [34].

Several post hoc analyses have since been conducted to determine whether alternative definitions of hypoglycemia and nocturnal hypoglycemia impact upon the risk reduction reported in the meta-analysis by Ratner et al. [37]. These analyses used the revised criteria of ‘only confirmed episodes of hypoglycemia with symptoms’; the ADA definition of hypoglycemia (symptomatic and PG ≤70 mg/dl); and an extension of the ‘nocturnal period’ to 8 h covering two different timescales (21:59–05:59 and 00:01–07:59), using the original definition (PG ≤56 mg/dl or severe hypoglycemia requiring assistance). With the exception of extending the nocturnal period to 07:59 in insulin-naive patients with Type 2 diabetes, increasing the sensitivity and specificity of the hypoglycemia definition did not markedly change the rate ratio or confidence intervals for insulin degludec compared with insulin glargine in different patient groups. These post hoc analyses suggest that the reduced rate ratio of nocturnal confirmed hypoglycemia with insulin degludec versus insulin glargine in patients with Type 2 diabetes will be seen in clinical practice [39]. Further post hoc meta-analyses of Phase III trial data have shown that, whilst HbA_1c was similar between treatment arms, patients with Type 1 diabetes on basal–bolus therapy and insulin-naive patients with Type 2 diabetes had significantly lower total daily insulin doses (-12%, p < 0.0001 and -10%, p < 0.0004, respectively) and significantly lower rates of nocturnal nonsevere hypoglycemia (RR: 0.83 [95% CI: 0.69; 0.99] and 0.64 [95% CI: 0.47; 0.86], respectively) when treated with insulin degludec compared with insulin glargine. This was despite significantly lower fasting PG concentrations at end-of-trial in the patients receiving insulin degludec compared with patients receiving insulin glargine (estimated treatment difference: Type 1 basal–bolus -0.61 [95% CI: -1.13; -0.1] and Type 2 insulin-naive -0.34 [95% CI: -0.54; -0.15]) [40].

Besides measuring hard clinical end points, the Phase III studies with insulin degludec also recorded patient-reported outcomes with the aim of capturing changes in quality of life. Evaluated using the 0–100 ranked Short-Form 36 (SF-36) version 2.0 health survey, a statistically significant improvement in vitality (+0.81 points [95% CI: 0.01; 1.59]) and physical health (+0.66 points [95% CI: 0.04; 1.28]) status was observed for insulin degludec compared with insulin glargine, in patients with Type 2 diabetes starting basal insulin. The improvement in physical health status was largely driven by a lower score in the domain for bodily pain (+1.10 points [95% CI: 0.22; 1.98]) among patients treated with insulin degludec [41]. A separate meta-analysis of data from six Phase III trials mapped SF-36 scores onto the EuroQol 5D health utility scale, which has a range of -0.59 (a state worse than death) to 1.00 (perfect health). Using this approach it was shown that insulin degludec provides a modest, but statistically significant, improvement in health utility compared with insulin glargine (+0.005 points [95% CI: 0.0006; 0.009], p < 0.024) [42]. Together, these analyses suggest that insulin degludec improves patients’ quality of life more than insulin glargine, even within the confines of the clinical trial.

Whilst the Phase III clinical trial program has successfully demonstrated the benefits of insulin degludec, limitations such as the treat-to-target study design and the exclusion criteria (e.g., patients who experience recurrent hypoglycemia) limit the evaluation of its real-world potential. One of the first real-world studies to report the benefits of insulin degludec in the clinic has recently published its findings. A maximum 37-week follow-up of 51 patients in routine practice in the UK, who were suffering from recurrent hypoglycemia on insulin glargine or insulin detemir, found that switching to insulin degludec resulted in clinically significant reductions in reported hypoglycemia, lower HbA_1c and improved treatment satisfaction. HbA_1c was reduced by an average of 0.5 ± 0.3% and 0.7 ± 0.3% for patients with Type 1 and Type 2 diabetes, respectively. The reduction in frequency of hypoglycemia was >90% [43].

The key clinical benefits of insulin degludec – the potential for flexible dosing and a reduced risk of hypoglycemia – have been clearly demonstrated in RCTs and real-world studies. Not only will these attributes help patients by providing greater freedom from same-time-daily dosing and reducing their exposure to the damaging mental and physical effects of hypoglycemia, they may also improve patients’ adherence. From the clinicians’ perspective, insulin degludec shows great potential, particularly in the treatment of patients experiencing recurrent hypoglycemia, and could assist in bringing patients to target HbA_1c levels without the fear of an excessive increase in the risk of hypoglycemia.

The cost–effectiveness of an intervention is driven by two factors: the improvement in health status/quality of life and the cost of treatment. Many healthcare payers use the quality-adjusted life-year (QALY) to assess whether a drug is cost-effective, applying a financial threshold at which cost–effectiveness is accepted or rejected (i.e., the National Institute for Health and Care Excellence sets this at £20,000–£30,000/year, depending upon the disease). QALYs can also be used when comparing the cost–effectiveness of two therapies, whereby the cost of treatment A minus the cost of treatment B is divided by the effectiveness of treatment A minus the effectiveness of treatment B. This is termed the incremental cost–effectiveness ratio (ICER). Typically, short-term cost–effectiveness modeling is used in conjunction with clinical trials. This is because treat-to-target trials are not optimized to show differences in long-term risk markers, for example HbA_1c. Conversely, real-world studies are better suited to long-term cost–effectiveness analyses.

Health economic evaluations assessing the cost–effectiveness of the first basal insulin analogs to be developed have shown them to be cost-effective, compared with neutral protamine Hagedorn insulin, in both the short and long term [44–46]. These studies have shown that it is possible to use the baseline demographics, changes in treatment and physiological parameters recorded in the trial setting to project total direct costs (the expense of providing drugs and treating potential complications) of an intervention over a patient's lifetime.

Data from a meta-analysis of Phase III clinical trials with insulin degludec were used to populate a short-term health economic model, and it was concluded that, for patients with Type 2 diabetes who are considered appropriate for treatment with a basal insulin analog, insulin degludec is a cost-effective treatment option compared with insulin glargine. This is a result of the lower risk of hypoglycemia, reduced insulin dose requirement and improved quality of life observed with insulin degludec. A base-case analysis, using hypoglycemia rates reported by the UK Hypoglycaemia Study Group [47], yielded estimated ICERs of £15,795 per QALY and £13,078 per QALY for insulin degludec and insulin glargine, respectively. In a subgroup of patients who experienced ≥1 hypoglycemic event per year, insulin degludec was even more cost-effective versus insulin glargine, with estimated ICERs of £4887 and £2625 per QALY, respectively [48]. Another short-term (1-year time horizon) study was carried out on a subpopulation of Swedish patients from the insulin degludec Phase III program. A cost–utility model determined that insulin degludec is associated with a gain in QALYs in patients with Type 1 diabetes (0.31 vs 0.26  QALYs), Type 2 diabetes treated on basal insulin-supported oral therapy (0.76 vs 0.69 QALYs) and Type 2 diabetes treated with basal–bolus therapy (0.56 vs 0.47 QALYs), versus insulin glargine. The study demonstrated that, from a societal perspective, based on the reduced incidence of hypoglycemia and possibility for flexibility around timing of dose administration, use of insulin degludec is likely to be cost-effective in Sweden compared with insulin glargine for the patient/treatment groups assessed [49].

Because patients experiencing recurrent hypoglycemia were excluded, by design, from the insulin degludec clinical trial program, understanding how the lower rates of hypoglycemia reported in those studies will translate into real-world clinical practice has been difficult. Recently, the cost–effectiveness of insulin degludec has been analyzed using the IMS CORE diabetes model, with real-world input data – unconstrained by treat-to-target trial design – from 19 patients with Type 1 diabetes who were experiencing recurrent hypoglycemia. The model showed that over a lifetime, treatment with insulin degludec is cheaper and more effective than treatment with insulin glargine or insulin detemir, with a cost per QALY of £13,220 (below the commonly used cost–effectiveness threshold of £20,000/QALY) [50]. This provides further evidence that the benefits of insulin degludec observed in RCTs are maintained in the real world.

When profiling new therapeutic entities for their clinical potential and cost–effectiveness, clinical trials are unparalleled in their evaluation of specific end points among a highly selected population under tightly controlled conditions. Yet, due to limitations in RCT design, assessing external validity is difficult; this is why real-world data are important in assisting regulatory agencies and healthcare professionals in their decision-making process.

Evidence from the insulin degludec clinical trial program and subsequent real-world studies demonstrate that the ultra-long duration of action and low variability in glucose-lowering activity are translated into real clinical benefits, in particular a reduced risk of hypoglycemic events. Furthermore, recently published studies have shown that insulin degludec is likely to be cost-effective compared with other basal insulin analogs, particularly for those patients who are struggling with recurrent hypoglycemia. Time trade-off surveys have shown that flexibility in basal insulin dose timing plays an important role in improving patients’ quality of life [51]. It is likely that those patients who choose to exercise this freedom in a real-world setting will also report improved quality of life. In summary, currently available data both from the clinical trial program and real-world studies show great promise for insulin degludec. However, further studies need to be conducted to evaluate the long-term benefits, and to assess whether there are other groups of patients for whom it would be of benefit, and to evaluate prescribing patterns.

Projections show that rising obesity rates will lead to a parallel in increase in Type 2 diabetes over the coming decades, both in developed and developing countries. Coupled with a growing elderly population, pressure on healthcare budgets will continue to grow and reimbursement decisions will face greater scrutiny. Clinical trials with insulin degludec have shown that it provides a significantly lower risk of hypoglycemia compared with other insulin analogs, whilst evidence from real-world studies demonstrate that insulin degludec is cost-effective, particularly in patients who experience recurrent hypoglycemia. Considering this evidence, we believe that insulin degludec may therefore help in alleviating the burden of hypoglycemia on healthcare budgets in coming years.

M Evans has received honoraria as an advisory panel member and speaker for Novo Nordisk, Sanofi Aventis, Novartis and Merck Sharp & Dohme. P McEwan has received research funding from Novo Nordisk, Sanofi, Bristol-Myers Squibb, AstraZeneca, Eli Lilly & Co., GlaxoSmithKline and Takeda. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Writing assistance was utilized in the production of this manuscript. The authors are grateful to P Tisdale and D Renshaw of Watermeadow Medical, UK, for providing medical writing and editorial/submission support, which was sponsored by Novo Nordisk A/S.

This work is licensed under the Creative Commons Attribution-NonCommercial 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/3.0/