Although insulin introduced nearly a century ago has saved the lives of countless people with diabetes, improvements of this therapy continue as significant unmet needs remain (1–4). Even persons with type 2 diabetes who, compared with those with type 1 diabetes, exhibit a much more stable, yet disordered, metabolic milieu only achieve an intended HbA1c target of 7% or lower without hypoglycemic events in a minority of cases (1,3,4). Hypoglycemia and weight gain represent the most important obstacles to achieving the desired glycemic control with insulin therapy, despite major advances in education and empowerment of people with diabetes, the introduction of self-monitoring of blood glucose, improvement in the technical application of insulin, and the availability of a wide array of insulin preparations including insulin analogs (1–4). Intermediate and long-acting basal insulins represent the cornerstone of any insulin therapy regimen, and human NPH insulin, insulin glargine, and insulin detemir are currently the most widely used members of this class of insulin preparations. Discernable progress has been observed with improved day-to-day variability and a lowering in the risk of hypoglycemia with the introduction of the soluble long-acting preparations insulin glargine and insulin detemir compared with NPH insulin, which requires appropriate resuspension prior to administration to reduce variability (5). Against the background of the aforementioned unmet needs, however, a new generation of long-acting basal insulins has emerged aiming at new standards in terms of pharmacokinetics, efficacy, and safety (Fig. 1).
Modern drug development occurs in an era of increasingly strict regulations by global institutions such as the U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMA) that rightly demand enhanced safety checks of new chemical entities for clinical use such as the new long-acting insulin analogs (6–8). Expensive clinical safety studies are, therefore, required, and cost has thereby become a major issue. This has direct impact on the prescribing of insulins in the emerging economies around the world and even in some developed countries where the cost of NPH insulin has increased. Furthermore, sales of one of the new-generation long-acting insulins, i.e., insulin degludec, have recently been halted by the manufacturer in a country like Germany, as no agreement could be reached with the German authorities over appropriate pricing. Hence, this article seeks to evaluate whether the clinical benefits of the new long-acting insulins indeed justify the cost.
Brief Description of Previous and New Long-Acting Basal Insulins
Figure 2 outlines the chemical structure and modifications derived from the human insulin molecule of insulins glargine (Fig. 2A) and detemir (Fig. 2B) as well as of the three new long-acting basal insulins discussed in this article (Fig. 2C, D, and E). In addition, Tables 1 and 2 provide an overview of clinically important time-action characteristics of current standard long-acting basal insulins and insulin analogs and this new generation of even longer-acting basal insulin analogs, respectively (3). These, apostrophized by some also as “ultra”-long-acting insulin analogs, clearly possess a duration of action far beyond 24 h, with a much reduced peak action, whereas current standard long-acting insulins act for up to 24 h and slightly beyond, with higher doses extending to a longer duration of action. However, the new longer-acting insulins also need to be dosed daily and approximately at the same time, albeit with some flexibility (see below), as the effect does wane beyond the 24-h period.
Chemical structure as derived from the human insulin molecule of long-acting basal insulin analogs.
Chemical structure as derived from the human insulin molecule of long-acting basal insulin analogs.
Pharmacological profile of current standard basal insulins
| Basal insulin classification | Insulin preparation | Onset (h) | Peak (h) | Duration (h) | Variability (CV%) | Timing of administration |
|---|---|---|---|---|---|---|
| Intermediate acting | NPH | 1–3 | 4–6 | 12–16 | 68 (refs. 1,4) | Usually twice daily |
| Long acting | Glargine | 0.5–2 | Flat, some peak 4–12 | ∼24 | 32–82 (refs. 1,38,56) | Once daily, same time of day |
| Detemir | 0.5–2 | Flat, some peak 7–14 | ∼20 | 27 (refs. 1,4) | Once or (often) twice daily |
| Basal insulin classification | Insulin preparation | Onset (h) | Peak (h) | Duration (h) | Variability (CV%) | Timing of administration |
|---|---|---|---|---|---|---|
| Intermediate acting | NPH | 1–3 | 4–6 | 12–16 | 68 (refs. 1,4) | Usually twice daily |
| Long acting | Glargine | 0.5–2 | Flat, some peak 4–12 | ∼24 | 32–82 (refs. 1,38,56) | Once daily, same time of day |
| Detemir | 0.5–2 | Flat, some peak 7–14 | ∼20 | 27 (refs. 1,4) | Once or (often) twice daily |
CV, coefficient of variation.
Pharmacological profile of new ultra-long-acting basal insulins (>24 h)
| Insulin preparation | Onset (h) | Peak (h) | Duration (h) | Within-subject variability (CV%) | Timing of administration | T1/2 (h) | Steady state (days) |
|---|---|---|---|---|---|---|---|
| Degludec | ND | Flat, no peak | >24 | 20 (refs. 38,39) | Once daily, any time of day (?) | 25* | 4.2 |
| Glargine U300 | ND | Flat, no peak | >24 | 17–35 (in T1D) (ref. 42) | Once daily ±3 h | 18 | 3–4 |
| PegLispro | ND | Flat, no peak | >24 | <18 (ref. 56) | Once daily | 39* | 6.6 |
| Insulin preparation | Onset (h) | Peak (h) | Duration (h) | Within-subject variability (CV%) | Timing of administration | T1/2 (h) | Steady state (days) |
|---|---|---|---|---|---|---|---|
| Degludec | ND | Flat, no peak | >24 | 20 (refs. 38,39) | Once daily, any time of day (?) | 25* | 4.2 |
| Glargine U300 | ND | Flat, no peak | >24 | 17–35 (in T1D) (ref. 42) | Once daily ±3 h | 18 | 3–4 |
| PegLispro | ND | Flat, no peak | >24 | <18 (ref. 56) | Once daily | 39* | 6.6 |
CV, coefficient of variation; ND, not determined due to the methodologic approach used; T1/2, half-life time; T1D, type 1 diabetes.
*Methodologic limitation, since the insulin assay detects only total, not free, insulin concentration.
In comparison, the duration of action of NPH insulin is considerably shorter, being up to 16 h and dose dependent, with a clear peak action after 4–6 h (1,3,4,9–14). The latter property may be responsible for nocturnal hypoglycemia sometime between 2:00 and 4:00 a.m., after the administration of NPH insulin at bedtime, or late morning hypoglycemia, when NPH insulin is given before breakfast and its peak action not sufficiently covered by an appropriate snack. The long-acting insulin detemir has a flatter peak action profile than NPH at low doses, but at higher doses a peak action is evident 7–14 h after administration (1,3,4,9–13,15). The long-acting insulin glargine has an even smoother peak action profile than insulin detemir but has a discernable peak action 3–4 h after administration while possessing a duration of action of 24 h or more, depending on the dose (1,3,4,9–13,15).
In a comparison of the three options of standard basal insulins in a “real-world” setting, some practically relevant differences become evident, as can be seen in a huge database of 51,964 individuals with type 1 or type 2 diabetes exclusively on insulin therapy from 336 centers in Germany and Austria. The injection frequencies and doses of basal and total insulin per day were lowest with the use of insulin glargine, which typically enables a once-a-day dosing regimen, compared with NPH insulin or insulin detemir, yet at similar glycemic control and rates of severe hypoglycemia (16). A meta-analysis looking into data from 9,548 participants with type 2 diabetes failing on oral agents from 22 randomized controlled trials (RCTs) assessed the comparative efficacy of adding the established long-acting insulin analogs to existing oral therapies, although without differentiating between insulin detemir and glargine, compared with adding NPH insulin or premixed insulins containing NPH insulin or other options like glucagon-like peptide 1 receptor agonists and rapid-acting insulin analogs (9). Implementing therapy with long-acting insulin analogs such as glargine or detemir in this context achieved glycemic control similar to that with NPH insulin or short-acting insulin analogs or glucagon-like peptide 1 receptor agonists, although slightly inferior to that with premixed insulins, but importantly with the least number of side effects, especially hypoglycemia.
In summary of these and findings from other comparative studies, the following practical, evidence-based notions become apparent (1,3,4,9–14,16–18):
Standard long-acting insulin analogs (insulin glargine and insulin detemir) are not superior to NPH insulin in efficacy terms as determined by the number of participants reaching HbA1c targets.
Compared with use of NPH insulin, the use of the standard long-acting insulin analogs with relatively flat action profiles is associated with up to a 50% reduced risk of nocturnal hypoglycemia.
All insulin analogs, both short- and long-acting insulin analogs, contribute to a reduced rate of overall hypoglycemia and less weight gain compared with therapies based on regular human insulin and NPH insulin.
In contrast to insulin detemir, insulin glargine typically lowers blood glucose for 24 h, thus allowing for a once-a-day basal insulin regimen in most patients. Insulin glargine is currently regarded as the “gold standard” basal insulin (19), i.e., excluding regimes using insulin infusion devices.
Hence, all representatives of the new generation of ultra-long-acting insulins, i.e., insulins degludec, pegylated lispro (PegLispro), and glargine U300, have mostly been evaluated in head-to-head comparisons in RCTs against insulin glargine as the reference standard (14,15,20–26).
Whether the benefits of long-acting insulin analogs translate long-term into a reduction in micro/macrovascular diabetes complications is a largely unaddressed issue, as only very few RCTs have been of long enough duration and include sufficient numbers of participants for a meaningful evaluation of this important aspect. One such trial is the Outcome Reduction With Initial Glargine Intervention (ORIGIN) trial, which looked into cardiovascular (CV) outcomes over a 6-year period in 12,612 persons (predominantly persons with early type 2 diabetes and some [12%] with impaired glucose tolerance). Glargine therapy was similar to more conventional drug therapy, mainly based on metformin and/or sulfonylureas, in terms of macrovascular and microvascular outcomes (27). The equivalence in outcomes also included prespecified, detailed cancer outcomes, thus dismissing earlier fears over a higher cancer risk related to the use of insulin glargine (27). Furthermore, some suggestive evidence in favor of insulin glargine was generated from the ORIGIN-GRACE substudy in reducing the progression of carotid-intima thickening, i.e., a surrogate of CV risk (28), and in a national administrative claims database for acute myocardial infarction (28).
The mega-trials UK Prospective Diabetes Study (UKPDS), Action in Diabetes and Vascular Disease: Preterax and Diamicron MR Controlled Evaluation (ADVANCE), and Veterans Affairs Diabetes Trial (VADT), although not specifically studying a particular long-acting insulin but, rather, seeking to intensify glycemic control by the use of multiple drugs including insulin in a randomized, controlled fashion, have recently provided evidence from open follow-up observations after the initial RCT that the earlier intensive glycemic control was ultimately effective in reducing long-term CV events by some 17% and, in the UKPDS, also in reducing overall mortality to a similar extent (29–31). A meta-analysis of these and other studies, looking selectively and post hoc at nonfatal myocardial infarction during the RCT phase of these trials, also showed a 15–17% reduction in the intensively treated subjects (32,33). As to subjects with type 1 diabetes, the long-term follow-up of the Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) provides comparable evidence (34,35). So, insulin therapy clearly is a major contributor to the successful reduction of long-term diabetes complications in both type 1 and type 2 diabetes.
The new generation of long-acting insulin analogs seeks to further improve on many features of an ideal basal insulin, which includes not only a duration of action beyond 24 h without possessing an action peak but also a lower intraindividual and interindividual day-to-day variability of action (the assessment of which, however, has sometimes yielded rather variable results [Tables 1 and 2]), along with a minimal risk for inducing hypoglycemia and incurring weight gain. Clearly, such attributes should result in an overall improvement in quality of life and increased patient compliance with treatment requirements.
Insulin degludec is a second-generation acylated insulin following insulin detemir (Fig. 2), again a soluble insulin formulation, which upon injection into the subcutaneous tissue forms multihexamer chains. The latter serves as a depot prior to degradation into dimers and monomers, which are then absorbed into the circulation (36). Degludec has a low affinity for the human IGF-1 receptor, comparable with human insulin, with a low mitogenic-metabolic potency ratio (37). It has a half-life of 25 h, which seems to be almost twice as long as glargine, with a lower degree of variability in its glucose-lowering effect over 24 h when at steady state (38,39) (although those measurements have some methodologic limitations). An extensive clinical development program with degludec in both type 1 and type 2 diabetes has demonstrated noninferiority to glargine in terms of efficacy with a reduced risk of nocturnal hypoglycemia (1,3,4,10,13,14,23,24,38,40). To date, insulin degludec has been approved for use in Europe (including its use in children), Asia, and Latin America and very recently also in the U.S. (September 2015), based on additional interim data from a CV outcome study (A Trial Comparing Cardiovascular Safety of Insulin Degludec Versus Insulin Glargine in Subjects With Type 2 Diabetes at High Risk of Cardiovascular Events [DEVOTE], clinical trial reg. no. NCT01959529, clinicaltrials.gov). This requirement was deemed necessary based on the original phase 3 trials, which suggested the presence of an enhanced CV risk with insulin degludec in comparison with insulin glargine, although this interpretation was not universal. The findings from the DEVOTE study in 7,500 high-risk individuals with type 2 diabetes will be invaluable owing to its prospective, double-blind design.
Glargine U300 is a new higher-strength formulation of insulin glargine at 300 IU/mL rather than the usual glargine at 100 IU/mL, which alters its pharmacokinetic and pharmacodynamic properties (1,3,4,22,25,26,41,42). Glargine U300 forms a more compact subcutaneous depot with its surface area reduced by half and its volume by two-thirds (Fig. 2), resulting in a retarded and prolonged dissolution from the subcutaneous tissue compared with generally available standard glargine (U100) (Tables 1 and 2). Also, its metabolic degradation to the metabolites M1 (GlyA21 insulin) and M2 (GlyA21, des-ThrB30) is similar to that of insulin glargine U100 (42). These metabolites show binding characteristics to the insulin receptor and IGF-1 receptor comparable with human insulin. A number of key phase 3 randomized parallel-group studies (EDITION), predominantly in persons with type 2 diabetes (22,25,26), demonstrate noninferiority to insulin glargine U100, although at a slightly higher dose requirement of ∼10–18%, but with a 25% reduced risk of nocturnal hypoglycemia over a 6-month period, already evident during the titration period (43). In addition, there was negligible weight gain with the new high-strength glargine regardless of age, BMI, and diabetes duration (44). Glargine U300 has recently been approved in the U.S. and several countries in Europe.
PegLispro (LY2605541), at this moment an investigational insulin only, is a long-acting insulin that comprises insulin lispro covalently bound, via a urethane bond at lysine B28, to a 20-kDa polyethylene glycol (PEG) chain (Fig. 2), thereby increasing the hydrodynamic size of the insulin-PEG complex (1,3,4,15,20,21,45–48). PegLispro has an even lower binding affinity for both the insulin receptor and IGF-1 receptor than lispro and a lower mitogenic potency than human insulin (45,47,48). It possesses a prolonged duration of action exceeding 36 h as a result of delayed absorption and reduced clearance, with a low intrasubject variability in glucose lowering and a lower and quite different tissue distribution compared with the other insulin preparations (45,47,48). Both animal and human studies suggested that PegLispro may have a preferential hepato-specific action on glucose metabolism compared with insulin glargine (49). Early phase 3 clinical studies (IMAGINE) include eight pivotal studies, three of which are double-blind, and most involve glargine (U100) as the comparator. These have demonstrated that PegLispro was consistently superior under these conditions to glargine in terms of lowering HbA1c and in causing less nocturnal hypoglycemia with a reduction of ∼40% (20,21,50–52). However, an increase in liver enzyme levels (alanine aminotransferase) with an increase in liver fat content has been noted, although no acute, severe, hepatocellular drug-induced liver injury has been observed to date with PegLispro in studies up to 78 weeks (20,21,51,53).
Arguments in Favor of the New Insulins
Hypoglycemia is a most feared and costly complication of insulin therapy and is well recognized as a major barrier to achieving the required glycemic control in the majority of insulin-requiring persons with diabetes (1). Insulin degludec has been clearly demonstrated to be equivalent to insulin glargine in lowering HbA1c in key RCTs in both type 1 and type 2 diabetes (14,23,24). However, the risk of hypoglycemia, especially nocturnal hypoglycemia, was further reduced by ∼25% during the maintenance period of the trials despite somewhat lower fasting plasma glucose concentrations (40). In addition, insulin degludec allows a more flexible approach to the timing of insulin administration to better accommodate the recipients’ ever-changing daily life pattern (1,3,4,10,13,50,54). Such improved convenience seems to result in better compliance with therapy and an enhancement in the patients’ quality of life (50,54). Finally, the use of insulin degludec has been found to be associated with a 10% lower dose requirement compared with both insulin detemir and insulin glargine in a meta-analysis of available RCTs (55). This seems particularly noteworthy, as the use of insulin glargine yields the lowest insulin doses among the current standard long-acting insulins (16).
Insulin glargine U300 also provides a flatter and more reproducible pharmacokinetic profile resulting in a more even daylong blood glucose–lowering action than its predecessor (insulin glargine U100) (1,3,4,22,25,26,41) (Table 2). The intraindividual variability measures were low and ranged from 17 to 35% (42). Head-to head comparisons of glargine U300 and standard glargine U100 in RCTs showed identical efficacy in terms of lowering HbA1c, although at a slightly higher dose requirement of ∼10–18%. An ∼25% decrease of nighttime confirmed or severe hypoglycemia, however, and only minimal weight gain were observed with the use of insulin glargine U300 in studies looking at subjects with type 2 diabetes (22,25,26,43). Due to the longer duration of action, insulin glargine U300 has been shown to provide some increased flexibility in the timing of the insulin injection of approximately ±3 h (58,59). Quality-of-life assessments have shown a slight improvement with the use of glargine U300 over insulin glargine U100 (1,3,4,22,25,26). In addition, the new high-strength glargine seems to have the substantial weight of indirect safety evidence, both CV and carcinogenesis, emanating from the ORIGIN trial (27).
The experimental PegLispro long-acting insulin analog also yields a relatively flat and peakless pharmacokinetic profile along with a more reproducible glycemic action profile compared with insulin glargine U100 and a much extended duration of action, exceeding 36 h, reaching steady state within 7–10 days (48) (Table 2). Based primarily on animal studies, PegLispro is catabolized to smaller peptides and/or amino acids and to lysine-PEG, the latter being eliminated via biliary and renal routes (1,3,4,45). There is also a lesser variability in glucose lowering with PegLispro compared with insulin glargine (56) accompanied by less weight gain and fewer nocturnal hypoglycemic episodes in early phase 2 studies with PegLispro in both type 1 (51) and type 2 diabetes (21,50,56). Its preferential hepatic (inhibition of glucose production) versus peripheral (glucose disposal) activity makes it perhaps more similar to endogenous insulin when compared with other exogenously administered insulins that possess a predominant effect on peripheral glucose disposal (1,3,4,15,20,21,49). In a series of phase 3 studies (IMAGINE), it shows a consistent, although marginal, but statistically greater HbA1c-lowering capacity (0.2–0.3% HbA1c) than glargine U100 in persons with either type 1 or type 2 diabetes on intensified insulin therapy or persons with type 2 diabetes on basal insulin in addition to oral therapy (20,51,52). A significant lower rate of nocturnal hypoglycemia in the range of 36–45% has been observed compared with that in those taking insulin glargine U100 (1,3,4,15,20,21,50–52). There was also some weight loss associated with the use of PegLispro in contrast to some weight gain with insulin glargine, amounting to an absolute difference of ∼2 kg (20,45,50,51), but warranting a closer look into whether the efficacy of this new insulin at peripheral tissues is adequate.
Arguments Challenging the Benefit of the New Long-Acting Insulins
Whereas the initial studies with insulin degludec seemed to demonstrate a reduction in episodes of hypoglycemia, especially nocturnal hypoglycemia, by ∼25% compared with insulin glargine, the FDA in its assessment highlighted the fact that this benefit was not a consistent feature seen across the different types of diabetes or definitions of hypoglycemia, in particular the more objective definitions of hypoglycemia (14,23,24,40,54,55,57). Also, interpreting the relative safety of these two basal insulins during a set period overnight (0000–0559 h) is difficult owing to the differences in their pharmacokinetic profiles and the assumption about the timing of administration of the main comparator insulin glargine, compounded by the inadequacy of data capture during this nighttime period. Extending the overnight period by 2 h negated the difference in the nocturnal hypoglycemic event rate between insulin degludec and glargine for both individuals with type 1 diabetes and with type 2 diabetes. The perceived safety benefit did not translate into an efficacy advantage either.
The FDA has been able to approve the drug for the U.S. only recently (September 2015), based on additional interim data obtained from the FDA-mandated specific CV outcome trial (DEVOTE), in line with the guidance put forward by this agency in 2008 for approval of new blood glucose medications (5). Approval requires a hazard ratio (HR) with an upper bound of the 95% CI <1.3 for major adverse cardiovascular events (MACE) (composite of CV death, nonfatal myocardial infarction, and nonfatal stroke) from RCTs during phase 2/3 of drug development, whereas >1.8 necessitates a preapproval CV outcome trial and a value in between requires a postapproval CV trial. At the time of the first FDA assessment in 2012, there were 132 case subjects with CV events available (MACE+, i.e., plus unstable angina) in >7,716 patient-years of exposure with an HR of 1.30 (95% CI 0.88–1.93), thus exceeding the 1.8 limit (4). Additional analysis based on MACE, i.e., excluding unstable angina, involved 91 case subjects with CV events resulting in a higher risk with an HR of 1.67 (1.01–2.75). As a consequence, the above-mentioned specific CV outcome trial was mandated by the FDA, as opposed to a less strict stance of the EMA, Japanese, and other authorities that have approved the drug. DEVOTE was commenced in November 2013, designed as a double-blind study comparing insulin degludec with insulin glargine involving ∼7,500 subjects with type 2 diabetes at high CV risk (6–8). Obviously, the FDA set safety margins for marketing approval with an upper bound of the 95% CI <1.8 have now been met, but it remains to be seen whether the full outcome results of the ongoing trial, once available, will ultimately demonstrate CV safety as required for permanent approval. Despite these uncertainties and its somewhat controversial benefits in reducing nocturnal hypoglycemia, the cost associated with the use of insulin degludec in countries where it is or, rather, has been available is substantially higher than with any other basal insulin, e.g., some 50–70% higher than insulin glargine or detemir and several-fold higher than NPH insulin in countries like Germany, Denmark, or the U.K. In fact, the manufacturer has decided to stop marketing the drug in Germany, since no agreement could be reached with the authorities over the appropriate pricing. Also, in other countries, the benefit-to-cost ratio has been challenged, and discussions on this issue are bound to continue, especially since insulin glargine U300, with pharmacokinetic/pharmakodynamic properties not too dissimilar to those of insulin degludec, has become available at a cost comparable with that of previous insulin glargine U100.
The new high-strength long-acting formulation of insulin glargine U300 inherits the indirect proof of long-term CV and cancer safety based on the ORIGIN trial involving insulin glargine U100 strength conducted over a period of ∼6 years (27). In RCTs (EDITION), glargine U300 was seen to be noninferior, though not superior, to standard glargine U100 (22,25,26,43,58). There was, however, a reduction of ∼25% in confirmed nocturnal hypoglycemia (≤70 mg/mL) or severe hypoglycemia over a 6-month study period with the difference already evident during a titration period and continuing thereafter (22,25,26,43). With the period of overnight observation extended from midnight to 8:00 a.m., the difference was 16%, confirming a clinically relevant benefit of glargine U300 compared with glargine U100 (59). Conversely, the reduction of overall hypoglycemia was much less impressive, at ∼10% at any time of the day, pointing to some potential problems during daytime. Indeed, in a meta-analysis of the phase 3 clinical trials on glargine U300, the highest prevalence of hypoglycemia was seen between 6:00 and 10:00 a.m. when insulin glargine was dosed at bedtime (43,59). As for insulin degludec, the lower risk of hypoglycemia did not translate to improved glycemic control despite the 9–18% higher daily dose required in the case of insulin glargine U300, which was accompanied by a minimal increase in weight (43). Both the FDA and EMA have approved insulin glargine U300 for clinical use in 2015. Per-unit cost for glargine U100 and glargine U300 should be comparable and indeed are in countries like the U.S. or U.K. In view of the higher daily dose requirement in the clinical trials, one might, however, argue that the price should be discounted by 10–20% for glargine U300 to better reflect equivalency.
Despite extensive basic and early clinical research leading to a portfolio of well-designed double-blind studies, insulin PegLispro is still at a relatively early phase of development, and the dossier for submission to the regulators seems to be delayed. PegLispro achieves a statistically significantly greater improvement in glycemic control than insulin glargine, and the risk of nocturnal hypoglycemia is reduced by nearly half compared with insulin glargine (1,3,4,15,20,21,43,51). However, daytime hypoglycemic events were significantly increased, possibly due to its hepato-preferential action resulting in a greater inhibition of hepatic glucose output postprandially (20,49). It remains to be seen whether a reduction in the dose of short-acting insulin or other hypoglycemic agents may abolish this untoward daytime hypoglycemia risk. This adaptation may, however, result in a loss of superiority in terms of efficacy compared with standard insulin glargine and perhaps unmask an insufficient efficacy at peripheral muscle and adipose tissues, as might be evidenced by increased lipolysis and some loss of body weight (19,50,51,57). This aspect clearly needs more evaluation, especially since 1 unit of insulin PegLispro contains 9 nmol insulin in contrast to 6 nmol in insulins degludec or glargine U300. The observed increase in triglyceride concentrations needs to be explained and possible consequences need to be explored as well as the very consistently observed increase of liver transaminases (still remaining within the normal range) (20,21,53,60). The observed increase in liver fat by MRI also requires a sound assessment before the overall benefit-to-risk ratio of this new ultra-long-acting insulin can be determined (53).
Consensus and Conclusions
Hypoglycemia remains a major obstacle in our ability to achieve the required glycemic target in the majority of our patients. The introduction of insulin glargine and insulin detemir, both soluble insulin preparations with a more protracted action up to 24 h, smoother time-action profile, and reduced peak action, has resulted in notable improvement and a significant reduction in the risk of hypoglycemia compared with NPH insulin, especially during nighttime.
The recently available new long-acting basal insulin analogs insulin degludec, the more concentrated insulin glargine U300, and PegLispro (the latter an investigational drug only) all demonstrate superior pharmacokinetic and pharmacodynamic properties with an even more protracted action, i.e., >24 h, with a much lower peak action profile compared with currently available basal insulin preparations.
The main clinical advantage of these new long-acting basal insulins is the substantial reduction in nocturnal hypoglycemia, although this is somewhat controversial in the case of insulin degludec.
Other benefits include flexibility of timing of insulin administration, the lack of weight gain or even weight loss, and some improvement in quality-of-life aspects.
For the time being, the use of the longer-acting insulin glargine U300 seems to present the least issues; it appears to offer advantages comparable with those of the other new long-acting basal insulins, yet at a price similar to that for insulin glargine U100. The higher daily dose requirement seen in the clinical drug development program might even warrant a 10–20% discount.
Based on additional interim safety data from an ongoing mandated CV outcome trial (DEVOTE), insulin degludec has recently been approved for marketing also in the U.S., as in many other countries in Europe and Asia before. An even more rigorous safety assessment including potential liver toxicity and insufficient peripheral action will be required for PegLispro, whereas insulin glargine U300 has the benefit of the long-term ORIGIN study.
The cost associated with use of insulin degludec in countries where it has become available is substantially higher than with any other basal insulin, e.g., some 50–70% higher compared with insulins glargine or detemir and several-fold higher than NPH insulin, which has generated considerable debate. In fact, insulin degludec has been withdrawn from marketing in Germany by the manufacturer owing to the inability to come to an agreement on the pricing. Furthermore, insulin glargine U300, with properties similar to those of insulin degludec, has meanwhile become available in many countries at a cost comparable with that of previous insulin glargine U100. As investigational insulin PegLispro is currently being evaluated in a large phase 3 clinical trial program before a labeling package can be submitted to the regulatory authorities, a final assessment of its cost–to–clinical benefit ratio seems premature. Meanwhile, Lilly has stopped the insulin PegLispro development program as a result of internal discussions (61).
This publication is based on the presentations at the 5th World Congress on Controversies to Consensus in Diabetes, Obesity and Hypertension (CODHy). The Congress and the publication of this supplement were made possible in part by unrestricted educational grants from AstraZeneca.
Article Information
Duality of Interest. E.S. has received lecturing honoraria and consultation fees from AstraZeneca, Bayer, Boehringer Ingelheim, Merck-Serono, Merck Sharp & Dohme/Merck, Novartis, and Sanofi. D.R.O. has received honoraria from Boehringer Ingelheim, Eli Lilly, Novo Nordisk, Roche, and Sanofi for lectures and involvement in advisory boards. No other potential conflicts of interest relevant to this article were reported.
