Healthcare Professional Check

This website is intended for diabetes healthcare professionals. Click one of the options below to access the content.

You are here

New forms of insulin and insulin therapies for the treatment of type 2 diabetes

The Lancet Diabetes & Endocrinology, Volume 3, Issue 8, August 2015, Pages 638 - 652


Insulin is a common treatment option for many patients with type 2 diabetes, and is generally used late in the natural history of the disease. Its injectable delivery mode, propensity for weight gain and hypoglycaemia, and the paucity of trials assessing the risk-to-safety ratio of early insulin use are major shortcomings associated with its use in patients with type 2 diabetes. Development of new insulins—such as insulin analogues, including long-acting and short-acting insulins—now provide alternative treatment options to human insulin. These novel insulin formulations and innovative insulin delivery methods, such as oral or inhaled insulin, have been developed with the aim to reduce insulin-associated hypoglycaemia, lower intraindividual pharmacokinetic and pharmacodynamic variability, and improve imitation of physiological insulin release. Availability of newer glucose-lowering drugs (such as glucagon-like peptide-1 receptor agonists, dipeptidyl peptidase-4 inhibitors, and sodium-glucose co-transporter-2 inhibitors) also offers the opportunity for combination treatment; the results of the first trials in this area of research suggest that such treatment might lead to use of reduced insulin doses, less weight gain, and fewer hypoglycaemic episodes than insulin treatment alone. These and future developments will hopefully offer better opportunities for individualisation of insulin treatment for patients with type 2 diabetes.


Insulin remains the cornerstone of diabetes treatment. More than 90 years of clinical data support use of this hormone, and it is still the most effective treatment to reduce glucose and HbA1c concentrations, even with the emergence of many new drug classes for diabetes treatment. In autoimmune type 1 diabetes, insulin is the sole treatment and has been used to save millions of lives since its discovery. In people with type 2 diabetes, insulin can be prescribed as a first-line treatment for those intolerant to other antidiabetes drugs, in the presence of advanced renal or hepatic failure, or in individuals with a primary β-cell defect—such as those with latent autoimmune diabetes of the adult ( panel ). Moreover, although many other pharmacological drugs are now available, insulin is still recommended as the preferred, if not the only, treatment for patients with type 2 diabetes who are not at the target HbA1c despite lifestyle changes and maximum dose of non-insulin treatment.1, 2, 3, and 4

Panel Indications for insulin treatment in patients with type 2 diabetes

Strong indications

  • New diagnosis or long-term diabetes with symptoms of hyperglycaemia (rescue treatment)
  • Ketoacidosis
  • Non-insulin hypoglycaemic treatments not tolerated or contraindicated
  • Acute medical events (eg, infection or myocardial infarction) or major surgery
  • Concomitant disease such as pancreatitis, cirrhosis, or chronic steroid treatment
  • Failure of non-insulin treatments (replacement treatment)

Potential indications

  • Glycaemic control not achieved with diet, exercise, non-insulin drugs (augmentation treatment), and latent autoimmune diabetes
  • Women with diabetes who are pregnant or planning a pregnancy
  • Patients admitted to hospital and unable to take their usual drug regimen or during enteral or parenteral nutrition
  • Increasingly flexible lifestyle or unplanned eating behaviour (older patients)

During the past decades many manipulations of the insulin molecule have attempted to provide an increasingly effective and safe treatment option for patients. Development of insulin treatment has given rise to long-acting and short-acting insulin analogues as alternatives to human insulin. These and emerging insulin preparations, and the availability of novel antihyperglycaemic drugs that can be combined with insulin, provide opportunities for treatment that are thought to reduce side-effects of insulin treatment and address the many pathogenetic mechanisms in type 2 diabetes. In this Review, we discuss new and emerging insulin formulations as well as strategies of insulin administration and their potential as treatments for people with type 2 diabetes. We also review evidence for combining insulin therapy with new non-insulin hypoglycaemic drugs and, finally, discuss intensification of insulin treatment with existing and emerging forms of insulin.

The present role of insulin in type 2 diabetes treatment

Type 2 diabetes is characterised by progressive deterioration of glycaemic control. Early intervention and maintenance of good glycaemic control from the time of diagnosis is strongly recommended in treatment guidelines1, 2, 3, and 4 because it is the most effective way to reduce the burden of long-term complications.5, 6, 7, and 8 However, to achieve glycaemic targets is a challenge for both patients and health-care providers ( figure 1 ). Oral hypoglycaemic agents (OHA) alone are frequently insufficient for maintenance of glycaemic control, and with the progressive loss of β-cell function that is characteristic of the disease treatment (including insulin therapy) might need to be intensified. In the UKPDS study, more than half of the participants with newly diagnosed type 2 diabetes and initially started on OHA needed additional insulin therapy within 6 years. Another study reported that 25% of patients with type 2 diabetes were given a prescription of insulin within 6 years after initiating OHA treatment, increasing to 42% after 10 years. This progressive increase shows either difficulties in ensuring adequate glycaemic control or need to intensify treatment to meet patients' glycaemic targets, which is an attempt to reduce the risk of diabetic complications. Yet, reservation still exists with respect to timely initiation of insulin therapy. The A1chieve study was a non-interventional 6 month study, including 66 726 patients with type 2 diabetes. Participants were both insulin (32·8%) and non-insulin (67·2%) users, had a mean duration of diabetes of 8 years at baseline, and mean HbA1c of 9·5% (SD 1·7) and 9·4% (1·8), respectively; 30–90% of these patients had diabetic complications. This observation suggests that insufficient intensification of treatment, including delayed introduction of insulin therapy and ineffective insulin treatment, leads to poor glycaemic control and increased risk of complications.


Figure 1 Reasons why health-care professionals and patients might refrain from starting insulin treatment Results from a case-control descriptive study of 92 patients with diabetes who needed initiation of insulin treatment because of failure (HbA1c >8·5%) on maximum oral drugs. Participants were interviewed about attitudes and thoughts regarding their illness and insulin treatment, and 157 family physicians completed a questionnaire regarding barriers to insulin treatment and answered an open-ended question about the criteria for starting insulin treatment. Figure is adapted from Nakar and colleagues, by permission of Elsevier.

Therefore the question arises as to why insulin therapy is often delayed and, if initiated, why it is not always effective. The major reservations expressed by both patients and physicians include fear of hypoglycaemia, concern for the often accompanying weight gain, the psychological stigma of initiating insulin therapy, inconvenience of injections, and need for qualified staff to instruct the patient about the use of insulin and its titration ( figure 1 ). These factors, the appropriate selection of patients with type 2 diabetes for insulin therapy, and the timing of insulin initiation have been reviewed elsewhere and will not be covered in depth in this Review. Among other concerns, long-term safety remains a factor limiting the initiation and intensification of insulin treatment. Despite insulin being used for almost a century for treatment of diabetes, few trials allow for the definition of a proper risk-to-safety ratio of insulin use in patients with type 2 diabetes. Many attitudes and suggestions about timely initiation of insulin treatment are based on a wide range of opinions rather than trial-generated evidence. A 3·3 year retrospective study, including 6484 people with type 2 diabetes who were progressing to treatment with insulin monotherapy, showed an association between insulin treatment and insulin dose and increased risk of all-cause mortality, major adverse cardiovascular events, and cancer. However, the retrospective aspect of the study calls for more specific intervention trials. Of the few published studies into this topic, the ORIGIN trial showed that use of insulin glargine to target normal fasting plasma glucose in people with type 2 diabetes and cardiovascular disease or cardiovascular risk factors, or those with impaired glucose tolerance and impaired fasting glucose, had a neutral effect on cardiovascular outcomes (hazard ratio 1·02, 95% CI 0·94–1·11, p=0·63) and cancer (1·00, 0·88–1·13, p=0·97). However, in this trial quite a low insulin dose was used compared with the higher insulin needed in patients with advanced stages of the disease, a fact to be taken into consideration in view of the dose–effect association between insulin and cardiovascular risk reported in retrospective analyses. Data from ORIGIN and UKPDS follow-up suggest that insulin treatment could be safe when introduced early in the disease course. By contrast, data from other studies suggest that intensive insulin treatment in patients with type 2 diabetes with more advanced stage and with overt complications (as those recruited in the trials ACCORD and VADT ) might increase the cardiovascular risk and all-cause mortality. Thus, appropriate use of insulin treatment needs better characterisation of the heterogeneity of the disease noted in participants with type 2 diabetes. Therefore, to decide when to initiate insulin and select an insulin treatment regimen, several elements should be considered, including actual and biological age, initial bodyweight, implications of future bodyweight gain, and presence of comorbidities—such as cardiovascular disease, renal insufficiency, impaired cognitive function, and visual impairment. Additional factors are the propensity to develop hypoglycaemia and impaired ability to recognise hypoglycaemic symptoms. Finally, personal needs have to be considered, such as acceptance of injections and willingness to monitor blood glucose, special requirements associated with jobs and leisure activities, and capacity of self-management.

Accurate education and empowerment of patients are necessary to cope with some of these aspects, but pharmacological technology might also contribute to make future insulin treatment easier, more acceptable, and, possibly, more effective with fewer drawbacks. Recent developments have attempted to address some of these drawbacks. Both weight gain and hypoglycaemia stem from the inability of the exogenously delivered insulin to mimic the physiological profile of insulin secretion. Novel insulin formulations and innovative insulin delivery methods should ensure a more physiological daily insulin profile and, thereby, decrease the risk of hypoglycaemia and the propensity for weight gain. Additionally, alternative routes of insulin administration can alleviate patients' needle phobia. Finally, combination of insulin with insulin sensitisers or novel antihyperglycaemic drugs might reduce insulin dose needed, occasionally resulting in lower risk of hypoglycaemia and less weight gain or even weight loss. However, whether these innovations could also alleviate long-term concerns so far related to classic insulin treatment needs careful surveillance and planned trials.

New insulin formulations

Long-acting analogues

Basal insulin is regarded as the optimum choice for most individuals with type 2 diabetes starting insulin treatment ( table 1 ). Generally injected at bedtime, basal insulin suppresses overnight hepatic glucose production and ensures better fasting plasma concentrations of glucose. Introduction of long-acting insulin analogues (ie, insulin glargine and insulin detemir) has been welcomed as an opportunity to achieve better glycaemic control with fewer side-effects than with neutral protamine hagedorn (NPH) insulin. Modification of the pharmacokinetics and pharmacodynamics of basal insulin results in flatter plasma insulin concentrations upon injection, better day-to-day reproducibility, and reduced risk of nocturnal hypoglycaemia. To further improve these features, new insulin formulations have been synthesised to extend their time–action profile to be longer than 24 h. Although these long-acting analogues offer the initial promise of less than once daily administration, no insulin formulation has achieved this goal.

Table 1 Pharmacokinetics of available insulin formulations

Regular human insulin 30–60 min 2–4 h 5–8 h
Aspart 12–18 min 30–90 min 3–5 h
Glulisine 12–30 min 30–90 min 3–5 h
Lispro 15–30 min 30–90 min 3–5 h
NPH 1–2 h 4–12 h 12–16 h
Lispro protamine 30–60 min 4–12 h 12–16 h
Detemir 1–2 h 6–8 h up to 24 h
Glargine 1–2 h None 20–26 h
Glargine U300 1–2 h None up to 36 h
Degludec 30–90 min None >42 h
70% NPH, 30% regular 30–60 min 2–4 h 10–16 h
50% NPH, 50% regular 30–60 min 2–5 h 10–16 h
30% aspart protamine, 70% aspart 5–15 min 1–4 h 10–16 h
50% aspart protamine, 50% aspart 15–30 min 1–4 h 10–16 h
70% aspart protamine, 30% aspart 15–30 min 1–12 h 10–16 h
50% lispro protamine, 50% lispro 10–15 min 1–4 h 10–16 h
75% lispro protamine, 25% lispro 10–15 min 1–12 h 10–16 h

NPH=neutral protamine hagedorn.

Degludec is a desB30 insulin acylated at the LysB29 residue with a glutamate linker and a hexadecandioyl fatty acid side chain, which was approved for licensing by the European Medicines Agency whereas the US Food and Drug Administration (FDA) has requested a cardiovascular safety trial before making any further decisions. Degludec shows an insulin-receptor binding specificity and a metabolic-to-mitogenic ratio that is comparable with that of human insulin. Its mean half-life is 24·5 h and its metabolic effect is still apparent 42 h after injection. In a pre-planned meta-analysis of phase 3 studies, which included 5299 people with type 2 diabetes, use of insulin degludec was associated with a significantly lower rate of overall confirmed and nocturnal episodes of hypoglycaemia than with insulin glargine (relative risk 0·83, [95% CI 0·74–0·94] and 0·68 [0·57–0·82], respectively). Whether this difference translates into clinical benefits in real-life settings, however, needs extensive assessment. On the basis of a study published in 2014, one patient needs to be treated for 4 months with insulin degludec versus glargine to avoid one confirmed hypoglycaemic episode, and two patients need to be treated for 1 year to avoid one nocturnal confirmed hypoglycaemic episode. Moreover, the cost-effectiveness for the use of degludec versus glargine in patients with type 2 diabetes has not yet been extensively assessed; however, analyses done on the basis of data generated by the phase 3 studies have shown that insulin degludec is a cost-effective treatment compared with glargine and offers additional benefits for people with type 2 diabetes and recurrent hypoglycaemia.

Although insulin degludec is the latest long-acting insulin analogue to be approved, other long-acting insulin formulations are being developed. PEGylated insulin (LY2605541), which consists of insulin lispro with a covalently bound polyethylene glycol moiety at lysine B28, slows the absorption of insulin from the injection site and reduces renal insulin clearance, resulting in a half-life as long as 75 h with a flat time–action profile. Initial results from phase 3 studies comparing it with insulin glargine reported similar efficacy in terms of HbA1c reduction. Use of PEGylated insulin was associated with a lower risk of hypoglycaemia and less apparent increase in bodyweight than with insulin glargine, but results from larger and longer-term studies are needed before crucial comparison with existing long-acting insulin analogues is possible.

Glargine U300 is a new, more concentrated (300 U/mL), formulation than insulin glargine U100 with no other molecular changes. This new formulation was approved in the beginning of 2015 for market use by the FDA and the European Medicines Agency. Upon injection, glargine U300 forms a compact subcutaneous depot with a small surface area that results in a more gradual, long-term, and flatter release than with standard glargine, enabling glucose control for up to 36 h. Glargine U300 has been compared with insulin glargine in clinical trials of patients with type 2 diabetes who were given basal-bolus insulin or basal insulin plus OHAs.36, 37, and 38 Basal insulin doses were titrated to target a similar fasting plasma glucose (4·4–5·6 mmol/L). After 6 months of treatment, HbA1c reduction was equivalent between regimens, although use of glargine U300 was associated with fewer instances of nocturnal confirmed hypoglycaemia (≥70 mg/dL) or severe hypoglycaemia.36, 37, and 38

Thus far, the ultra-long pharmacokinetic profile of the novel, long-acting insulin analogues provides the clinical benefit of reduced cases of nocturnal hypoglycaemia. Whether the use of these and other insulins in development with increasingly long and flat profiles will lead to improved glycaemic control has yet to be shown. Moreover, possible benefits of the novel analogues will still need to be balanced by long-term safety and efficacy data.

Rapid-acting analogues

Although basal insulin is preferred for initiation of insulin treatment in patients with type 2 diabetes, many individuals then later advance to a disease stage when prandial insulin is necessary. Rapid-acting insulin analogues were introduced to the market in the mid-1990s and have undergone continuous developments ( table 2 ). These analogues have a faster onset and shorter duration of action than human insulin and provide more control of post prandial glucose than does human insulin, with no significant difference in the risk of hypoglycaemia. However, the pharmacokinetic profile of rapid-acting insulins is far from ideal because they only estimate the pattern of physiological insulin release. In particular, these insulin analogues cannot fully reproduce the prompt increase in circulating insulin upon ingestion of a meal. In the past decade, several attempts have been made to develop ultra-rapid-acting insulins, many of which have focused on conjugation with excipients or enzymes that accelerate absorption of insulin monomers, whereas other approaches have attempted to change insulin pharmacokinetics by affecting route of delivery ( table 2 ).

Table 2 Future investigational rapid-acting insulins for treatment of diabetes

BioD-090 (VIAject) Recombinant insulin plus edetic acid (EDTA) Loosely packed insulin multimers with rapid dissociation into monomers and dimers
rHuPH20 (Hylenex) Recombinant insulin plus hyaluronidase Increased subcutaneous insulin spreading, accelerated pharmacokinetics
Ultra-fast-acting insulin aspart (FIAsp) Recombinant insulin plus nicotinamide and arginine Increased local blood flow
BioChaperone ultra-rapid-acting insulin BioChaperone plus lispro insulin Enhanced insulin diffusion to aid absorption into blood circulation
Local heating of injection site Patch devices to apply mild heat on subcutaneous injection site Increased tissue perfusion, accelerated insulin pharmacokinetics
Intradermal delivery Intradermal insulin application Rapid insulin absorption
Inhaled insulin Technosphere inhalation Large surface area but bioavailability, high doses needed but peak insulin concentration reached early after administration
Oral insulin spray Absorption of insulin through mucosa of oropharynx Accelerated insulin pharmacokinetics, increased size of absorption area
Jet delivery Subcutaneous jet injection of insulin Dispersion of insulin over a large subcutaneous area
Injectable nano-network (Smart insulin) Dextran nanoparticles loaded with insulin and glucose-specific enzymes Glucose-dependent insulin release

Changes in the formulation of subcutaneously injected insulin to increase insulin absorption rate has been attempted by different manipulations. Addition of edetic acid, a chelator of zinc, to the insulin hexamer (BioD-090, VIAject; Biodel Inc, Danbury, CT, USA) causes a more rapid dissociation of the hexamer into dimers and monomers when subcutaneously injected, resulting in a faster maximum glucose-lowering effect. Overall experience of rapid-acting analogues in patients with type 2 diabetes is small and very rare. In 14 patients with type 2 diabetes, a single dose of BioD-090, human regular insulin, or insulin lispro before a test of a standardised liquid meal was associated with comparable absolute plasma glucose concentrations and similar timecourse of plasma glucose concentrations between insulin types.

Injection of a rapid-acting insulin with recombinant human hyaluronidase (rHuPH20) leads to disruption of the hyaluronic acid in the subcutaneous adipose layer, enabling a more rapid spread of the locally injected insulin, and better absorption than rapid-acting insulins without rHuPH20. In a trial including 21 patients with type 2 diabetes, use of insulin lispro with rHuPH20 provided better control of glycaemic excursion than with lispro alone, with lower insulin requirements and lower hypoglycaemic excursion, as indicated by an improvement of both hyperglycaemic (area under the curve [AUC]0–4h >7·8 mmol/L, 56% of control, p=0·48) and hypoglycaemic (AUC0–8 h <3·9 mmol/l, 34% of control, p=0·33) excursions.

Fast-acting insulin aspart (FIAsp; Novo Nordisk, Bagsvaerd, Denmark), developed as a modification of insulin aspart, is at advanced phase 3 trials in both patients with type 1 diabetes and those with type 2 diabetes. This insulin is a combination of rapid-acting insulin aspart, nicotinamide, and arginine, resulting in a faster initial absorption after subcutaneous injection. Fast-acting insulin aspart is thought to increase local blood flow, yet little information is available about its mechanism of action.

In December, 2014, Eli Lilly (Indianapolis, IN, USA) announced an alliance with Adocia (Lyon, France) for the development of an ultra-rapid-acting insulin based on BioChaperone technology. By forming a physical complex with rapid-acting insulin lispro, BioChaperone protects from enzymatic degradation and exerts a stabilising and solubilising effect that increases absorption and bioavailability with accelerated action. In an initial study including 37 participants with type 1 diabetes, BioChaperone lispro showed a faster rate of absorption after subcutaneous injection than insulin lispro, with earlier insulin exposure, higher peak concentration, and more rapid clearance of this hormone. In this study, participants with type 1 diabetes were given BioChaperone with U100 lispro; a BioChaperone with U300 lispro is in development for use in people who are obese and have type 2 diabetes.

Other approaches to optimise insulin delivery

Physical methods to change insulin pharmacokinetics have also been explored as a means to increase speed of action of injected insulin. Topical heating of the subcutaneous injection site is a novel approach to increase insulin's bioavailability. A small patch device (InsuPatch; Insuline Medical, Petach Tikvah, Israel) locally applies mild heat (40°C) to the skin at the location of the subcutaneous insulin injection. Raised skin temperature leads to vasodilatation and increased tissue perfusion, promoting absorption of the subcutaneously injected insulin. Data are limited to people with type 1 diabetes, with the device being studied in 17 patients who underwent injection of 0·2 U/kg bolus-insulin aspart with and without the warming device under the conditions of a euglycaemic clamp. The warming device led to an earlier peak of insulin action than without the device (43 min vs 73 min), whereas the total AUC for the time–action profile and the peak action did not differ with and without infusion warming. This product has received the European Conformity (CE) mark and is marketed for use in Germany and Israel and additional studies are underway aiming for FDA approval.

Intradermal injection of insulin leads to a rapid increase in its systemic availability. Compared with the subcutaneous layer, the dermis layer of skin allows faster absorption of the injected protein because it has greater vascularisation and a denser network of lymph vessels. Furthermore, regional capillaries have thinner vessel walls and reduced endothelial barrier, both of which promote rapid absorption. Again, clinical information about dermal insulin administration is limited to 22 individuals with type 1 diabetes in whom intradermal versus subcutaneous delivery of insulin led to earlier attainment of maximum insulin concentrations (36 min vs 52 min), yet did not change post prandial concentrations of glucose.

Faster delivery of insulin to the systemic circulation can be achieved with routes of delivery other than subcutaneous injection. Of the different approaches explored, insulin inhalation and absorption through lung alveoli has received the greatest attention. Inhaled insulin differs from injectable insulins not only in the route of administration, but also by dosing units, patient eligibility, and need for periodic safety testing. Inhaled insulin is contraindicated in people with asthma or chronic obstructive pulmonary disease, because it can cause acute bronchospasm. A detailed medical history, physical examination, and spirometry might be necessary before initiation of inhaled insulin to identify potential lung disease. Due to insulin dispersion through airways, high doses of inhaled insulin must be given to achieve a therapeutic response equivalent to injected insulin. Exubera (Pfizer, New York, NY, USA) was the first inhaled, rapid-acting insulin and was first available in August, 2006, but was withdrawn from the market in October, 2007, because it did not gain acceptance by physicians and patients. Exubera had a bulky device to dispense insulin and little dosing flexibility. By contrast, Afrezza (Sanofi, Bridgewater, CT, USA) is a novel form of inhaled insulin and was approved by the FDA in June, 2014, as a short-acting insulin. It is based on a technosphere platform that contains recombinant human insulin dissolved in dry powder. Onset of action of this insulin is faster than that of the rapid-acting analogues; peak insulin concentration is reached in 15–30 min. Afrezza is administered via a small thumb-sized device and has greater flexibility in dosing than Exubera. A pilot study undertaken with eight people with type 2 diabetes showed that Afrezza can be used irrespective of variations in meal carbohydrate content with improvements noted in HbA1c during 19 weeks of treatment, although the study did not include a control treatment group. A larger study with 618 patients showed that treatment with Afrezza could be implemented with no negative effects on patients' health-related quality of life and had few diabetes-related concerns, as shown by specific questionnaires completed.

Needle-free delivery of insulin could be seen as a potential advantage by some patients, and a role for inhaled insulin to help with insulin initiation has been suggested but not proven.54 and 55 Moreover, improvements in patients' perception of insulin treatment, treatment satisfaction, and treatment preference did not differ with use of Afrezza compared with pre-mixed insulin. Furthermore, cost of the delivery device and the insulin preparation might be a drawback for most people with type 2 diabetes.

Many other routes of administration have been tested, with little success. Oral insulin56 and 57 is still not available in clinical practice and, despite several approaches attempted, only a few have progressed to clinical trials.58 and 59 Other strategies to deliver insulin include enteric-coated capsules and use of adjuvants, liposomes, and nanoparticles (which have been reviewed elsewhere ). Oral spray insulin and subcutaneous jet delivery of insulin have been attempted but not widely accepted.

Although smart insulin might be the future of insulin treatment, it is still in the early stages of development. Injectable enzyme nanocapsules (made of a pH-responsive chitosan matrix and recombinant human insulin) can act as a self-regulating valve system to release insulin at basal rates in normoglycaemic conditions and at higher rates in response to hyperglycaemia. These forms are being tested in animal models of diabetes.

Outlook for insulin formulations

In summary, many attempts have been made in the past decade to improve insulin treatments ( figure 2 ) directed at both improving the formulation of insulin and increasing patient compliance and acceptability with respect to delivery mode. However, whether these pharmacological and technological developments of insulin treatment will have a truly tangible effect on the long-term maintenance of metabolic control and related outcomes will need careful assessment, especially with respect to cost-effectiveness and long-term safety-to-efficacy ratio. All of these new insulin options are expected to be more expensive, further increasing what is already a costly treatment. Moreover, several concerns are associated with chronic insulin treatment, which are yet to be properly addressed. Chronic hyperinsulinaemia, a common feature in type 2 diabetes perpetuated by exogenous insulin administration, has raised concern because of the hormone's atherogenic and mitogenic properties. In 2013, retrospective studies of 84 622 patients from the UK general practice research database reported that exogenous insulin, compared with OHAs, was associated with increased risk of diabetes-related complications, cancer, and all-cause mortality. The same authors reported an increased risk of mortality in 20 005 patients with type 2 diabetes who need insulin treatment because of OHA failure. Several retrospective analyses19, 64, 65, and 66 have repeatedly reported an increased mortality in patients with type 2 diabetes treated with insulin. However, these were retrospective studies, and their findings should on the one hand be viewed with caution, but on the other hand cannot be ignored, since prospective studies have raised concern about side-effects and the safety of insulin treatment. Moreover, insulin treatment was associated with an increase in non-fatal cardiac events during the follow-up of the Digami 2 trial. The ACCORD trial was prematurely stopped because of an excess of mortality in the intensively treated patients, most of them receiving, among many treatments, insulin. Although high-insulin dose and increased duration of treatment have been also associated with an increased risk of cancer,19 and 68 these results should be taken with caution because of multiple confounders by indication, poor recording of exposure time, and intensity of diagnosis. So far, only the ORIGIN trial has reported acceptable safety margins of insulin treatment, although the study population included participants with recently diagnosed short-duration type 2 diabetes or even individuals with impaired glucose tolerance who require quite low doses of insulin (0·4 U/kg). In these participants, no excess of cardiovascular events, cardiovascular mortality, and cancer were reported compared with non-insulin treatments (mainly metformin and sulfonylureas). Moreover, glargine treatment was associated with a reduced incidence of microvascular complications of the kidney and the retina in those individuals with a baseline HbA1c of 6·4% or more (hazard ratio 0·90, 95% CI 0·81–0·99). However, the extent to which these results can be applied to people with type 2 diabetes with longer duration of the disease or worse initial glycaemic control remains questionable. Therefore more long-term randomised clinical trials are needed that specifically address the risk-to-safety ratio of insulin treatment.70 and 71 Because of all these reasons, insulin treatment in people with type 2 diabetes is likely to develop in the future in terms of newer pharmacological or technological formulations to better mimic physiological insulin profiles, reducing the risk of unnecessary chronic hyperinsulinaemia, hypoglycaemia, and bodyweight gain. Possibly at the same time, novel combination treatments might emerge as a way to reduce some of the potential side-effects of insulin treatment and also address the many pathogenetic mechanisms responsible for disease progression.


Figure 2 Timeline for the development of short-acting, long-acting, and future rapid-acting analogues of insulin NPH=neutral protamine hagedorn. PEG=polyethylene glycol. EDTA=edetic acid.

Combination of insulin treatment with other antidiabetic drugs

Combination of insulin treatment with insulin sensitisers or with incretin-based treatments might enable attainment of similar, or possibly better, glycaemic goals than insulin treatment alone, but with lower insulin needs, less weight gain, and less hypoglycaemia.

Insulin resistance is a common disorder in people with type 2 diabetes and to address this disorder at the time of insulin initiation is generally considered a rational approach. Metformin is an effective adjunct to insulin treatment in people with type 2 diabetes. Its addition to insulin, or its use with initiation of insulin treatment, results in improved glycaemic control, lower insulin need, and no weight gain compared with insulin treatment alone,72, 73, and 74 although final evidence of cardiovascular protection of metformin is yet to be confirmed.

In the PROactive study, addition of pioglitazone versus placebo to patients receiving insulin was studied in the subgroup of patients receiving insulin at baseline. Mean HbA1c concentrations and insulin dose decreased in the pioglitazone group compared with the placebo group. However, the combination of insulin and pioglitazone could increase fluid retention and risk of bone fractures.76 and 77

Incretin-based therapies (consisting of glucagon-like peptide-1 [GLP-1] receptor agonists and dipeptidyl peptidase-4 [DPP-4] inhibitors) have shown benefit as a monotherapy and in combination with different antihyperglycaemic drugs. GLP-1 also exerts complementary action with insulin and combinations of insulin with incretin-based treatments have been investigated as a pharmacological approach.

Even with the restricted experience in the addition of a DPP-4 inhibitor to pre-existing insulin treatment in individuals with type 2 diabetes, these inhibitors have been shown to reduce HbA1c compared with insulin therapy alone, with no significant effect on bodyweight and variable response in terms of hypoglycaemia ( table 3 ).80, 81, 82, 83, 84, 85, 86, 87, and 88 To draw conclusions on the basis of trials that tested this combination treatment is difficult because of different patients and background treatments in every study. Insulin regimens used were quite variable, including intermediate, long-acting, and pre-mixed insulin with or without metformin. Because of this variability and differing study designs (ie, allowing80, 81, 86, and 88 or not allowing82, 83, 84, and 85 dose adjustments), insulin-sparing effects of combining a DPP-4 inhibitor cannot be ascertained from the available evidence.

Table 3 Synopsis of clinical trials assessing safety and efficacy of adding a dipeptidyl peptidase-4 inhibitor to existing insulin treatment

            Symptomatic (%) Severe (%)  
Fonseca et al (2007) Poorly controlled glucose, insulin monotherapy 144 (vildagliptin 50 mg bid), 152 (placebo) Vildagliptin: 81·2 (SD 44·8); placebo: 81·9 (SD 49·4) 24 Vildagliptin: −0·5 (SD 0·1); placebo: 0·2 (SD 0·1) Vildagliptin: 1·95; placebo: 2·96 Vildagliptin: none ; placebo: 0·1 Vildagliptin: 1·3 (SD 0·3); placebo: 0·6 (SD 0·3)
Lukashevich et al (2013) Poorly controlled glucose with renal impairement, had insulin with or without OHA treatment 100 (vildagliptin 50 mg qd), 78 (placebo) Vildagliptin: 53·1 (SD 35·9); placebo: 49·6 (SD 41·6) 24 Vildagliptin: −0·9 (SD 0·4); placebo: −0·6 (SD 0·2) Vildagliptin: 19·0; placebo: 14·1 Vildagliptin: 2·0; placebo: 2·6 Vildagliptin: −0·1; placebo: 0·7
Kothny et al (2013) Poorly controlled glucose, had insulin with or without metformin treatment 228 (vildagliptin 50 mg bid), 221 (placebo) Vildagliptin: 39·9 (SD 18·1); placebo: 41·9 (SD 20·4) 24 Vildagliptin: −0·8 (SD 0·1); placebo: −0·1 (SD 0·1) Vildagliptin: 8·4; placebo: 7·2 Vildagliptin: 0·9; placebo:0·9 Vildagliptin: 0·1; placebo: −0·4
Kozlovski and colleagues (2013) Poorly controlled, had insulin with or without metformin treatment 87 (vildagliptin 50 mg bid), 86 (placebo) Vildagliptin: 39·5 (SD 15·8); placebo: 39·5 (SD 15·3) 24 Vildagliptin: −0·8 (SD 0·2); placebo: −0·03 (SD 0·2) Vildagliptin: 8·0; placebo: 8·1 None Vildagliptin: 0·3; placebo: −0·2
Barnett et al (2012) Poorly controlled glucose, had insulin with or without metformin treatment 304 (saxagliptin qd), 151 (placebo) Saxagliptin: 53·6 (range 19–150); placebo: 55·3 (range 30–149) 24 Saxagliptin: −0·73; placebo: −0·32 Saxagliptin: 18·4; placebo: 19·9 Saxagliptin: 1·0; placebo: 1·3 Saxagliptin: 0·39; placebo: 0·18
Hong et al (2012) Poorly controlled glucose, had insulin with or without OHA treatment 61 (sitagliptin 100 mg bid), 63 (placebo) Sitagliptin: 39·6 (SD 19·1); placebo: 35·4 (SD 16·3) 24 Sitagliptin: −0·6 (SD 0·1); placebo: −0·2 (SD 0·1) Sitagliptin: 8·2; placebo: 17·5 Sitagliptin: 1·6; placebo: 4·8 Sitagliptin: −0·7 (SD 0·1); placebo: 1·1 (SD 0·4)
Arnolds et al (2010) Poorly controlled glucose, had insulin with or without OHA treatment 16 (sitagliptin 100 mg bid), 16 (placebo) Sitagliptin: 33·4; placebo: 32·3 4 Sitagliptin: −1·5 (SD 0·7); placebo: −1·2 (SD 0·5) Sitagliptin: 3·3; placebo: 1·6 None Sitagliptin: 0·1 (SD 1·6); placebo: 0·4 (SD 1·5)
Visbøll et al (2010) Poorly controlled glucose, had insulin with or without metformin treatment 322 (sitagliptin 100 mg bid), 319 (placebo) Sitagliptin: 44·2 (SD 29·9); placebo: 44·5 (SD 25·7) 24 Sitagliptin: −0·6 (range −0·7 to −0·5); placebo: 0 (range −0·1 to 0·1) Sitagliptin: 16; placebo: 8 Sitagliptin: 0·6; placebo: 0·03 Sitagliptin: 0·1; placebo: 0·1
Yki-Järvinen et al (2013) Poorly controlled glucose, had insulin with or without metformin treatment 543 (linagliptin 5 mg qd), 520 (placebo) Linagliptin: 41·5 (SD 31·9); placebo: 40·1 (SD 27·3) 76 Linagliptin: −0·48 (SD 0·08); placebo: 0·05 (SD 0·08) Linagliptin: 31·4; placebo: 32·9 Linagliptin: 1·7; placebo: 1·1 Linagliptin: −0·3 (SD 0·19); placebo: −0·4 (SD 0·18)

* p<0·01.

Events per patient-year.


§ p<0·05.

SD or range not reported.

Least squares mean change (95% CI).

OHA=oral hypoglycaemic agents. Bid=twice daily. Qd=once daily.

Compared with combination treatment that includes a DPP-4 inhibitor, better glycaemic control could be anticipated with the combination of insulin and a GLP-1 receptor agonist, because increased concentrations of circulating GLP-1 can be achieved with a GLP-1 receptor agonist. Because of greater activation of GLP-1 receptor gastric emptying will be delayed, contributing to reduction of post prandial glucose excursions and providing minimal weight gain or weight loss in patients on combination treatments. Both short-acting and long-acting GLP-1 receptor agonists result in improved glycaemic control, weight loss, and reduced insulin dose compared with insulin treatment alone ( figure 3 ). Besides observational data (reviewed elsewhere79 and 89) Buse and colleagues reported that 30 weeks after addition of exenatide twice daily to existing insulin glargine, HbA1c was decreased by 1·74% compared with a 1·04% reduction in the placebo control group (between-group difference −0·69%, 95% CI −0·93 to −0·46%; p<0·001). Patients assigned to the glargine group needed higher insulin doses by the end of the trial than did those assigned to exenatide. Additionally, those patients in the glargine group gained 0·96 kg compared with a 1·78 kg weight reduction in the exenatide group (−2·7 kg, −3·7 to −1·7; p<0·001). Similar results have been obtained with the addition of lixisenatide once daily to the drug regimen of patients with type 2 diabetes previously treated with long-acting (glargine or detemir) or pre-mixed insulin.91, 92, and 93 Lixisenatide was associated with a greater reduction in HbA1c, bodyweight, and insulin need than placebo.91, 92, and 93 Symptomatic hypoglycaemia and gastrointestinal adverse events were more common in the patients treated with lixisenatide than those given placebo.


Figure 3 Effects of insulin added to existing treatment with glucagon-like peptide-1 receptor agonist or vice versa on glycaemic control (HbA1c), bodyweight, and insulin dose Summary of results from observational studies. All three variables (HbA1c, bodyweight, and insulin dose) decreased from baseline to end of treatment (endpoint). Duration of treatment varied from 26 weeks to 48 months. Every coloured line represents one study. Figure is reproduced from Balena and colleagues, by permission of Wiley. GLP-1 RA=glucagon-like peptide-1 receptor agonist.

Although the addition of a short-acting GLP-1 receptor agonist to control meal-related hyperglycaemia might seem rational, another appealing option is the combination of long-acting GLP-1 receptor agonists. In the few controlled studies79 and 89 that assessed this combination of long-acting GLP-1 receptor agonists with insulin, an overall improvement in glycaemic control was reported in the absence of bodyweight gain, and with variable levels of hypoglycaemia, which were generally low.79 and 89 To establish whether use of a short-acting rather than a long-acting GLP-1 receptor agonist is more advantageous, prospective head-to-head comparisons are needed. Nonetheless, the combination of long-acting GLP-1 receptor agonists and basal insulin is worth considering, because an increasing number of patients will already be on a GLP-1 receptor agonist before insulin is considered as add-on therapy.

How should clinicians and their patients choose between basal insulin and an incretin-based drug when OHA therapy is unsuccessful? Similar to insulin, GLP-1 receptor agonists have to be injected once or twice per day, but once-a-week preparations have now also become available. Efficacy of GLP-1 receptor agonists is well documented compared with basal insulin treatment. In the trial by Bunck and colleagues, use of insulin glargine once daily and exenatide twice daily and thrice daily ensured satisfactory β-cell function and glycaemic control with no difference between treatments. Diamant and colleagues reported results of the DURATION-3 trial after a 3 year follow-up that compared use of exenatide once per week and insulin glargine once daily in addition to their existing oral glucose-lowering regimens. After the 3 years both treatments were associated with significant and comparable improvements to HbA1c concentrations (exenatide 6·6% [SD 0·2] and insulin glargine 6·9 [0·2], p=0·186). However, hypoglycaemic events were more common in the glargine group than the insulin group. Similarly, liraglutide added to metformin and sulfonylurea produced significant improvements in glycaemic control and bodyweight than placebo or insulin glargine. Although all of these trials included only a small number of patients, equivalent glycaemic control between GLP-1 receptor agonists and basal insulin has been documented in a network meta-analysis reported in 2013. On the basis of these results,94, 95, and 96 GLP-1 receptor agonists seem to be a potential alternative to insulin at the time of OHA failure, prompting the question of how to select one treatment rather than the other. The reduced risk of hypoglycaemia and potential beneficial effect in preventing gains in bodyweight could help guide treatment selection, with insulin still available for patients in whom GLP-1 receptor agonists might be contraindicated. A third treatment option can be the combined use of both drugs from the time of failure of OHA. For example, a recent trial compared the use of insulin degludec or liraglutide with a fixed-ratio combination of the two (to create IDegLira [NovoNordisk, Bagsvaerd, Denmark]) in patients with type 2 diabetes inadequately controlled with oral hypoglycaemic drugs. After 26 weeks of treatment, mean HbA1c was lowered by 1·4% to 6·9% with insulin degludec, and by 1·3% to 7·0% with liraglutide. The reported decreases in HbA1c with the two drugs used as a fixed-ratio combination (−1·9% to 6·4%) was non-inferior to insulin degludec (−0·47%, 95% CI −0·58 to −0·36%; p<0·0001) and better than liraglutide (−0·64%, −0·75 to −0·53; p<0·0001). With the fixed-ratio combination lower daily doses of both insulin degludec and liraglutide were necessary for adequate glycaemic control and the main side-effects of basal insulin (weight gain and hypoglycaemia) and GLP-1 receptor agonists (gastrointestinal adverse events) were attenuated with IDegLira compared with the components used separately. Similar results have been reported in patients with established treatment using basal insulin who were randomly assigned to either the once per day fixed-combination group or just degludec insulin group; however, in this study the rate of hypoglycaemia was similar in both treatment groups (24% vs 25%). Whether early introduction of the insulin and GLP-1 receptor agonist combination will result in a more sustained glycaemic control than either component alone, or whether this might relieve some of the concerns associated with insulin treatment is yet to be established.

Selective inhibitors of sodium-glucose co-transporter-2 (SGLT2) are a novel class of oral antidiabetic drugs recently approved in many countries. Clinical data show benefits in terms of glycaemic control, weight loss, and blood pressure reduction. Their long-term safety is being assessed in large-scale clinical trials. If used as a combination therapy with insulin, SGLT2 inhibitors promote weight loss in addition to lowering glucose concentrations, thereby curbing the weight gain associated with insulin treatment. In a trial, 808 people with type 2 diabetes with HbA1c of 7·5–10·5% and on a stable insulin dose of 30 IU/day or more (most were receiving basal-bolus insulin), with or without up to two oral antidiabetic drugs before participation, were randomly assigned to receive placebo or 2·5 mg/day, 5 mg/day, or 10 mg/day of the SGLT2 inhibitor dapagliflozin for 104 weeks. At 48 weeks, patients on dapagliflozin 5 mg/day were switched to 10 mg/day. At the study end, differences from placebo in HbA1c changes from baseline were −0·4% (95% CI −0·6 to −0·2, p=0·0002) in the 5 mg/day to 10 mg/day group and −0·4% (−0·6 to −0·2, p=0·0007) in the 10 mg group. In the placebo group, mean insulin dose increased progressively by 18·3 IU/day (95% CI 13·7–22·9) and weight increased by 1·83 kg (1·05–2·9), whereas in the dapagliflozin groups, the insulin dose was stable. However, both dapagliflozin groups had significant differences in bodyweight changes from baseline compared with placebo (5 mg/day to 10 mg/day group −2·86 kg [95% CI −3·92 to −1·80], p<0·0001; 10 mg/day group −3·33, [−4·38 to −2·27], p<0·0001). A similar effect was reported in a trial including 563 patients with inadequately controlled type 2 diabetes who were obese (BMI 30–45 kg/m2) and on more than 60 IU/day of insulin (multiple injections) alone or in combination with metformin. After a 2 week placebo run-in, patients were randomly assigned to receive either once per day empagliflozin (at a dose of 10 mg or 25 mg) or placebo. At week 52, empagliflozin groups at 10 mg/day and 25 mg/day had reduced insulin doses by 9 IU/day or 11 IU/day and weight by 2·4 kg or 2·5 kg, respectively, whereas placebo had no difference in the rate of hypoglycaemia in the three groups. Overall, adjusted mean HbA1c was reduced by 0·81% (SE 0·08%) with placebo, 1·18% (0·08%) with 10 mg/day empagliflozin, and 1·27% (0·08%) with 25 mg/day empagliflozin. The insulin substudy of the CANVAS trial included 278 individuals receiving insulin monotherapy or insulin in combination with other antihyperglycaemic drugs. Patients were randomly assigned to receive 100 mg or 300 mg canagliflozin, or placebo once per day. After week 18, the mean change from baseline HbA1c was −0·76% (SD 0·08) for those receiving 100 mg/day canagliflozin, −0·79% (0·07%) for 300 mg/day canagliflozin, and 0·1% (0·08%) for placebo (p<0·001). Mean change of bodyweight from baseline was −1·8% (0·3%), −2·7 (0·3%), and no change (0·3%), respectively, (p<0·001).

In summary, the little data on combination of SGLT2 inhibitors with insulin are promising. These drugs can be added at any stage of disease (providing glomerular filtration rate is more than 60), resulting in improved glycaemic control, even in patients who have taken complex insulin regimens for many years.

Intensification of insulin treatment

Type 2 diabetes is a progressive disorder and intensification of treatment is commonly needed with time. Insulin is generally added to ongoing oral antidiabetes treatment with subsequent treat-to-target dose adjustments. The simplest approach is to add basal insulin, such as NPH or a long-acting analogue.2, 3, and 4 A consensus statement from the American Association of Clinical Endocrinologists and the American College of Endocrinology strongly recommends against NPH insulin use in favour of long-acting insulin analogues. Other guidelines, such as the National Institute for Health and Care Excellence, recommend the use of long-acting insulin analogues (detemir and glargine) only in specific situations—eg, repeated episodes of hypoglycaemia or otherwise need for two injections per day. Although efficacy of long-acting insulin analogues is comparable with that of the NPH insulin combination, risk of hypoglycaemia is lower. Furthermore, detemir is associated with decreased weight gain, although two injections per day might be necessary with this insulin form. Despite these advantages, a high percentage of participants do not achieve a target HbA1c of less than 7·0%. In clinical trials in which long-acting insulin analogues were compared with NPH insulin, about 40% of participants did not meet the glycaemic target on optimum titration of basal insulin and oral drugs (mostly metformin and sulfonylureas).106 and 107 Whether insulin degludec or glargine U300 can increase the number of participants with type 2 diabetes achieving their glycaemic target is still to be determined; however, despite some further reduction in the risk of nocturnal hypoglycaemia, no advantages could be detected in clinical trials compared with insulin glargine.108 and 109

Addition of non-insulin hypoglycaemic drugs to present basal insulin treatment is a simple and often successful strategy. However, several patients might not respond to or progressively lose response to this approach so that after intensified basal insulin fails, and having maximised the potential of non-insulin drugs, thought should be given to the addition of prandial insulin.

Several protocols have been proposed for insulin intensification, including bolus-plus (addition of prandial insulin before breakfast or the main meal), a full basal-bolus regimen, or use of pre-mixed insulin. The relative safety and efficacy to initiate a pre-mixed rather than a basal-bolus or basal-plus insulin regimen in individuals naive to insulin has been assessed in several trials. The trials showed similar110, 111, and 112 or lower glycaemic efficacy, and the frequency of hypoglycaemia was not different in some studies111 and 112 and higher107 and 110 in the other ones. A lower safety profile of pre-mixed insulin might arise from the fixed, inflexible ratio of the prandial to basal insulin and also from the use of protamine insulin as the long-acting component, which has discernible peak activity 4–12 h after administration. The little data available suggests that the pre-mixed formulation of insulin degludec-insulin aspart might decrease the frequency of hypoglycaemia compared with older pre-mixed insulins,113 and 114 and therefore its use could combine the benefits of a pre-mixed formulation (ie, a single injection) without the failings. At present, use of protamine-based pre-mixed insulin is not generally encouraged unless particular circumstances arise in which benefit surpasses the limitations. Most of these circumstances are patient-related, including reluctance or incapacity to embark in more complex insulin regimens.

A bolus-plus regimen has been proposed in the guidelines from the American Diabetes Association and the European Association for the Study of Diabetes as the next step to insulin intensification. This treatment requires addition of one bolus of short-acting insulin to existing basal insulin. Several clinical trials have reported non-inferiority of one prandial injection compared with many injections for attainment of glycaemic targets. Intensification with a complex insulin regimen should be saved for individuals who might truly benefit from it, such as those with substantial loss of β-cell function and in whom maintenance of strict glycaemic control is advisable to prevent microvascular complications. This judgment, once again, calls for a process of insulin intensification to be undertaken in an individualised manner, accounting for each patient's glycaemic target, eating habits, lifestyle, and personal preferences for treatment.

Use of an insulin pump could improve absorption and reduce variability of the basal component of insulin infusion and enhance compliance to prandial insulin injections. In a randomised trial, 331 people with type 2 diabetes needing high doses of insulin (0·7–1·8 U/kg per day) and with poor glycaemic control (HbA1c 8–12%) were randomly assigned (after a 2 month insulin dose optimisation run-in treatment) to treatment with an insulin pump or continuation with multiple drug injections. After 6 months, the mean total daily insulin dose was 97 U (SD 56) in the group using the pump compared with 122 U (68) in the group given multiple drug injections (p<0·0001). HbA1c had decreased in both groups from a mean of 9·0% to 7·9% (mean change −1·1% [SD 1·2]) in those using pump and from 9·0% to 8·6% (mean change −0·4% [SD 1·1]) in the group given multiple drug injections, with a significant difference between the two groups (0·7%, 95% CI −0·9 to −0·4%; p<0·0001). Both treatment groups reported similar weight gains (mean 1·5 [SD 3·5] kg vs 1·1 [SD 3·6] kg, p=0·25) and no difference in hypoglycaemic events as recorded by a masked 6 day continuous monitoring of glucose concentrations done at the end of the trial. Although these data seem promising, the cost and the manpower needed to train patients in pump use could represent a major limitation. Insulin pump treatment might be considered in patients with type 2 diabetes who are quite young and highly motivated because the associated technology can be demanding. Additionally, this treatment can be considered particularly for those with a very active life who need as much flexibility as possible in insulin delivery. Administration of novel short-acting insulin analogues via insulin pumps is being investigated and might offer even further benefits to those reported with available short-acting analogues.


A major shift in the treatment of type 2 diabetes has happened in the past decade. After the findings from the ACCORD, VADT, and ADVANCE trials, a more individualised2, 5, and 118 and patient-centred approach has become part of treatment strategies. This approach needs careful identification of individual glycaemic targets and, accordingly, an individual pharmacological strategy both at the time of initiation and during the process of intensification of insulin treatment.

Upon failure with oral antihyperglycaemic drugs the dilemma might be in the decision of whether basal insulin or a GLP-1 receptor agonists should be initiated. Although these two forms of treatment are both delivered via subcutaneous injection, GLP-1 receptor agonists might be preferred over insulin, with new longer formulations of these agonists providing further advantages in terms of frequency of injections needed. However, another option could just consist of early combination of the two drugs. Combination of insulin with SGLT2 inhibitors also has had promising results when initiated in patients with uncontrolled glucose on insulin treatment, although their benefit in delaying initiation of insulin treatment or their possible benefit above prandial insulin injections is yet to be investigated.

Several attempts have been made to develop novel insulin analogues, both basal and prandial. With the patent of old analogues soon to expire and the expected reduction in their costs, the new insulin analogues will need to show a clear clinical benefit to justify their use. Debate about the definition of cost-effectiveness of insulin has stemmed from the need to more clearly define clinically significant endpoints. Are HbA1c reduction and abatement of hypoglycaemia sufficient endpoints themselves, or should so-called softer endpoints (ie, quality of life, rate of compliance, and fear from hypoglycaemia) be used in determining cost-effectiveness of treatments?119, 120, and 121 Emerging clinical data should be carefully scrutinised to assess whether new insulin analogues truly represent a new era in insulin treatment and an improvement in the ability to control glycaemia, or whether their net benefit is minimal and these drugs would be better reserved for only patients whose clinical circumstances justify the excessive cost and less well established safety associated with their use.

Recent trials have indicated that, although insulin theoretically addresses β-cell failure by being the physiological replacement therapy, insulin treatment continues to raise concerns, which are mainly because of the chronic hyperinsulinaemia sustained by exogenous insulin administration in patients who are insulin resistant.17 and 19 Furthermore, associated weight gain and hypoglycaemia often leads clinicians to reconsider the justification of continued intensive insulin treatment. Reduction from insulin treatment could then be considered for patients who have acquired many comorbidities over the years, making a more lenient glycaemic target now appropriate. The role of prandial insulin in these patients who are more susceptible should be carefully thought of because its use increases the risk of hypoglycaemia, an event associated with increased risk of cardiovascular disease and all-cause mortality in people with type 2 diabetes treated with insulin. Moreover, apparent β-cell failure that could encourage introduction of short-acting insulin even in patients with long-term disease might be sustained by glucotoxicity rather than loss of functional β cells. In keeping with this practice, bariatric surgery is often followed by remission of diabetes, even in patients with long-term diabetes. Therefore, in a patient with progressive disease and in whom insulin treatment was initiated several years previously, addition of novel, non-insulin, glucose-lowering drugs or even reintroduction of drugs that were stopped because of poor efficacy, could be of benefit when given with insulin treatment if glucotoxicity is successfully abated. Once identified in such a patient, the individual glycaemic target, the possibility of replacing prandial insulin with another drug, and reduction or even stopping of basal insulin might be explored.

In conclusion, even after 90 years since insulin discovery, insulin treatment is still developing in terms of new formulations, novel routes of administration, and treatment strategies. These new developments might offer better opportunities to achieve and maintain individualised glycaemic targets, although the dilemma regarding the timing of insulin initiation remains debatable, mainly because randomised clinical trials with a sufficient duration of follow-up are absent in this area of research. Additionally, insulin intensification is a challenge for clinicians. Initiation and intensification of insulin treatment needs careful discussion and decisions with the patient. Perhaps, the comment made 85 years ago by Elliot P Joslin, the founder of modern diabetology, should still be regarded as one of the few cornerstones in the treatment of diabetes: “Insulin is a remedy primarily for the wise and not for the foolish, be they patients or doctors…. Everyone knows it requires brains to live long with diabetes, but to use insulin successfully requires more brains”. Indeed, we believe that use of insulin needs a good deal of communication between the doctor's and the patient's brains.

Search strategy and selection criteria

We searched PubMed and Google Scholar with the search terms “insulin therapy“, “type 2 diabetes mellitus”, and “T2DM“ mainly for articles published up to Dec 31, 2014, that focused on the insulin treatment of type 2 diabetes. We identified mainly full-text manuscripts written in English. We also searched for information about ongoing clinical trials in type 2 diabetes.

SDP, RM, and AD did the initial literature review. AC and SDP wrote the first draft of the manuscript. RM and AD provided critical review and redrafting of the text and figures, and assisted with additional literature review.

SDP has received honoraria for advisory work and lectures from AstraZeneca, Boehringer Ingelheim, Bristol-Myers Squibb, Eli Lilly, GlaxoSmithKline, Intarcia, Janssen, Merck Sharp and Dohme, Novartis, Novo Nordisk, Roche Diagnostics, Sanofi-Aventis, and Takeda. SDP has received research support from Bristol-Myers Squibb, Merck Sharp and Dohme, Novartis, and Novo Nordisk. AC has received honoraria for advisory work and lectures from AstraZeneca, Boehringer Ingelheim, Merck Sharp and Dohme, Novartis, Novo Nordisk, Sanofi Aventis, and TEVA. RM and AD declare no competing interests. SDP and RM were supported by grants from the Italian Ministry of Education and Research (SDP: PRIN grant 2010YK7Z5K_006, and RM: PRIN grant 2010JS3PMZ_002). No payment was made by any pharmaceutical company to support the writing of this article.


  • 1 SE Kahn, ME Cooper, S Del Prato. Pathophysiology and treatment of type 2 diabetes: perspectives on the past, present, and future. Lancet. 2014;383:1068-1083
  • 2 SE Inzucchi, RM Bergenstal, JB Buse, et al. Management of hyperglycaemia in type 2 diabetes, 2015: a patient-centred approach. Update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetologia. 2015;58:429-442
  • 3 AJ Garber, MJ Abrahamson, JI Barzilay, et al., the American Association of Clinical Endocrinologists. AACE comprehensive diabetes management algorithm 2013. Endocr Pract. 2013;19:327-336
  • 4 Canadian Diabetes Association 2013 Clinical Practice Guidelines Expert Committee, G Booth, AY Cheng. Canadian Diabetes Association 2013 clinical practice guidelines for the prevention and management of diabetes in Canada: methods. Can J Diabetes. 2013;37(suppl 1):S1-S212
  • 5 JS Skyler, R Bergenstal, RO Bonow, et al., the American Diabetes Association, the American College of Cardiology Foundation, the American Heart Association. Intensive glycemic control and the prevention of cardiovascular events: implications of the ACCORD, ADVANCE, and VA diabetes trials: a position statement of the American Diabetes Association and a scientific statement of the American College of Cardiology Foundation and the American Heart Association. Diabetes Care. 2009;32:187-192
  • 6 Y Ohkubo, H Kishikawa, E Araki, et al. Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective 6-year study. Diabetes Res Clin Pract. 1995;28:103-117
  • 7 UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet. 1998;352:837-853
  • 8 UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998;352:854-865
  • 9 S Nakar, G Yitzhaki, R Rosenberg, S Vinker. Transition to insulin in Type 2 diabetes: family physicians' misconception of patients' fears contributes to existing barriers. J Diabetes Complications. 2007;21:220-226
  • 10 SJ Griffin, K Borch-Johnsen, MJ Davies, et al. Effect of early intensive multifactorial therapy on 5-year cardiovascular outcomes in individuals with type 2 diabetes detected by screening (ADDITION-Europe): a cluster-randomised trial. Lancet. 2011;378:156-167
  • 11 C Weyer, C Bogardus, DM Mott, RE Pratley. The natural history of insulin secretory dysfunction and insulin resistance in the pathogenesis of type 2 diabetes mellitus. J Clin Invest. 1999;104:787-794
  • 12 A Wright, AC Burden, RB Paisey, CA Cull, RR Holman, the U.K. Prospective Diabetes Study Group. Sulfonylurea inadequacy: efficacy of addition of insulin over 6 years in patients with type 2 diabetes in the U.K. Prospective Diabetes Study (UKPDS 57). Diabetes Care. 2002;25:330-336
  • 13 A Ringborg, P Lindgren, DD Yin, M Martinell, J Stålhammar. Time to insulin treatment and factors associated with insulin prescription in Swedish patients with type 2 diabetes. Diabetes Metab. 2010;36:198-203
  • 14 P Home, NE Naggar, M Khamseh, et al. An observational non-interventional study of people with diabetes beginning or changed to insulin analogue therapy in non-Western countries: the A1chieve study. Diabetes Res Clin Pract. 2011;94:352-363
  • 15 L Litwak, S-Y Goh, Z Hussein, R Malek, V Prusty, ME Khamseh. Prevalence of diabetes complications in people with type 2 diabetes mellitus and its association with baseline characteristics in the multinational A1chieve study. Diabetol Metab Syndr. 2013;5:57
  • 16 P Home, M Riddle, WT Cefalu, et al. Insulin therapy in people with type 2 diabetes: opportunities and challenges?. Diabetes Care. 2014;37:1499-1508
  • 17 CJ Currie, JA Johnson. The safety profile of exogenous insulin in people with type 2 diabetes: justification for concern. Diabetes Obes Metab. 2012;14:1-4
  • 18 DS Lasserson, P Glasziou, R Perera, RR Holman, AJ Farmer. Optimal insulin regimens in type 2 diabetes mellitus: systematic review and meta-analyses. Diabetologia. 2009;52:1990-2000
  • 19 SE Holden, S Jenkins-Jones, CL Morgan, G Schernthaner, CJ Currie. Glucose-lowering with exogenous insulin monotherapy in type 2 diabetes: dose association with all-cause mortality, cardiovascular events and cancer. Diabetes Obes Metab. 2015;17:350-362
  • 20 HC Gerstein, J Bosch, GR Dagenais, et al., the ORIGIN Trial Investigators. Basal insulin and cardiovascular and other outcomes in dysglycemia. N Engl J Med. 2012;367:319-328
  • 21 RR Holman, SK Paul, MA Bethel, DR Matthews, HA Neil. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359:1577-1589
  • 22 HC Gerstein, ME Miller, RP Byington, et al., the Action to Control Cardiovascular Risk in Diabetes Study Group. Effects of intensive glucose lowering in type 2 diabetes. N Engl J Med. 2008;358:2545-2559
  • 23 W Duckworth, C Abraira, T Moritz, et al., the VADT Investigators. Glucose control and vascular complications in veterans with type 2 diabetes. N Engl J Med. 2009;360:129-139
  • 24 ML Shank, S Del Prato, RA DeFronzo. Bedtime insulin/daytime glipizide. Effective therapy for sulfonylurea failures in NIDDM. Diabetes. 1995;44:165-172
  • 25 P Rossetti, F Porcellati, CG Fanelli, G Perriello, E Torlone, GB Bolli. Superiority of insulin analogues versus human insulin in the treatment of diabetes mellitus. Arch Physiol Biochem. 2008;114:3-10
  • 26 I Jonassen, S Havelund, T Hoeg-Jensen, DB Steensgaard, PO Wahlund, U Ribel. Design of the novel protraction mechanism of insulin degludec, an ultra-long-acting basal insulin. Pharm Res. 2012;29:2104-2114
  • 27 E Nishimura, AR Sorensen, BF Hansen, et al. Insulin degludec, a new generation ultra-long acting basal insulin designed to maintain full metabolic effect while minimizing mitogenic potential. Diabetologia. 2010;53(suppl 1):S388-S389
  • 28 T Heise, L Nosek, SG Bøttcher, H Hastrup, H Haahr. Ultra-long-acting insulin degludec has a flat and stable glucose-lowering effect in type 2 diabetes. Diabetes Obes Metab. 2012;14:944-950
  • 29 RE Ratner, SC Gough, C Mathieu, et al. Hypoglycaemia risk with insulin degludec compared with insulin glargine in type 2 and type 1 diabetes: a pre-planned meta-analysis of phase 3 trials. Diabetes Obes Metab. 2013;15:175-184
  • 30 P Hollander, AB King, S Del Prato, et al. Insulin degludec improves long-term glycaemic control similarly to insulin glargine but with fewer hypoglycaemic episodes in patients with advanced type 2 diabetes on basal-bolus insulin therapy. Diabetes Obes Metab. 2014;17:202-206
  • 31 M Evans, M Wolden, J Gundgaard, B Chubb, T Christensen. Cost-effectiveness of insulin degludec compared with insulin glargine for patients with type 2 diabetes treated with basal insulin—from the UK health care cost perspective. Diabetes Obes Metab. 2014;16:366-375
  • 32 TM Caparrotta, M Evans. PEGylated insulin Lispro, (LY2605541)—a new basal insulin analogue. Diabetes Obes Metab. 2014;16:388-395
  • 33 J Rosenstock, RM Bergenstal, TC Blevins, et al. Better glycemic control and weight loss with the novel long-acting basal insulin LY2605541 compared with insulin glargine in type 1 diabetes: a randomized, crossover study. Diabetes Care. 2013;36:522-528
  • 34 US FDA. FDA Approval of Toujeo (U-300 insulin glargine). (2015) (accessed May 20, 2015).
  • 35 European Medicines Agency. Toujeo (previously Optisulin). (2015) (accessed May 20, 2015).
  • 36 RM Bergenstal, J Rosenstock, RF Arakaki, et al. A randomized, controlled study of once-daily LY2605541, a novel long-acting basal insulin, versus insulin glargine in basal insulin-treated patients with type 2 diabetes. Diabetes Care. 2012;35:2140-2147
  • 37 H Yki-Järvinen, R Bergenstal, M Ziemen, et al., the EDITION 2 Study Investigators. New insulin glargine 300 units/mL versus glargine 100 units/mL in people with type 2 diabetes using oral agents and basal insulin: glucose control and hypoglycemia in a 6-month randomized controlled trial (EDITION 2). Diabetes Care. 2014;37:3235-3243
  • 38 MC Riddle, GB Bolli, M Ziemen, I Muehlen-Bartmer, F Bizet, PD Home, the EDITION 1 Study Investigators. New insulin glargine 300 units/mL versus glargine 100 units/mL in people with type 2 diabetes using basal and mealtime insulin: glucose control and hypoglycemia in a 6-month randomized controlled trial (EDITION 1). Diabetes Care. 2014;37:2755-2762
  • 39 E Mannucci, M Monami, N Marchionni. Short-acting insulin analogues vs. regular human insulin in type 2 diabetes: a meta-analysis. Diabetes Obes Metab. 2009;11:53-59
  • 40 L Heinemann, L Nosek, F Flacke, et al. U-100, pH-Neutral formulation of VIAject® : faster onset of action than insulin lispro in patients with type 1 diabetes. Diabetes Obes Metab. 2012;14:222-227
  • 41 T Forst, A Pfützner, F Flacke, et al. Postprandial vascular effects of VIAject compared with insulin lispro and regular human insulin in patients with type 2 diabetes. Diabetes Care. 2010;33:116-120
  • 42 L Morrow, DB Muchmore, M Hompesch, EA Ludington, DE Vaughn. Comparative pharmacokinetics and insulin action for three rapid-acting insulin analogs injected subcutaneously with and without hyaluronidase. Diabetes Care. 2013;36:273-275
  • 43 M Hompesch, DB Muchmore, L Morrow, E Ludington, DE Vaughn. Improved postprandial glycemic control in patients with type 2 diabetes from subcutaneous injection of insulin lispro with hyaluronidase. Diabetes Technol Ther. 2012;14:218-224
  • 44 L Heinemann, DB Muchmore. Ultrafast-acting insulins: state of the art. J Diabetes Sci Technol. 2012;6:728-742
  • 45 Eli Lilly. Lilly and Adocia announce alliance to co-develop ultra-rapid insulin based on BioChaperone technology. (2014) (accessed Dec 20, 2014).
  • 46 Adocia. Adocia reports positive preliminary results from dose response clinical study of ultra-fast acting biochaperone lispro U100 in patients with type 1 diabetes. (2014) (accessed Feb 28, 2015).
  • 47 Adocia. BioChaperone Lispro U300: concentrated ultra-fast prandial insulin. (accessed Feb 28, 2015).
  • 48 E Cengiz, SA Weinzimer, JL Sherr, et al. Faster in and faster out: accelerating insulin absorption and action by insulin infusion site warming. Diabetes Technol Ther. 2014;16:20-25
  • 49 E McVey, L Hirsch, DE Sutter, et al. Pharmacokinetics and postprandial glycemic excursions following insulin lispro delivered by intradermal microneedle or subcutaneous infusion. J Diabetes Sci Tech. 2012;6:743-754
  • 50 T Santos Cavaiola, S Edelman. Inhaled insulin: a breath of fresh air? A review of inhaled insulin. Clin Ther. 2014;36:1275-1289
  • 51 W Nuffer, JM Trujillo, SL Ellis. Technosphere insulin (Afrezza): a new, inhaled prandial insulin. Ann Pharmacother. 2015;49:99-106
  • 52 H Zisser, L Jovanovic, K Markova, et al. Technosphere insulin effectively controls postprandial glycemia in patients with type 2 diabetes mellitus. Diabetes Technol Ther. 2012;14:997-1001
  • 53 M Peyrot, RR Rubin. Patient-reported outcomes in adults with type 2 diabetes using mealtime inhaled technosphere insulin and basal insulin versus premixed insulin. Diabetes Technol Ther. 2011;13:1201-1206
  • 54 S Del Prato, L Blonde, L Martinez, et al., the EXPERIENCE Trial Team. The effect of the availability of inhaled insulin on glycaemic control in patients with Type 2 diabetes failing on oral therapy: the evaluation of Exubera as a therapeutic option on insulin initiation and improvement in glycaemic control in clinical practice (EXPERIENCE) trial. Diabet Med. 2008;25:662-670
  • 55 T Strack, L Martinez, S Del Prato, et al. Assessing the impact of a new delivery method of insulin on glycemic control using a novel trial design. Diabetes Metab Syndr Obes. 2009;2:1-10
  • 56 P Fonte, F Araújo, S Reis, B Sarmento. Oral insulin delivery: how far are we?. J Diabetes Sci Tech. 2013;7:520-531
  • 57 M Kanzarkar, PP Pathak, M Vaidya, C Brumlik, A Choudhury. Oral insulin-delivery system for diabetes mellitus. Pharm Pat Anal. 2015;4:29-36
  • 58 C Kapitza, E Zijlstra, L Heinemann, MC Castelli, G Riley, T Heise. Oral insulin: a comparison with subcutaneous regular human insulin in patients with type 2 diabetes. Diabetes Care. 2010;33:1288-1290
  • 59 SD Luzio, G Dunseath, A Lockett, TP Broke-Smith, RR New, DR Owens. The glucose lowering effect of an oral insulin (Capsulin) during an isoglycaemic clamp study in persons with type 2 diabetes. Diabetes Obes Metab. 2010;12:82-87
  • 60 L Heinemann. New ways of insulin delivery. Int J Clin Pract Suppl. 2012;66:35-39
  • 61 S Cernea, M Kidron, J Wohlgelernter, P Modi, I Raz. Dose-response relationship of oral insulin spray in healthy subjects. Diabetes Care. 2005;28:1353-1357
  • 62 EE Engwerda, EJ Abbink, CJ Tack, BE de Galan. Improved pharmacokinetic and pharmacodynamic profile of rapid-acting insulin using needle-free jet injection technology. Diabetes Care. 2011;34:1804-1808
  • 63 Z Gu, TT Dang, M Ma, et al. Glucose-responsive microgels integrated with enzyme nanocapsules for closed-loop insulin delivery. ACS Nano. 2013;7:6758-6766
  • 64 CJ Currie, CD Poole, M Evans, JR Peters, CL Morgan. Mortality and other important diabetes-related outcomes with insulin vs other antihyperglycemic therapies in type 2 diabetes. J Clin Endocrinol Metab. 2013;98:668-677
  • 65 CJ Currie, JR Peters, A Tynan, et al. Survival as a function of HbA(1c) in people with type 2 diabetes: a retrospective cohort study. Lancet. 2010;375:481-489
  • 66 LG Mellbin, K Malmberg, A Norhammar, H Wedel, L Rydén, the DIGAMI 2 Investigators. Prognostic implications of glucose-lowering treatment in patients with acute myocardial infarction and diabetes: experiences from an extended follow-up of the Diabetes Mellitus Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI) 2 Study. Diabetologia. 2011;54:1308-1317
  • 67 DJ Margolis, O Hoffstad, BL Strom. Association between serious ischemic cardiac outcomes and medications used to treat diabetes. Pharmacoepidemiol Drug Saf. 2008;17:753-759
  • 68 B Carstensen, DR Witte, S Friis. Cancer occurrence in Danish diabetic patients: duration and insulin effects. Diabetologia. 2012;55:948-958
  • 69 S Hernández-Díaz, HO Adami. Diabetes therapy and cancer risk: causal effects and other plausible explanations. Diabetologia. 2010;53:802-808
  • 70 JM Gamble, SH Simpson, DT Eurich, SR Majumdar, JA Johnson. Insulin use and increased risk of mortality in type 2 diabetes: a cohort study. Diabetes Obes Metab. 2010;12:47-53
  • 71 CJ Currie, SE Holden. Optimizing clinical outcomes resulting from glucose-lowering therapies in type 2 diabetes: increased confidence about the DPP-4 inhibitors and continued concerns regarding sulphonylureas and exogenous insulin. Diabetes Obes Metab. 2014;16:881-884
  • 72 LS Hermann, J Kalén, P Katzman, et al. Long-term glycaemic improvement after addition of metformin to insulin in insulin-treated obese type 2 diabetes patients. Diabetes Obes Metab. 2001;3:428-434
  • 73 MG Wulffelé, A Kooy, P Lehert, et al. Combination of insulin and metformin in the treatment of type 2 diabetes. Diabetes Care. 2002;25:2133-2140
  • 74 IF Douek, SE Allen, P Ewings, EA Gale, PJ Bingley, the Metformin Trial Group. Continuing metformin when starting insulin in patients with Type 2 diabetes: a double-blind randomized placebo-controlled trial. Diabet Med. 2005;22:634-640
  • 75 R Boussageon, I Supper, T Bejan-Angoulvant, et al. Reappraisal of metformin efficacy in the treatment of type 2 diabetes: a meta-analysis of randomised controlled trials. PLoS Med. 2012;9:e1001204
  • 76 B Charbonnel, R DeFronzo, J Davidson, et al., the PROactive investigators. Pioglitazone use in combination with insulin in the prospective pioglitazone clinical trial in macrovascular events study (PROactive19). J Clin Endocrinol Metab. 2010;95:2163-2171
  • 77 J Kung, RR Henry. Thiazolidinedione safety. Expert Opin Drug Saf. 2012;11:565-579
  • 78 VR Aroda, RR Henry, J Han, et al. Efficacy of GLP-1 receptor agonists and DPP-4 inhibitors: meta-analysis and systematic review. Clin Ther. 2012;34:1247-1258 e22.
  • 79 R Balena, IE Hensley, S Miller, AH Barnett. Combination therapy with GLP-1 receptor agonists and basal insulin: a systematic review of the literature. Diabetes Obes Metab. 2013;15:485-502
  • 80 V Fonseca, A Schweizer, D Albrecht, MA Baron, I Chang, S Dejager. Addition of vildagliptin to insulin improves glycaemic control in type 2 diabetes. Diabetologia. 2007;50:1148-1155
  • 81 V Lukashevich, A Schweizer, JE Foley, S Dickinson, PH Groop, W Kothny. Efficacy of vildagliptin in combination with insulin in patients with type 2 diabetes and severe renal impairment. Vasc Health Risk Manag. 2013;9:21-28
  • 82 W Kothny, J Foley, P Kozlovski, Q Shao, B Gallwitz, V Lukashevich. Improved glycaemic control with vildagliptin added to insulin, with or without metformin, in patients with type 2 diabetes mellitus. Diabetes Obes Metab. 2013;15:252-257
  • 83 P Kozlovski, J Foley, Q Shao, V Lukashevich, W Kothny. Vildagliptin-insulin combination improves glycemic control in Asians with type 2 diabetes. World J Diabetes. 2013;4:151-156
  • 84 AH Barnett, B Charbonnel, M Donovan, D Fleming, R Chen. Effect of saxagliptin as add-on therapy in patients with poorly controlled type 2 diabetes on insulin alone or insulin combined with metformin. Curr Med Res Opin. 2012;28:513-523
  • 85 ES Hong, AR Khang, JW Yoon, et al. Comparison between sitagliptin as add-on therapy to insulin and insulin dose-increase therapy in uncontrolled Korean type 2 diabetes: CSI study. Diabetes Obes Metab. 2012;14:795-802
  • 86 S Arnolds, S Dellweg, J Clair, et al. Further improvement in postprandial glucose control with addition of exenatide or sitagliptin to combination therapy with insulin glargine and metformin: a proof-of-concept study. Diabetes Care. 2010;33:1509-1515
  • 87 T Vilsbøll, J Rosenstock, H Yki-Järvinen, et al. Efficacy and safety of sitagliptin when added to insulin therapy in patients with type 2 diabetes. Diabetes Obes Metab. 2010;12:167-177
  • 88 H Yki-Järvinen, J Rosenstock, S Durán-Garcia, et al. Effects of adding linagliptin to basal insulin regimen for inadequately controlled type 2 diabetes: a ≥52-week randomized, double-blind study. Diabetes Care. 2013;36:3875-3881
  • 89 JJ Holst, T Vilsbøll. Combining GLP-1 receptor agonists with insulin: therapeutic rationales and clinical findings. Diabetes Obes Metab. 2013;15:3-14
  • 91 Y Seino, KW Min, E Niemoeller, A Takami, the EFC10887 GETGOAL-L Asia Study Investigators. Randomized, double-blind, placebo-controlled trial of the once-daily GLP-1 receptor agonist lixisenatide in Asian patients with type 2 diabetes insufficiently controlled on basal insulin with or without a sulfonylurea (GetGoal-L-Asia). Diabetes Obes Metab. 2012;14:910-917
  • 92 MC Riddle, R Aronson, P Home, et al. Adding once-daily lixisenatide for type 2 diabetes inadequately controlled by established basal insulin: a 24-week, randomized, placebo-controlled comparison (GetGoal-L). Diabetes Care. 2013;36:2489-2496
  • 93 MC Riddle, T Forst, R Aronson, et al. Adding once-daily lixisenatide for type 2 diabetes inadequately controlled with newly initiated and continuously titrated basal insulin glargine: a 24-week, randomized, placebo-controlled study (GetGoal-Duo 1). Diabetes Care. 2013;36:2497-2503
  • 94 MC Bunck, A Cornér, B Eliasson, et al. Effects of exenatide on measures of β-cell function after 3 years in metformin-treated patients with type 2 diabetes. Diabetes Care. 2011;34:2041-2047
  • 95 M Diamant, L Van Gaal, B Guerci, et al. Exenatide once weekly versus insulin glargine for type 2 diabetes (DURATION-3): 3-year results of an open-label randomised trial. Lancet Diabetes Endocrinol. 2014;2:464-473
  • 96 D Russell-Jones, A Vaag, O Schmitz, et al., the Liraglutide Effect and Action in Diabetes 5 (LEAD-5) met+SU Study Group. Liraglutide vs insulin glargine and placebo in combination with metformin and sulfonylurea therapy in type 2 diabetes mellitus (LEAD-5 met+SU): a randomised controlled trial. Diabetologia. 2009;52:2046-2055
  • 97 DA Scott, KS Boye, L Timlin, JF Clark, JH Best. A network meta-analysis to compare glycaemic control in patients with type 2 diabetes treated with exenatide once weekly or liraglutide once daily in comparison with insulin glargine, exenatide twice daily or placebo. Diabetes Obes Metab. 2013;15:213-223
  • 98 SCL Gough, B Bode, V Woo, et al., on behalf of the NN9068-3697 (DUAL-I) trial investigators. Efficacy and safety of a fixed-ratio combination of insulin degludec and liraglutide (IDegLira) compared with its components given alone: results of a phase 3, open-label, randomised, 26-week, treat-to-target trial in insulin-naive patients with type 2 diabetes. Lancet Diabetes Endocrinol. 2014;2:885-893
  • 99 JB Buse, T Vilsbøll, J Thurman, et al., the NN9068-3912 (DUAL-II) Trial Investigators. Contribution of liraglutide in the fixed-ratio combination of insulin degludec and liraglutide (IDegLira). Diabetes Care. 2014;37:2926-2933
  • 100 E Ferrannini, A Solini. SGLT2 inhibition in diabetes mellitus: rationale and clinical prospects. Nat Rev Endocrinol. 2012;8:495-502
  • 101 JP Wilding, V Woo, K Rohwedder, J Sugg, S Parikh, the Dapagliflozin 006 Study Group. Dapagliflozin in patients with type 2 diabetes receiving high doses of insulin: efficacy and safety over 2 years. Diabetes Obes Metab. 2014;16:124-136
  • 102 J Rosenstock, A Jelaska, G Frappin, et al., the EMPA-REG MDI Trial Investigators. Improved glucose control with weight loss, lower insulin doses, and no increased hypoglycemia with empagliflozin added to titrated multiple daily injections of insulin in obese inadequately controlled type 2 diabetes. Diabetes Care. 2014;37:1815-1823
  • 103 J Rosenstock, M Davies, R Dumas, M Desai, M Alba, G Meininger. Effects of Canagliflozin added on to basal insulin with or without other antihyperglycemic agents in type 2 diabetes. Diabetes. 2013;62(suppl 1):A280
  • 104 National Institute of Health and Care Excellence. Diabetes in adults quality standard: NICE quality standard [QS6]. (2011) (accessed Dec 12, 2014).
  • 105 RR Holman, KI Thorne, AJ Farmer, et al., the 4-T Study Group. Addition of biphasic, prandial, or basal insulin to oral therapy in type 2 diabetes. N Engl J Med. 2007;357:1716 1713
  • 106 MC Riddle, J Rosenstock, J Gerich, the Insulin Glargine 4002 Study Investigators. The treat-to-target trial: randomized addition of glargine or human NPH insulin to oral therapy of type 2 diabetic patients. Diabetes Care. 2003;26:3080-3086
  • 107 RR Holman, AJ Farmer, MJ Davies, et al., the 4-T Study Group. Three-year efficacy of complex insulin regimens in type 2 diabetes. N Engl J Med. 2009;361:1736-1747
  • 108 HW Rodbard, B Cariou, B Zinman, et al., the BEGIN Once Long trial investigators. Comparison of insulin degludec with insulin glargine in insulin-naive subjects with Type 2 diabetes: a 2-year randomized, treat-to-target trial. Diabet Med. 2013;30:1298-1304
  • 109 Y Onishi, Y Iwamoto, SJ Yoo, P Clauson, SC Tamer, S Park. Insulin degludec compared with insulin glargine in insulin-naïve patients with type 2 diabetes: A 26-week, randomized, controlled, Pan-Asian, treat-to-target trial. J Diabetes Investig. 2013;4:605-612
  • 110 MC Riddle, J Rosenstock, A Vlajnic, L Gao. Randomized, 1-year comparison of three ways to initiate and advance insulin for type 2 diabetes: twice-daily premixed insulin versus basal insulin with either basal-plus one prandial insulin or basal-bolus up to three prandial injections. Diabetes Obes Metab. 2014;16:396-402
  • 111 R Malek, F Ajili, SH Assaad-Khalil, A Shinde, JW Chen, E Van den Berg. Similar glucose control with basal-bolus regimen of insulin detemir plus insulin aspart and thrice-daily biphasic insulin aspart 30 in insulin-naive patients with type 2 diabetes: results of a 50-week randomized clinical trial of stepwise insulin intensification. Diabetes Metab. 2014; 10.1016/j.diabet.2014.11.002 published online Dec 4.
  • 112 K Bowering, VA Reed, JS Felicio, J Landry, L Ji, J Oliveira. A study comparing insulin lispro mix 25 with glargine plus lispro therapy in patients with type 2 diabetes who have inadequate glycaemic control on oral anti-hyperglycaemic medication: results of the PARADIGM study. Diabet Med. 2012;29:e263-e272
  • 113 S Kaneko, F Chow, DS Choi, et al., on behalf of BOOST: Intensify All Trial Investigators. Insulin degludec/insulin aspart versus biphasic insulin aspart 30 in Asian patients with type 2 diabetes inadequately controlled on basal or pre-/self-mixed insulin: a 26-week, randomised, treat-to-target trial. Diabetes Res Clin Pract. 2015;1:139-147
  • 114 GR Fulcher, JS Christiansen, G Bantwal, et al., the BOOST: Intensify Premix I Investigators. Comparison of insulin degludec/insulin aspart and biphasic insulin aspart 30 in uncontrolled, insulin-treated type 2 diabetes: a phase 3a, randomized, treat-to-target trial. Diabetes Care. 2014;37:2084-2090
  • 115 DR Owens. Stepwise intensification of insulin therapy in type 2 diabetes management—exploring the concept of the basal-plus approach in clinical practice. Diabet Med. 2013;30:276-288
  • 116 Y Reznik, O Cohen, R Aronson, et al., for the OpT2mise Study Group. Insulin pump treatment compared with multiple daily injections for treatment of type 2 diabetes (OpT2mise): a randomised open-label controlled trial. Lancet. 2014;384:1265-1272
  • 117 A Patel, S MacMahon, J Chalmers, et al., the ADVANCE Collaborative Group. Intensive blood glucose control and vascular outcomes in patients with type 2 diabetes. N Engl J Med. 2008;358:2560-2572
  • 118 I Raz, MC Riddle, J Rosenstock, et al. Personalized management of hyperglycemia in type 2 diabetes: reflections from a Diabetes Care Editors' Expert Forum. Diabetes Care. 2013;36:1779-1788
  • 119 N Waugh, E Cummins, P Royle, et al. Newer agents for blood glucose control in type 2 diabetes: systematic review and economic evaluation. Health Technol Assess. 2010;14:1-248
  • 120 G Grunberger. Insulin analogs-are they worth it? Yes!. Diabetes Care. 2014;37:1767-1770
  • 121 MB Davidson. Insulin analogs—is there a compelling case to use them? No!. Diabetes Care. 2014;37:1771-1774
  • 122 KJ Chang-Chen, R Mullur, E Bernal-Mizrachi. Beta-cell failure as a complication of diabetes. Rev Endocr Metab Disord. 2008;9:329-343
  • 123 K Khunti, M Davies, A Majeed, BL Thorsted, ML Wolden, SK Paul. Hypoglycemia and risk of cardiovascular disease and all-cause mortality in insulin-treated people with type 1 and type 2 diabetes: a cohort study. Diabetes Care. 2015;38:316-322
  • 124 BP Müller-Stich, JD Senft, R Warschkow, et al. Surgical versus medical treatment of type 2 diabetes mellitus in nonseverely obese patients: a systematic review and meta-analysis. Ann Surg. 2015;261:421-429
  • 125 EP Joslin. The treatment of diabetes mellitus 4th edn. (Lea & Febiger, Philadelphia, 1928)



a Diabetes Unit, Department of Internal Medicine, Hadassah Hebrew University Hospital, Jerusalem, Israel

b Department of Clinical and Experimental Medicine, Section of Diabetes and Metabolic Diseases, University of Pisa, Pisa, Italy

* Correspondence to: Prof Stefano Del Prato, Department of Clinical and Experimental Medicine, Section of Diabetes and Metabolic Diseases, University of Pisa, 56124 Pisa, Italy