Rationale for a titratable fixed-ratio co-formulation of a basal insulin analog and a glucagon-like peptide 1 receptor agonist in patients with type 2 diabetes

Abstract

Objective: Achieving and maintaining recommended glycemic targets, including those for glycated hemoglobin A1c (A1C), is key to improving outcomes in patients with type 2 diabetes (T2D). As fasting plasma glucose and postprandial glucose contribute to overall A1C, targeting both is essential for sustaining glycemic control.

Methods: This review examines the complementary mechanisms of action of glucagon-like peptide 1 (GLP-1) receptor agonists and basal insulin; they both enhance glucose-stimulated insulin release and suppress glucagon secretion. GLP-1 receptor agonists also slow gastric emptying and increase satiety.

Results: Adding a GLP-1 receptor agonist to therapy with a basal insulin analog has been associated with improved overall glycemic control, with comparable risk of hypoglycemia and no weight gain. Titratable fixed-ratio co-formulations of basal insulin and a GLP-1 receptor agonist have been shown to improve glycemic control, with less complex dosing schedules, possibly increasing treatment adherence. The slow titration of fixed-ratio co-formulations has been shown to reduce the occurrence and severity of gastrointestinal adverse events associated with the use of a separate GLP-1 receptor agonist. Titratable fixed-ratio co-formulations also mitigate insulin-associated weight gain, and show a comparable risk of hypoglycemia to basal insulin use alone.

Conclusions: The efficacy and safety of titratable fixed-ratio co-formulations have been demonstrated for insulin degludec/liraglutide and insulin glargine/lixisenatide in the DUAL and LixiLan trials, respectively, in both insulin-naive and -experienced patients. Titratable fixed-ratio co-formulations represent an attractive treatment option for many patients with T2D.

Keywords:

• Glucagon-like peptide 1
• Insulin degludec
• Insulin glargine
• Liraglutide
• Lixisenatide

Introduction

Type 2 diabetes (T2D) is a chronic metabolic disease characterized by an insulin secretory defect in pancreatic β-cells that, in most individuals, is combined with insulin resistance in peripheral tissues, such as skeletal muscle, fat, and liver. This is clinically manifested by increases in fasting plasma glucose (FPG) and postprandial glucose (PPG) levels, both of which independently contribute to an overall increase in glycated hemoglobin (A1C)^1,2.

Achieving glycemic control is central to preventing diabetes-associated microvascular and macrovascular complications in those with T2D³. As their disease progresses and β-cell function declines, most patients with T2D will require treatment escalation from initial metformin monotherapy to combinations of metformin with other oral anti-hyperglycemic drugs (OADs) and/or injectable therapies—commonly basal insulin and/or, more recently, a glucagon-like peptide 1 (GLP-1) receptor agonist³.

Basal insulin analogs are intended to mimic the basal component of normal physiological endogenous insulin secretion by pancreatic β-cells, and are primarily effective in controlling A1C via reductions in FPG levels. However, clinical trials and real-world studies have demonstrated that many patients with T2D do not reach (or fail to maintain) glycemic targets with basal insulin alone, and are in need of therapeutic escalation^4,5. This may be related to a number of factors, including clinical inertia⁶, a plateau effect on FPG with basal insulin^7,8, and the contribution of PPG levels to A1C^1,2.

Treatment guidelines recommend intensification of therapy for patients on basal insulin when A1C remains above target, despite acceptable FPG levels (suggesting PPG is above goal) or where basal insulin has been titrated to >0.5 units (U)/kg/day and FPG is still not at goal^3,9,10. Under this circumstance, the American Diabetes Association (ADA) recommends consideration of initiation of combination injectable therapy³. GLP-1 receptor agonists, short- or rapid-acting insulin preparations, dipeptidyl peptide-4 (DPP-4) inhibitors, and sodium-glucose co-transporter type 2 inhibitors are among the recommended strategies to control PPG levels^3,10. However, patients and clinicians are sometimes reluctant to escalate therapy because of concerns which may include weight gain, hypoglycemia, and tolerability issues.

Strategies which can improve glycemic control while minimizing adverse events have clear advantages. The combination of a basal insulin analog and a GLP-1 receptor agonist has theoretical advantages based on the complementary mechanisms of action of the two agents affecting both FPG (predominantly controlled by basal insulins) and PPG (predominantly controlled by GLP-1 receptor agonists). In addition, the combination has potential to reduce hypoglycemia and weight gain that can be associated with basal insulin therapy alone^11–18. The aim of this review is to explore the rationale and benefits of using a once-daily, titratable fixed-ratio co-formulation (combination) of a basal insulin analog and a GLP-1 receptor agonist for the management of T2D.

The importance of adherence in the treatment of those with T2D

Adherence to therapy is critical for achieving optimal glycemic control in patients with T2D. However, it is widely recognized that there are a range of patient- and treatment-related factors that can negatively impact adherence (Table 1)^19–21. For patients treated with insulin-based regimens, the most commonly cited concerns are hypoglycemia and weight gain²². Other adverse events associated with additional agents may also raise concerns and/or impact on adherence; for example, GLP-1 receptor agonists are associated with an increased risk of gastrointestinal adverse events²³, while the addition of some DPP-4 inhibitors might infrequently be associated with hypersensitivity, skin-related reactions, and joint pain^24,25. The increased number of daily injections and increasing regimen complexity associated with intensification of basal insulin therapy with the addition of prandial insulin (basal-plus and basal-bolus regimens) or a GLP-1 receptor agonist, may further erode adherence (Table 1). For many patients, the need to inject has a significant impact on multiple aspects of their daily life²⁶. Being “too busy” to inject, concern about injection discomfort, having multiple daily injections, interference with daily activities, and embarrassment of injecting in public have been given as reasons for non-adherence and insulin-injection omission^22,27,28. Not unexpectedly, most insulin-treated patients want to reduce the number of injections they administer each day²⁸. Furthermore, having fewer injections has been shown to have a positive impact on health-related quality-of-life²⁹. Reducing regimen complexity/injection frequency by the use of titratable fixed-ratio co-formulations has the potential to improve adherence and patient satisfaction with treatment^21,22,30.

Table 1. Factors affecting adherence among T2D patients^19–21.

Factors reducing adherence	Factors improving adherence
Polypharmacy, complex regimen, injection medications, increasing number of injections	Reduced complexity of regimen, reduced frequency of administration
Adverse events (e.g. weight gain, hypoglycemia, nausea, vomiting, diarrhea, headaches, water retention, cardiovascular events)	Newer medications that enable patients to achieve glycemic targets while reducing adverse events
Perceived efficacy and safety (both patients and physicians)	Education and increased knowledge. Improve patients’ perception of their personal ability to influence their diabetes
Poor patient–physician relationship	Improved continuity of care. Diabetes nurses at facilities. Improved communication between patients and physician. Patients feeling more informed about treatment options and decisions
Economic considerations	Benefits outweigh costs. Financial incentives. Reducing co-payments and limiting the payments patients are required to make
Self-monitoring of blood glucose

Rationale for therapy combining a basal insulin analog and a GLP-1 receptor agonist

The scientific rationale for combined therapy with a basal insulin analog and a GLP-1 receptor agonist lies in the complementary mechanisms of action of these two classes of glucose-lowering agents (Table 2)^31,32. Basal insulin analogs are intended to mimic the normal pattern of physiological basal insulin secretion by providing consistent, long-acting insulin levels. The main role of basal insulin is, therefore, to reduce FPG levels by preventing lipolysis and hepatic glucose production in the fasting state, especially during the night³³.

Table 2. The complementary mechanisms of action of basal insulin and a GLP-1 receptor agonist.

	Basal insulin	GLP-1 receptor agonist
Primary effects	↓ Mainly FPG ↓ Interprandial glucose	↓ PPG^a ↓ FPG^b
Mechanism	↓ Hepatic glucose production ↑ Non-glucose-dependent endogenous insulin ↓ Glucagon secretion ↑ Insulin concentration	↑ Glucose-dependent insulin secretion^b ↓ Glucose-dependent glucagon secretion ↓ Gastric emptying rate^a ↑ Satiety ↓ Food intake
Effect on weight	↑ Body weight	↓ Body weight

Reproduced and adapted with permission from Balena et al.³¹

^a short-acting formulations.

^b long-acting formulations.Abbreviations: FPG, fasting plasma glucose; GLP-1, glucagon-like peptide 1; PPG, postprandial glucose.

The effects of GLP-1 receptor agonists are glucose-dependent, acting to reduce blood glucose levels through the enhancement of glucose-stimulated insulin secretion by β-cells and suppression of glucagon secretion by α-cells (Figure 1). GLP-1 receptor agonists can be classified as either short- or long-acting based on their duration of action. Although both classes share some effects, there are also differences (Figure 1). In addition to their effects on the central nervous system and glucose dependent increase in insulin and decrease in glucagon secretion, short-acting GLP-1 receptor agonists (such as exenatide and lixisenatide) also delay gastric emptying as their primary mechanism of action³⁴. This limits the rate and extent of meal-derived glucose presentation to β-cells, contributing to a predominant effect on postprandial glucose excursions, most markedly for the meal following administration³⁴. Long-acting GLP-1 receptor agonists (such as semaglutide, albiglutide, dulaglutide, exenatide extended release, and liraglutide) predominantly lower blood glucose levels through stimulation of insulin secretion and reduction of glucagon levels, with a decreasing effect on gastric emptying over time likely due to tachyphylaxis^34–36. This results in a lesser effect on PPG excursions compared with short-acting GLP-1 receptor agonists but a longer overall effect on glucose levels reducing FPG³⁴. GLP-1 receptor agonists have also been shown to improve β-cell function^37,38. The action of GLP-1 receptor agonists on the central nervous system results in increased satiety, which contributes to a reduced food intake³⁹ and helps mitigate the weight gain often associated with insulin therapy.

Figure 1. Mechanisms of GLP-1 and GLP-1 receptor agonist action. GLP-1, glucagon-like peptide 1. Reproduced with permission from Meier 2012³⁴.

The combination of a basal insulin analog to primarily target FPG and a GLP-1 receptor agonist to benefit PPG control has, therefore, the potential to improve overall glycemic control and to provide relatively stable blood glucose levels. The combination also addresses seven of the eight pathophysiologic abnormalities contributing to hyperglycemia in T2D patients and described as the “ominous octet”⁴⁰. In clinical studies, adding a GLP-1 receptor agonist to basal insulin led to comparable glycemic control as adding prandial insulin, with lower risk for hypoglycemia, and weight loss^13,41,42. Although a greater proportion of patients on GLP-1 receptor agonists experienced gastrointestinal adverse events than those on prandial insulin, treatment with GLP-1 receptor agonists was associated with greater treatment satisfaction and better quality-of-life⁴¹. Overall, systematic reviews and meta-analyses of observational/clinical practice studies and clinical trials have found that combining a basal insulin analog and a GLP-1 receptor agonist improves glycemic control without weight gain or an increased risk of hypoglycemia^31,43.

Titratable fixed-ratio co-formulation of a basal insulin analog and a GLP-1 receptor agonist: benefits for titration, tolerability, adherence, and persistence

Given the potential effects of complex regimens on adherence to therapy, the ability to co-administer a basal insulin analog and a GLP-1 receptor agonist as a titratable fixed-ratio co-formulation supplied in a single injection device has several advantages. In addition to the potential efficacy benefits resulting from their complementary mechanisms of action, such a co-formulations may provide additional benefits over other combination strategies. Compared with basal-bolus insulin therapy, a titratable fixed-ratio co-formulation allows a less complex and more convenient dosing schedule, with fewer insulin injections and self-monitored blood glucose (SMBG) measurements, as only pre-breakfast FPG levels are required for titration, compared with multiple daily measurements for patients on intensive insulin regimens. Titration is also easier, as one dose adjustment is required rather than two or more separate adjustments. There is also evidence that fixed ratio co-formulations may be insulin sparing compared with insulin alone. In clinical studies where insulin dose was not capped, final insulin doses were lower in the combination treatment group, while providing superior glycemic control (Table 3)^11,13,15,47. Similarly, final GLP-1 receptor agonist dose in these studies was also lower with the co-formulation compared with the agent alone, due to the more gradual, incremental titration possible with the fixed-ratio co-formulations^11,13. Use of a titratable fixed ratio co-formulation could provide an option when changing from existing regimens with basal insulin or with a GLP-1 receptor agonist. Although it is currently off-label in the US, a titratable fixed-ratio co-formulation could also provide an “easy-start” option for initiating basal insulin therapy when patients on OADs are unable to reach glycemic target. In keeping with studies using separate injections, titratable fixed-ratio co-formulations have been shown to improve glycemic control with no increased risk of hypoglycemia, and to mitigate insulin-associated weight gain^{11–13,15–18}. The slow titration of a fixed-ratio co-formulation has been shown to increase the tolerability of GLP-1 receptor agonists compared with their use individually, reducing the occurrence and severity of gastrointestinal adverse events (e.g. nausea, vomiting, and diarrhea) which often lead to decreased adherence or persistence with treatment^{11–13,15–18}. Fewer adverse events and reduced risk for insulin-associated weight gain or even leading to modest weight loss may improve patient adherence, which can lead to early and sustained achievement of glycemic control.

Table 3. Summary of key clinical trials of IDegLira and iGlarLixi, including main efficacy, safety, and tolerability findings.

Clinical trial	Study	Results		Conclusions
		Final study drug doses (mean)	Efficacy, safety and tolerability
LixiLan phase 2b NCT01476475^16	24-week RCT (n = 323)^a Mean baseline A1C: 8.0% Background treatment: metformin Dosage: once-daily iGlarLixi + metformin or iGlar + metformin	Insulin: iGlarLixi group 36 units vs iGlar group 39 units	A1C change: −1.82% vs −1.64% (iGlarLixi vs iGlar). Estimated treatment difference between iGlarLixi and iGlar: −0.17% [95% CI = −0.312, −0.037; p = 0.01] Mean weight change: −1.16 kg vs +0.39 kg (iGlarLixi vs iGlar, respectively; p < 0.0001) Incidence of symptomatic hypoglycemia: ∼ 25% in both groups	iGlarLixi led to greater A1C reduction with weight reduction and no increased risk of hypoglycemia compared with iGlar alone.
LixiLan-O phase 3 NCT02058147^17	30-week RCT (n = 1170) Mean baseline A1C: ∼ 8.1% Background treatment: metformin Dosage: once-daily iGlarLixi + metformin, iGlar + metformin or 20 μg lixisenatide + metformin Ratio iGlar/lixisenatide: 2:1 (1 U iGlar/0.5 µg lixisenatide) for doses between 10 U/5 µg and 40 U/20 µg, or 3:1 (1 U iGlar/0.33 µg lixisenatide) for doses between 30 U/10 µg and 60 U/20 µg^b Maximum dose: 60 U iGlar for iGlarLixi and iGlar alone	Insulin: iGlarLixi group 39.8 units vs iGlar group 40.3 units	A1C change: −1.6%, −1.3%, −0.9% (iGlarLixi, iGlar, lixisenatide).Estimated treatment difference between iGlarLixi and iGlar: −0.3% [95% CI = −0.4, −0.2; p < 0.0001] Mean weight change: −0.3 kg, +1.1 kg, −2.3 kg (iGlarLixi, iGlar, lixisenatide). Estimated treatment difference between iGlarLixi and iGlar: −1.4 kg [95% CI = −1.9, −0.9; p < 0.0001] Incidence of symptomatic hypoglycemia (plasma glucose ≤70 mg/dL): 25.6%, 23.6%, and 6.4% (iGlarLixi, iGlar, and lixisenatide, respectively) Overall adverse events were similar; incidence of nausea: 9.6%, 3.6%, and 24.0%; incidence of vomiting: 3.2%, 1.5%, and 6.4%^c (iGlarLixi, iGlar, and lixisenatide, respectively)	iGlarLixi when added to metformin, significantly improved glycemic control compared with each of its components alone, and was associated with weight loss and no increased risk of hypoglycemia compared with iGlar.
LixiLan-L phase 3 NCT02058160^18	30-week RCT (n = 736) Mean baseline A1C: 8.1% Background treatment: ± metformin Dosage: once-daily iGlarLixi ± metformin, iGlar ± metformin Ratio iGlar/lixisenatide: 2:1 (1 U iGlar/0.5 µg lixisenatide) for doses between 10 U/5 µg and 40 U/20 µg, 3:1 (1 U iGlar/0.33 µg lixisenatide) for doses between 30 U/10 µg and 60 U/20 µg^b Maximum dose: 60 U iGlar for iGlarLixi and iGlar alone	Insulin: 47 U both groups Lixi: 17 μg both groups	A1C change: −1.1%, −0.6% (iGlarLixi, iGlar).Estimated treatment difference between iGlarLixi and iGlar: −0.5% [95% CI = −0.6, −0.4; p < 0.0001] Mean weight change: −0.7 kg, +0.7 kg (iGlarLixi, iGlar). Estimated treatment difference between iGlarLixi and iGlar: −1.4 kg [95% CI = −1.8, −0.9; p < 0.0001) Incidence of symptomatic hypoglycemia (plasma glucose ≤70 mg/dL): 40.0%, 42.5% (iGlarLixi, iGlar, respectively) Overall adverse events were similar; incidence of nausea: 10.4%, 0.5%. Incidence of vomiting: 3.6%, 0.5%^c(iGlarLixi and iGlar, respectively)	iGlarLixi showed superior glycemic control to iGlar, with beneficial effect on body weight, no additional risk of hypoglycemia and a low rate of gastrointestinal adverse events.
DUAL I trial, phase 3 NCT01336023^11	26-week RCT (n = 1663) Baseline A1C: 7.0–10.0% Background treatment: metformin with or without pioglitazone Dosage: once-daily IDegLira + metformin or IDeg + metformin or Lira + metformin Ratio IDeg/Lira: 1 U/0.036 mg Maximum dose: 50 U IDeg + 1.8 mg liraglutide for IDegLira, and no maximum doses for IDeg alone	Insulin: IDegLira group 38 U vs IDeg group 53 U (estimated treatment difference between IDegLira and IDeg –14.90 U, 95% CI = –17.14, –12.66; p < 0.0001) Lira: IDegLira group 1.4 mg vs Lira group 1.8 mg	A1C change: −1.9%, −1.4%, and −1.3% (IDegLira, IDeg, and Lira, respectively) Estimated treatment difference between IDegLira and IDeg: −0.47% [95% CI = −0.58, −0.36; p < 0.0001] Mean weight change: −0.5 kg, +1.6 kg and −3.0 kg (IDegLira, IDeg, and Lira, respectively; p < 0.0001) Confirmed hypoglycemia: 32%, 39%, and 7% (IDegLira, IDeg, and Lira, respectively) Overall adverse events were similar; incidence of nausea: 9%, 4%, and 20% (IDegLira, IDeg, and Lira, respectively)	IDegLira combined the clinical advantages of basal insulin and GLP-1 receptor agonist treatment, and mitigated their principal side-effects.
DUAL II trial, phase 3, double blind NCT01392573^12	26-week RCT (n = 413) Baseline A1C: 7.5–10.0% Background treatment: basal insulin analog and metformin with or without sulfonylurea/glinides Dosage: once-daily IDegLira + metformin or IDeg + metformin Maximum dose: 50 U IDeg +1.8 mg liraglutide for IDegLira, and 50 U for IDeg	Insulin: IDeg 45 units both groups	A1C change: −1.9% vs −0.9% (IDegLira vs IDeg) Estimated treatment difference: −1.1% [95% CI = −1.3, −0.8; p < 0.0001] Mean weight change of −2.7 kg vs no weight change (IDegLira vs IDeg; p < 0.0001) Hypoglycemia was comparable in both groups (∼ 25%) Overall adverse events were similar; incidence of nausea: 6.5% vs 3.5% (IDegLira vs IDeg)	IDegLira led to superior glycemic control compared to IDeg at equivalent insulin doses, with weight loss and no increased risk of hypoglycemia.
DUAL III trial, phase 3 NCT01676116^44	26-week RCT (n = 438) Background treatment: metformin ± pioglitazone ± sulfonylurea Dosage: once-daily IDegLira + continued OADs or unchanged GLP-1 receptor agonist therapy (liraglutide once daily or exenatide twice daily) + continued OADs	Insulin: 43 U Lira: 1.5 mg	A1C change (from baseline to end of study): from 7.8% to 6.4% vs from 7.7% to 7.4% (IDegLira vs unchanged GLP-1 receptor agonist therapy) Estimated treatment difference: −0.94% [95% CI = −1.11–0.78; p < 0.001] Mean weight change: +2.0 kg vs −0.8 kg (IDegLira vs unchanged GLP-1 receptor agonist therapy; p < 0.001) Rate of confirmed hypoglycemia: 2.82 events/PYE vs 0.12 events/PYE (IDegLira vs unchanged GLP-1 receptor agonist therapy)	IDegLira provided superior glycemic control vs unchanged GLP-1 receptor agonist therapy, with weight reduction.
DUAL IV trial, phase 3 NCT01618162^45	26-week RCT (n = 435) Baseline A1C: 7–9.0% Background treatment: metformin ± sulfonylurea Dosage: once-daily IDegLira or placebo	Insulin: 28 units Lira: 1.0 mg	A1C change (from baseline to end of study): from 7.9% to 6.4% vs from 7.9% to 7.4% (IDegLira vs placebo) Estimated treatment difference: −1.02% (p < 0.001) Mean weight change: +0.5 kg vs −1.0 kg (IDegLira vs placebo; p < 0.001) Incidence of confirmed severe hypoglycemia: 41.7% vs 17.1% (IDegLira vs placebo)	IDegLira provided superior glycemic control vs placebo, with a low rate of hypoglycemia. Hypoglycemia rates were greater to those seen in other Dual trials, possibly due to concomitant therapy with sulfonylurea.
DUAL V trial, phase 3 NCT01952145^15	26-week RCT (n = 557) Baseline A1C: 7–10.0% Background treatment: basal insulin analog (mean 32 U, no maximum) + metformin Dosage: once-daily IDegLira + metformin or continued insulin glargine uptitration + metformin Maximum dose: 50 U for IDegLira, and no maximum doses for IDeg alone	Insulin: IDegLira 41 U vs iGlar 66 U (estimated treatment difference between IDegLira and iGlar –25.47 U [95% CI = –28.9, –22.05]; p < 0.001) Lira: 1.48 mg	A1C change: −1.81% vs −1.13% (IDegLira vs iGlar) Estimated treatment difference: −0.59% [95% CI = −0.74, −0.45; p < 0.0001] Mean weight change: −1.4 kg vs +1.8 kg (IDegLira vs iGlar; p < 0.0001) Incidence of confirmed hypoglycemia: 28.4% vs 49.1% (IDegLira vs iGlar)	IDegLira was superior to iGlar in A1C reduction, with lower risk of hypoglycemia and weight reduction.
DUAL VI trial, phase 3 NCT02298192^46	32-week RCT (n = 420) Background treatment: metformin ± pioglitazone Dosage: once-daily IDegLira + metformin ±pioglitazone IDegLira titration: titrated once weekly based on the mean of 2 breakfast plasma glucose readings (1W); titrated twice weekly based on the mean of 3 breakfast plasma glucose readings (2W)	Insulin: 41 units both groups Lira: 1.48 units both groups	A1C change (from baseline to end of study): from 8.2% to 6.1% (1W); from 8.1% to 6.0% (2W) Estimated treatment difference: −0.12% [95% CI = −0.04, 0.28; p = 0.012] Mean weight change: −1.0 kg, −2.0 kg (1W, 2W, respectively) Rate of confirmed hypoglycemia: 0.16 events/PYE, 0.76 events/PYE (1W, 2W, respectively)	Titration of IDegLira with 1W or 2W adjustments resulted in comparable glycemic efficacy and safety profiles.
DUAL VII trial, phase 3 NCT02420262^47	26-week RCT Mean baseline A1C: 8.2% Background treatment: metformin and 20–50 U iGlar Dosage: IDegLira or iGlar + insulin aspart up to 4 times per day	Insulin: IDegLira group 40 units vs basal-bolus group 52 units basal (estimated treatment difference between IDegLira and basal-bolus –12.6 units [95% CI = –14.9, –10.3]; p < 0.0001) Lira: 1.44 mg	A1C change: −1.49 vs −1.48 (IDegLira vs iGlar + insulin aspart) Estimated treatment difference: −0.02 [95% CI = −0.16, 0.12] Mean weight change: −0.92 kg vs +2.64 (IDegLira vs iGlar + insulin aspart; p < 0.0001) Rate of overall hypoglycemia: 1.07 events/PYE vs 8.17 events/PYE (IDegLira vs iGlar + insulin aspart; p < 0.0001)	In patients with A1C > 7.0% on metformin and iGlar, IDegLira resulted in similar A1C reductions, greater weight loss, and lower risk of hypoglycemia events compared with iGlar plus insulin aspart.
DUAL I (extension), phase 3 NCT01336023^13	26-week extension RCT of Dual I(n = 1311) Background treatment: metformin ± pioglitazone Dosage: once-daily IDegLira + metformin or IDeg + metformin or Lira + metformin Maximum dose: 50 U IDeg +1.8 mg liraglutide for IDegLira, and no maximum doses for IDeg alone	Insulin: IDegLira 39 units vs IDeg 62 units (estimated treatment difference between IDegLira and IDeg −23.4 units, 95% CI = −26.4, −20.3; p < 0.0001) Lira: IDegLira group 1.4 mg vs Lira group 1.8 mg	A1C change: −1.84%, −1.40%, and −1.21% (IDegLira, IDeg, Lira) Estimated treatment difference between IDegLira and IDeg: −0.46% [95% CI = −0.57, −0.34; p < 0.0001] Mean weight change: −0.4 kg, +2.3 kg, and −3.0 kg (IDegLira, IDeg, and Lira, respectively; p < 0.0001) Rate of confirmed hypoglycemia events per 100 patient-years: 176.7, 279.1, and 19.1 (IDegLira, IDeg, and Lira, respectively) Overall adverse events and nausea were similar (IDegLira, IDeg, and Lira)	Over the 52 weeks of treatment, IDegLira was superior to IDeg alone in A1C reduction and Lira alone, with weight reduction and lower risk of hypoglycemia.

^a Patients from the modified-to-treat population with relevant baseline.

^b In the US, iGlarLixi is approved for use in the single patient use SoloSTAR pen injector in a 3:1 dosing ratio of iGlarLixi and lixisenatide; iGlar and lixisenatide are premixed in the cartridge.

^c Rates of vomiting were not reported for iDegLira in the DUAL trials.

Abbreviations: A1C, glycated hemoglobin A_1c; CI, confidence interval; GLP-1, glucagon-like peptide 1; iGlarLixi, lixisenatide/insulin glargine; iGlar, insulin glargine; IDeg, insulin degludec; IDegLira, insulin degludec/liraglutide; Lira, liraglutide; OAD, oral anti-hyperglycemic drug; PYE, patient-year of exposure; RCT, randomized controlled trial; U, units.

Clinical evaluation of titratable fixed-ratio co-formulations of a basal insulin analog and a GLP-1 receptor agonist

iGlarLixi

iGlarLixi⁴⁸, a co-formulation of insulin glargine 100 U/mL (iGlar) and lixisenatide (Lixi), is a single daily dose titratable fixed-ratio co-formulation injection. It has been approved by the US Food and Drug Administration (FDA) as an adjunct to diet and exercise to improve glycemic control in adults with T2D inadequately controlled on basal insulin (less than 60 U daily) or lixisenatide. iGlarLixi has also received marketing authorization from the European Medicines Agency (EMA) for the treatment of adults with T2D to improve glycemic control when this has not been provided by metformin alone or metformin combined with another oral glucose-lowering medicinal product or with basal insulin⁴⁹.

The efficacy of iGlarLixi compared with iGlar was investigated in a 24-week phase 2b proof-of-concept study in insulin-naive patients with T2D on OADs¹⁶. Once-daily iGlarLixi demonstrated greater A1C reduction than iGlar, with a similar rate of symptomatic hypoglycemia being observed in both groups¹⁶ (Table 3). Two phase 3 clinical trials investigated the efficacy of iGlarLixi compared with one or both of the individual components. The LixiLan-O trial compared iGlarLixi with iGlar and lixisenatide in patients with T2D inadequately controlled on metformin alone or on metformin combined with a second OAD¹⁷, whereas the LixiLan-L trial compared iGlarLixi with iGlar in patients inadequately controlled on basal insulin with or without OADs¹⁸. In both trials, the once-daily iGlar dose was capped at 60 U. These trials showed that treatment with once-daily single injections of iGlarLixi resulted in a statistically superior reduction in A1C compared with either lixisenatide or iGlar, with a safety profile reflecting those of the individual components (Table 3). Furthermore, patients on iGlarLixi were more likely to achieve A1C < 7.0% compared with iGlar in both LixiLan-O (73.7% vs 59.4%; p < 0.0001) and LixiLan-L (54.9% vs 29.6%; p < 0.0001) or compared with lixisenatide in LixiLan-O (73.7% vs 33.0%; p < 0.0001)^17,18.

In both LixiLan-O and LixiLan-L, patients treated with iGlarLixi experienced slight weight loss, whereas patients treated with iGlar experienced weight gain (p < 0.0001 for both studies); in LixiLan-O, patients treated with lixisenatide alone had the greatest weight loss. In both of these studies, the incidences and event rates of documented symptomatic hypoglycemia were similar in iGlarLixi- and iGlar-treated patients (Table 3)^17,18. Furthermore, in the LixiLan-O trial, the prevalence of nausea and vomiting, and associated treatment discontinuation, were lower with iGlarLixi than with lixisenatide (Table 3)¹⁷.

Post-hoc analyses of the LixiLan trials have shown similar results to the primary analyses^50,51. Data from three individual studies (ACT6011, DRI6012 [NCT00299871], and LixiLan-O) evaluating the impact of lixisenatide over the dose range delivered by iGlarLixi (5–20 μg) concluded that, even at lower doses, lixisenatide provides clinically meaningful improvements in PPG and A1C. A lixisenatide dose of 5 μg gave a 44% reduction in PPG under the curve compared with a 20 μg dose, and the maximum reduction in PPG area under the curve was achieved at 12.5 μg, indicating that a 20 μg dose is not necessary for the greatest reduction in PPG⁵¹. Efficacy and safety of iGlarLixi also appear to be maintained in special populations. In older patients (≥65 years) more iGlarLixi-treated patients achieved A1C < 7.0% than those treated with iGlar alone in both LixiLan-O (78.0% vs 54.4%; p < 0.001) and LixiLan-L (51.8% vs 21.0%; p < 0.001). In addition, there were reductions in the number of hypoglycemic events in older patients treated with iGlarLixi compared with those treated with iGlar alone in both LixiLan-O (1.42 vs 1.89 events/patient-year; p = 0.035) and LixiLan-L (2.84 vs 4.91 events/patient-year; p < 0.001). The proportion of older adults achieving the composite endpoint of A1C < 7.0% with no weight gain and no documented hypoglycemia (SMPG ≤70 mg/dL) was also greater with iGlarLixi than with iGlar alone in both LixiLan-O (30.1% vs 14.0%; p = 0.002) and LixiLan-L (26.4% vs 8.4%; p < 0.001)⁵². Results from another post-hoc analysis of the LixiLan trials indicate that the lixisenatide component of iGlarLixi mitigates iGlar-associated weight gain while being associated with better gastrointestinal tolerance⁵³. This result is attributed to the more gradual titration of lixisenatide when it is a component of iGlarLixi, rather than a separate injection⁵³.

A further post-hoc analysis of the LixiLan trials investigated the time taken to achieve glycemic control with iGlarLixi vs iGlar alone. In LixiLan-O, the A1C < 7.0% target was reached by 50% of patients in approximately half the time when treated with iGlarLixi compared with those patients treated with iGlar (median time to target = 85.0 days vs 166.0 days; p < 0.0001). In LixiLan-L, it took a median time of 153.0 days for 50% of iGlarLixi-treated patients to reach the A1C target, whereas the target was never achieved by 50% of patients treated with iGlar (p < 0.0001). The time to FPG control was similar between the two treatment arms, indicating that the PPG-lowering effect of lixisenatide contributes to patients achieving overall glycemic control⁵⁴. An additional post-hoc analysis of LixiLan-L determined that patients with A1C levels at screening that were either ≤8%, 8% < A1C ≤ 9%, or >9%, receiving iGlarLixi, had greater reductions in A1C at study end compared with those receiving iGlar, indicating that iGlarLixi is more effective than iGlar at controlling A1C, regardless of A1C value at screening⁵⁵.

IDegLira

IDegLira⁵⁶ is a once-daily single-injection of a titratable fixed-dose co-formulation of the long-acting basal insulin analog degludec (IDeg) and the once-daily GLP-1 receptor agonist liraglutide (Lira) that has been approved by the FDA as an adjunct to diet and exercise to improve glycemic control in adults with T2D inadequately controlled on basal insulin (<50 U daily) or liraglutide (≤1.8 mg daily). IDegLira has also been granted marketing authorization by the EMA for treatment of adults with T2D to improve glycemic control in combination with oral glucose-lowering medicinal products when these alone, or combined with a GLP-1 receptor agonist or basal insulin, do not provide adequate glycemic control⁵⁷.

The DUAL clinical development program for IDegLira encompasses a series of phase 3, 26-week, randomized clinical trials in patients with T2D. Data indicate that treatment with IDegLira results in superior A1C reductions compared with basal insulin or liraglutide alone. This was with a lower risk of hypoglycemia than with IDeg (in trials where the comparator insulin was not capped) and with additional benefits in terms of weight change in both insulin-naive and insulin-experienced patients^11,12,15,44 (Table 3). In DUAL III, patients who switched to IDegLira showed superior glycemic control to those who continued on GLP-1 receptor agonist therapy (liraglutide once daily or exenatide twice daily) (Table 3). A significantly higher proportion of IDegLira-treated patients achieved the secondary end-point of A1C < 7.0% compared with IDeg alone in both DUAL I (81% vs 65%; p < 0.0001) and DUAL II (60.3% vs 23.2%; p < 0.0001)^11,12. Similar results were shown in the DUAL V study of patients with T2D not adequately controlled on iGlar, where A1C < 7% was an exploratory pre-specified end-point; 76% of patients treated with IDegLira achieved an A1C < 7.0% vs 47% who escalated their dose of insulin glargine (p < 0.001)¹⁵. Data from the 26-week extension study to DUAL I demonstrated that the benefits of IDegLira were sustained, with a greater reduction in A1C, significantly greater weight loss, and lower rate of hypoglycemia at 52 weeks with IDegLira compared with IDeg. Rates of hypoglycemia were higher, and weight loss lower, for IDegLira compared with liraglutide alone (Table 3)¹³. A prospective sub-study from DUAL I using continuous glucose monitoring showed that, compared with each of its components, IDegLira improved mean PPG excursions in patients with T2D; this was possibly due to the greater endogenous insulin secretion and improved β-cell function observed in those patients, rather than to enhanced glucagon suppression⁵⁸. Overall, patients receiving IDegLira in the DUAL studies required a significantly lower dose of insulin by the end of the study than those using IDeg or iGlar (especially in trials where the comparator basal insulin was not capped)^13,15. In addition, a post-hoc analysis of data from the DUAL II and DUAL IV studies reported that the proportion of patients experiencing gastrointestinal adverse events during treatment with IDegLira, IDeg alone, or placebo was similar, and lower than that reported in patients treated with liraglutide alone in DUAL I. This is likely due to the slow, steady titration of IDegLira⁵⁹.

DUAL VII⁴⁷ compared IDegLira with basal-bolus therapy (insulin glargine plus insulin aspart). Data suggested that, at least in basal-insulin-experienced patients, transition to IDegLira may be superior to addition of a fast-acting insulin. The decrease in mean A1C was similar between the two regimens (−1.49% vs −1.48%, respectively; p < 0.0001 for non-inferiority). FPG at week 26 was also comparable for patients treated with IDegLira and insulin glargine plus insulin aspart (109.7 mg/dL vs 115.1 mg/dL, respectively). Patients treated with IDegLira had weight loss rather than weight gain (−0.92 kg vs +2.64 kg; p < 0.0001) and had markedly fewer episodes per patient year of overall hypoglycemia (1.07 vs 8.17; p < 0.0001) and nocturnal hypoglycemia (0.13 vs 1.66; p < 0.0001). Additionally, a greater number of patients achieved the triple composite endpoint of A1C < 7%, with no hypoglycemia episodes and no weight gain with IDegLira (34.9% vs 4.7%; p < 0.0001)⁴⁷. To date, no similar trial has been conducted using iGlarLixi; however, GetGoal-Duo-2 compared the addition of lixisenatide (20 μg once daily) or insulin glulisine (once daily or thrice daily) in patients treated with insulin glargine, allowing for an indirect comparison¹⁴. Changes in glycemic parameters were similar between treatment groups. Lixisenatide-treated patients experienced weight loss (−0.6 kg), whereas those treated with insulin glulisine experienced weight gain in both once daily (+1.0 kg) and thrice daily (+1.4 kg) treatment groups. Furthermore, a higher percentage of patients did not experience weight gain when treated with lixisenatide (64.7%) compared with insulin glulisine once (36.6%) or thrice daily (30.5%). Symptomatic hypoglycemia occurred in fewer patients treated with lixisenatide compared with insulin glulisine once daily (35.9% vs 46.5%; p = 0.01) and thrice daily (35.9% vs 52.4%; p = 0.0001). No severe symptomatic hypoglycemia occurred in the lixisenatide or thrice-daily insulin glulisine groups, and only two (0.7%) occurred in the insulin glulisine once-daily group¹⁴.

Although the LixiLan trials and the DUAL studies differed in design, their findings indicate that such co-formulations can indeed improve glycemic control and tolerability compared with either basal insulin or GLP-1 receptor agonists alone, with an increase in the number of patients reaching glycemic targets and low rates of adverse events (hypoglycemia, weight gain, and gastrointestinal adverse events). Furthermore, data from patient-reported outcomes questionnaires completed by participants in the DUAL V study showed that, compared to iGlar, IDegLira-treated patients had significantly greater improvements in treatment-related satisfaction, particularly treatment burden and diabetes management, and in patient-reported physical health¹⁵. A possible limitation of the LixiLan and DUAL trials is that neither co-formulation was compared directly with co-administration of the GLP-1 receptor agonist and basal insulin as separate injections. Head-to-head trials would be needed to determine any additional benefit of the co-formulations over co-administration of the individual agents. However, indirect comparisons of the results of the LixiLan and GetGoal-Duo trials suggest favorable results with the co-formulation in comparable populations^17,18. Overall, results of the trials to date suggest that titratable fixed-ratio co-formulations of a basal insulin analog and a GLP-1 receptor agonist are effective therapies for both insulin-naive and insulin-experienced patients, although labeling in the US is limited to insulin-experienced patients only or those inadequately controlled on lixisenatide or liraglutide, respectively^{11–13,15–18,44,45}.

Optimizing the use of titratable fixed-ratio co-formulation in T2D therapy

The optimal stage in the progression of T2D at which to add a titratable fixed-ratio co-formulation of basal insulin and a GLP-1 receptor agonist will depend on individual patient characteristics and preferences. A potential limitation of the currently available fixed ratio co-formulations is the top dose of insulin available in each formulation (50 U IDegLira and 60 U for iGlarLixi), influenced by the maximum approved GLP-1 receptor agonist dose and the co-formulation ratio of the basal insulin and the GLP-1 receptor agonist. There are also no studies directly comparing sequential administration of basal insulin and a GLP-1 receptor agonist with simultaneous administration of a co-formulation. Despite this lack of head-to-head studies, comparison of results from separate studies suggests that co-administration, rather than sequential addition of a basal insulin analog and a GLP-1 receptor agonist, might be an optimal strategy for many patients^11,12,60, as these results indicate that early simultaneous treatment with a titratable fixed-ratio co-formulation of basal insulin plus a GLP-1 receptor agonist may be more effective than sequential administration. Real-world evidence also suggests limited persistency with the free dose combination of a GLP-1 receptor agonist and basal insulin, with a median time to treatment discontinuation of 133 days⁶¹. Furthermore, a titratable fixed-ratio co-formulation of these therapies may provide more effective treatment, with potentially improved gastrointestinal tolerability, than administration of the two components separately. Given the multiple pathophysiological abnormalities observed in patients with T2D, and the progressive nature of β-cell dysfunction, there might well be a potential benefit in initiating combination therapy with agents that correct different pathogenic defects in T2D earlier in the disease course⁶².

Summary

The underlying pathophysiology of T2D involves the progressive loss of pancreatic β-cell function and incretin activity. As the disease progresses, most patients will eventually require the use of combinations of anti-hyperglycemic agents with complementary mechanisms of action.

The combination of a basal insulin analog together with a GLP-1 receptor agonist is currently recommended in ADA and American Association of Clinical Endocrinologists (AACE)/American College of Endocrinology (ACE) guidelines, algorithms, and position statements as an initial strategy for patients with T2D not achieving A1C goal on basal insulin plus OADs, and is also indicated for patients with T2D inadequately controlled on GLP-1 receptor agonist and OADs^3,9,10,48,56. The use of titratable fixed-ratio co-formulations of a basal insulin analog and a GLP-1 receptor agonist obviates the need for two separate injections and reduces regimen complexity associated with the administration of each agent individually, and so may have a positive effect on treatment initiation and adherence. In addition to this benefiting the patient, the simplified dosing regimen could also benefit healthcare professionals, therefore reducing a potential barrier to treatment initiation or intensification. Data from clinical trials show that titratable fixed-ratio co-formulations improve glycemic control, allowing more patients with T2D (both insulin-naive and insulin-experienced) to reach their glycemic targets when compared with each medication used separately. The use of these titratable fixed-ratio co-formulations has often been associated with less weight gain, or even weight loss, compared with basal insulin alone, and a lower rate of gastrointestinal adverse events compared with a GLP-1 receptor agonist alone. Reducing regimen complexity has the potential to curtail clinical inertia and, along with fewer adverse events, improve adherence and persistence to therapy, which may improve clinical outcomes.

Transparency

Declaration of funding

Writing/editorial support funded by Sanofi US, Inc.

Declaration of financial/other relationships

L.B. has disclosed he has received grants/research support from AstraZeneca, Janssen Pharmaceuticals, Inc., Lexicon Pharmaceuticals, Inc., Merck & Co., Novo Nordisk, and Sanofi; is a Speaker for AstraZeneca, Janssen Pharmaceuticals, Inc., Merck & Co., Novo Nordisk, and Sanofi; and a Consultant for AstraZeneca, GlaxoSmithKline, Intarcia Therapeutics, Inc., Janssen Pharmaceuticals, Inc., Merck & Co., Inc., Novo Nordisk, and Sanofi. J.E.A. has disclosed that he is an Advisory Board member for Abbott Diabetes, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Janssen, Merck, Novo Nordisk, and Sanofi; and a Speakers Bureau member for AstraZeneca, Boehringer Ingelheim, Eli Lilly, Janssen, and Sanofi; he is a stock shareholder for NuSirt Biopharma. P.C. has disclosed that he has received a research and travel grant from Sanofi; is a Speaker for Janssen and Novo Nordisk; and a Consultant for Pfizer. J.A.D. has no relevant financial or other relationships to disclose. CMRO peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors would like to thank Dr R. Aronson for his contributions and critical review of the manuscript. The authors received writing/editorial support in the preparation of this manuscript provided by Catarina Fernandes, PhD, and Keisha Peters, MSc, of Excerpta Medica, funded by Sanofi US, Inc.