Options for intensification of basal insulin in type 2 diabetes: Premeal insulin or short-acting GLP-1 receptor agonists?

Transcription

1 Diabetes & Metabolism 41 (2015) 6S21-6S27 Options for intensification of basal insulin in type 2 diabetes: Premeal insulin or short-acting GLP-1 receptor agonists? P. Darmon, D. Raccah* Pôle Endocrinologie, Nutrition, Diabète et Obésité, Hôpital de la Conception, 147 Boulevard Baille, 13285, Marseille, France INRA, UMR 1260 ; INSERM, UMR 1062 ; Nutrition, Obésité et Risque Thrombotique ; Aix-Marseille Université, Faculté de Médecine, Marseille, France Abstract Type 2 diabetes is an evolutive disease with a progressive defect of beta-cell insulin secretion. This characteristic points to a need for treatment that takes into account such a natural history. When oral antidiabetic drugs fail to achieve the patient s target HbA 1c level, basal insulin treatment is usually initiated and titrated in association with oral drugs to manage fasting hyperglycaemia. Over a period of time, it is enough to simply achieve the HbA 1c target. However, when even a good fasting blood glucose level is no longer sufficient to control overall glycaemia, then prandial treatment must be combined with the titrated basal insulin to deal with the postprandial hyperglycaemia responsible for the elevation of HbA 1c. Of the different therapeutic options now available for this, rapid-acting insulins and GLP-1 receptor agonists (RAs) can be used. Rapidacting insulins can be added either at each meal, achieving full insulin supplementation with a basal-bolus regimen, or at the main meal only as a basal-plus regimen. Compared with the full basal-bolus, the basal-plus strategy is associated with fewer injections, yet provides similar efficacy in terms of HbA 1c improvement, but with less weight gain and lower hypoglycaemic risk. As for GLP-1 RAs, numerous studies, and especially those using short-acting GLP-1 RAs, have demonstrated more pronounced effects on postprandial hyperglycaemia, good complementary effects with basal insulin, and significant improvement of HbA 1c with no weight gain and a low risk of hypoglycaemia. Similarly, direct and indirect comparisons of the use of rapid-acting insulins and GLP-1 RAs to intensify basal insulin have shown comparable efficacy in terms of HbA 1c control, but with less weight gain and fewer hypoglycaemic episodes with GLP-1 RAs Elsevier Masson SAS. All rights reserved Keywords: GLP-1 receptor agonists; Postprandial hyperglycaemia; Basal insulin; Premeal insulin 1. Introduction When oral antidiabetic drugs (OADs) become insufficient to achieve glycaemic goals in type 2 diabetes (T2D), adding basal insulin is the unanimously recommended strategy for improving glucose control [1]. However, the progressive decline of β-cell function in T2D and the inability of basal insulin to control postprandial glucose spikes mean that this regimen will inexorably need to be intensified over time to counteract the deterioration of glucose control. When HbA 1c remains over target levels despite adequate, structured titration of basal insulin to control fasting plasma glucose (FPG), the next therapeutic step needs to address postprandial plasma glucose (PPG) excursions [2]. This may be achieved by intensifying insulin therapy, usually with the addition of one injection of rapid-acting insulin analogue before the meal inducing the highest postprandial glucose spike ( basal-plus ) or three injections of rapid-acting insulin analogue before each meal ( basal-bolus ) or, in some selected patients, the less flexible use of premixed formulations of intermediate- and rapid-acting insulin in fixed ratios [1]. Nevertheless, although intensifying insulin therapy generally provides a significant decrease in HbA 1c, this happens at the cost of hypoglycaemia, weight gain and, in some patients, difficulty with, or a reluctance and/ or unwillingness to manage, a multidose insulin regimen. However, the recent updated American Diabetes Association (ADA)/European Association for the Study of Diabetes (EASD) position statement on the management of hyperglycaemia in T2D now presents the option of adding a glucagon-like peptide-1 receptor agonist (GLP-1 RA) for T2D patients whose diabetes remains uncontrolled despite a combination of basal insulin and OADs [1]. The rationale for such an option is justified by physiological and pharmacological considerations, and is now strengthened by a growing amount of robust clinical data [3-8]. Furthermore, it should be emphasized that such a combination is now also approved by health authorities. *Corresponding author. address: denis.raccah@ap-hm.fr (D. Raccah) Elsevier Masson SAS. All rights reserved.

2 S22 P. Darmon, D. Raccah / Diabetes & Metabolism 41 (2015) 6S21-6S27 Table 1 Key efficacy and safety parameters in the intensification of basal insulin treatments with rapid-acting insulin Change from baseline in HbA1c (%) HbA 1c treatment difference (%) Basal-bolus 0.3 to -2.1 vs Premix: to -0.2 vs Basal-plus: Change from baseline in PPG (mmol/l) Change from baseline in FPG (mmol/l) Change from baseline in body weight (kg) Hypoglycaemia (% of patients) to 4.5 Symptomatic: 11 to 89 Severe: 1 to 2 Nocturnal: 9 to 59 Basal-plus 0.4 to -1.1 vs Insulin glargine: to -0.3 PPG: postprandial plasma glucose; FPG: fasting plasma glucose 0.8 to to 1.8 Symptomatic: 46 to 52 Severe: 0 to 2 Nocturnal: 11 to Intensification of basal insulin with rapid-acting insulins (RAIs) Treatment intensification with RAIs can be administered as either a basal-plus regimen (basal insulin plus one RAI injection at the main meal) or basal-bolus regimen (basal insulin plus an injection of RAI before each meal). Basal-bolus treatment is the more physiological approach, while basal-plus was developed to provide a simplified dosing regimen. Key efficacy and safety parameters for intensification of basal insulin with rapid acting insulins are developed below and summarized in Table Basal-bolus treatment Thirteen studies have reported on the efficacy of basal-bolus treatment for T2D [9-21]. Study durations ranged from 16 weeks to 1 year, and the majority of patients were obese at enrolment with an HbA 1c > 8%. For all these studies, the mean changes from baseline in HbA 1c and body weight ranged from -0.3% to -2.1% and from 0.6 kg to 4.5 kg, respectively. Dailey et al. [22] reported a significant reduction from baseline in PPG with a basal-bolus regimen in a pooled analysis of five clinical trials (baseline: 183 mg/dl; endpoint: 152 mg/dl), while self-monitored plasma glucose (SMPG) was reduced vs a comparator agent in seven basal-bolus reports [10-12,18-21]. One 18-month open-label study examined the most effective way to initiate RAI therapy by comparing administration of thrice daily (TD) insulin glulisine from treatment onset with a progressive increase in the number of insulin glulisine injections [21]. During a 6-month optimalization period, insulin glargine was administered as a single injection while insulin glulisine was titrated according to PPG. At 18 months, optimal glycaemic control (HbA 1c < 7%) was not achieved in any of the study groups. However, a progressive increase in insulin glulisine did result in significantly smaller weight increases [21], suggesting that this approach may be appropriate for patients who are overweight or obese, as it may attenuate weight gain compared with basal-bolus treatment. Yet another, 32-week, study showed that stepwise prandial insulin intensification provided glycaemic control that was non-inferior to a full basal-bolus regimen with a significantly lower hypoglycaemia risk and better patient satisfaction [12]. A retrospective observational study conducted between 2001 and 2009 compared basal-bolus treatment with premixed insulins and found no significant differences in the number of propensity-matched patients attaining an HbA 1c < 7% at 1 year [14]. However, treatment persistence and adherence were significantly higher in the basal-bolus group, which the authors suggested may have been due to the fact that approximately half of premix-treated patients switched back to insulin glargine (with or without an RAI) during follow-up, owing to better treatment outcomes with glargine-based regimens [14] Basal-plus treatment This strategy for treatment intensification is based on three arguments: prandial glycaemia is part of HbA 1c ; prandial glucose excursions are not the same after each meal of the day; and correction of the highest prandial glucose excursion will have an impact on the rest of the glycaemia, especially preprandial episodes, so avoiding the need to cover every meal of the day with RAI [23]. Eight studies examined the efficacy of basal-plus regimens for the treatment of T2D [21,22,24-29]; five of them had no control arm [22,25-27,29], but are included for completeness. Study durations ranged from 12 weeks to 1 year and the majority of patients were overweight at enrolment, with HbA 1c levels > 7.5%. Mean changes from baseline in HbA 1c, PPG and body weight ranged from -0.4% to -1.1%, -0.8 mmol/l to -3.9 mmol/l and 0.5 kg to 1.8 kg, respectively. In a 6-month, phase-iv, controlled study that compared insulin glargine with insulin glargine plus insulin glulisine, a basal-plus regimen significantly improved glycaemic control, albeit with a non-significant increase in body weight [27]. Two meta-analyses confirmed these data, and demonstrated that basal-plus treatment resulted in non-significant or small weight increases and was associated with a low prevalence of hypoglycaemias [25,30]. In one pooled study without a control arm in which patients who initiated basal insulin (titrated according to a protocoldefined blood glucose in two out of four pooled studies), Owens et al. [28] demonstrated that basal-plus with insulin glulisine resulted in significant improvements from baseline at week 24 in HbA 1c and PPG, regardless of T2D duration. These data suggest that the basal-plus is an effective regimen

3 P. Darmon, D. Raccah / Diabetes & Metabolism 41 (2015) 6S21-6S27 6S23 even in patients who may have advanced disease and poor β-cell function, such as elderly individuals. In another 24-week study, a single daily injection of insulin glulisine titrated at the investigator s discretion on top of basal insulin ( 3 months of treatment at baseline; dose adjusted according to target FPG) provided effective glycaemic control irrespective of whether it was given at breakfast or at the main meal. However, more patients achieved an HbA 1c 7% at week 24 in the main-meal group (52.2%) compared with the breakfast group (36.5%) [31]. Moreover, yet another study comparing titrated insulin glargine plus 0-1 insulin glulisine injections (basal-plus) with titrated insulin glargine plus 0-3 insulin glulisine injections (basal-bolus) for 60 weeks found that the reduction in HbA 1c with basal-plus was statistically non-inferior to that with basal-bolus. Although body weight increased progressively in all groups, at week 60, weight increases were significantly lower in the basal-plus group [29]. Taken altogether, these data demonstrate that basal-plus regimens can provide glycaemic control comparable to basalbolus regimens, but with a flexibility of dosing that, alongside the smaller number of injections and smaller weight increases, may improve compliance [32]. 3. Intensification of basal insulin with GLP-1 RAs Combining basal insulin with a GLP-1 RA has a logical and compelling rationale, as these therapies have complementary properties [3]. On the one hand, basal insulin primarily targets FPG control by inhibiting hepatic glucose production overnight, but induces fat storage and weight gain. On the other hand, GLP-1 RAs: i) display glucose-dependent insulin-enhancing and glucagon-suppressing effects that provide additional lowering of PPG without increasing the risk of hypoglycaemic events, so totally preserving this counterregulatory response to hypoglycaemia; and ii) delay gastric-emptying and promote satiety, which may contribute to the reduction of insulin-induced weight gain and insulin dose requirements. In addition, a theoretical benefit of such an association is that exogenous insulin may give b cells a rest and reduce glucotoxicity, enabling greater recovery of the endogenous prandial insulin response, a major target of GLP-1 RAs. Furthermore, incretin-stimulated insulin secretion is more physiological than exogenous meal-time insulin administration, as insulin is secreted via the portal system. Indeed, recent data suggest that the combination of insulin and GLP-1 in T2D is more effective than either insulin or GLP-1 alone for improving endothelial dysfunction, inflammation and oxidative stress [33]. In recent years, numerous observational and interventional studies have demonstrated the benefits of combining basal insulin and GLP-1 RAs in T2D (irrespective of the sequence of their introduction) [3-8]. In a systematic review and metaanalysis of 15 randomized controlled trials, the combination of basal insulin and a GLP-1 RA was shown to significantly improve glycaemic control compared with other antidiabetic treatments, along with no increases in either weight loss or hypoglycaemia; there was even a decrease in hypoglycaemia risk compared with basal-bolus therapy [34]. Of the GLP-1 RAs, short-acting ones such as exenatide and lixisenatide may represent more logical and suitable options after failure of a basal insulin regimen because of their more pronounced effects on PPG through their preferential action on gastric motility [35]; in contrast, long-acting GLP-1 RAs, administered once daily (liraglutide) or once weekly (exenatide QW, albiglutide, dulaglutide, semaglutide), preferentially target FPG. Table 2 summarizes the design of seven randomized controlled studies evaluating the addition of a short-acting GLP-1 RA in T2D patients who were uncontrolled despite basal insulin. Their main findings are summarized below and in Table Results of randomized controlled studies of exenatide after basal insulin In a study by Shao et al. [36], 60 obese patients with non-alcoholic fatty liver disease and newly diagnosed with diabetes were randomized to receive exenatide twice daily or insulin aspart thrice daily in addition to insulin glargine for 12 weeks: at the end of the intervention, FPG, PPG and HbA 1c were all significantly decreased in both groups, whereas body weight and waist circumference significantly decreased in the exenatide group, but increased in the intensive insulin group (P < 0.001). Also, levels of hepatic enzymes in the exenatide group were significantly lower than in the intensive insulin group (P < 0.001). Furthermore, the reversal rate of fatty liver was significantly higher in the exenatide group (93.3%) than in the intensive insulin group (66.7%; P < 0.01). With the aim of comparing twice-daily exenatide injections vs placebo on HbA 1c levels in T2D patients treated by basal insulin, Buse et al. [37] conducted a randomized controlled trial of 261 adults with T2D and HbA 1c levels of %; all were using insulin glargine alone or in combination with metformin and/or pioglitazone and assigned to exenatide 10 µg or placebo twice daily for 30 weeks. At week 30, HbA 1c levels decreased by 1.74% with exenatide and by 1.04% with placebo (between-group difference: -0.69%, 95% CI: to -0.46%; P < 0.001), and the difference was primarily due to a greater reduction in PPG with exenatide. Weight decreased by 1.8 kg with exenatide and increased by 1.0 kg with placebo (betweengroup difference: -2.7 kg, 95% CI: -3.7 to -1.7 kg). Average increases in insulin dosages with exenatide and placebo were 13 U/d and 20 U/d, respectively. The estimated rate of minor hypoglycaemia was similar between groups. On exploratory post-hoc analysis, these results were confirmed irrespective of baseline HbA 1c, diabetes duration and body mass index [38], and regardless of the extent of insulin titration [39]. As expected, gastrointestinal side-effects were more frequent with exenatide vs placebo (nausea: 41% vs 8%; diarrhoea: 18% vs 8%; vomiting: 18% vs 4%; and constipation: 10% vs 2%, respectively). In an open-label, multicentre, randomized non-inferiority trial (4B Study), Diamant et al. [40] compared exenatide twice daily (10 20 µg/d) and mealtime insulin lispro thrice daily (titrated to a premeal glucose level of mmol/l) in 637 T2D patients whose glycaemic control remained poor with

4 S24 P. Darmon, D. Raccah / Diabetes & Metabolism 41 (2015) 6S21-6S27 Table 2 Randomized controlled trials evaluating the association of basal insulin and short-acting glucagon-like peptide-1 receptor agonists (GLP-1 RAs) Mean baseline HbA 1c (%) Mean diabetes duration (years) Mean baseline weight (kg) Mean age (years) Subjects (n) Interventions in study groups Duration of interventions (weeks) Background treatments in study groups EXE BD vs PBO Buse et al Glargine ± MET, PIO or both Seino et al (GetGoal-L Asia) Basal insulin ± SU LIXI QD vs PBO Riddle et al (GetGoal-L) Basal insulin ± MET LIXI QD vs PBO LIXI QD vs PBO 24 (after 12-week glargine titration) Riddle et al., 2013 (GetGoal-Duo 1) Glargine + MET ± thiazolidinedione Diamant et al., 2014 (4B Study) Glargine + MET EXE BD vs lispro TD 30 (after 12-week glargine titration) Shao et al., 2014 Glargine EXE BD vs aspart TD Newly diagnosed Rosenstock et al., 2015 (GetGoal-Duo 2) Basal insulin ± MET LIXI QD vs glulisine QD or TD 26 (after 12-week glargine titration) MET: metformin; PIO: pioglitazone; EXE: exenatide; BD: twice daily; PBO: placebo; SU: sulphonylurea; LIXI: lixisenatide; QD: once daily; TD: thrice daily metformin and insulin glargine, despite a 12-week period of up-titration. At week 30, HbA 1c reduction with exenatide was non-inferior to the reduction observed with insulin lispro [-1.13% and -1.10%, respectively; treatment differences: (95% CI: to 0.11) in per-protocol and (95% CI: to 0.11) in intention-to-treat populations]. Exenatide was associated with significantly lower FPG (6.5 mmol/l vs 7.2 mmol/l; P = 0.002) and weight (-2.5 kg vs +2.1 kg; P < 0.001) vs basal-bolus therapy. Incidence of minor diurnal hypoglycemia was lower with exenatide than with lispro (30% vs 41%) whereas no difference was seen for nocturnal hypoglycemia (25% vs 27%). Two exenatide and seven lispro recipients displayed at least one major hypoglycemic episode. Exenatide resulted in more gastrointestinal adverse events than lispro. Nevertheless, more patients reported treatment satisfaction and better quality of life with exenatide than with lispro, despite the larger proportion of patients experiencing adverse events with exenatide Results of randomized controlled studies of lixisenatide after basal insulin GetGoal-L was a randomized controlled trial conducted to compare the efficacy and safety of adding lixisenatide 20 μg once daily vs placebo to previous basal insulin therapy administered alone or together with metformin in 495 patients with uncontrolled T2D [41]. The basal insulin dose remained unchanged throughout the study except when necessary to limit hypoglycaemia. At week 24, the placebo-corrected change of HbA 1c from baseline with lixisenatide was -0.4% (P = ). In addition, lixisenatide significantly reduced 2-h plasma glucose levels after a standardized breakfast (placebo-corrected: -3.8 mmol/l; P < ), body weight (placebo corrected: -1.3 kg; P < ) and insulin daily doses (placebo corrected: -3.7 U/d; P = 0.012). Symptomatic hypoglycaemia was reported by 28% of patients treated with lixisenatide and 22% of patients taking a placebo (severe hypoglycaemia: 1.2% vs 0%, respectively). The GetGoal-L Asia trial had the same design and involved 311 Asian T2D patients who were insufficiently controlled by basal insulin therapy administered either alone or with or without sulphonylureas [42]. At week 24, the placebocorrected change of HbA 1c from baseline with lixisenatide was -0.88% (P < ). Lixisenatide also significantly improved 2-h PPG and glucose excursions, average 7-point self-monitored blood glucose levels and FPG. Symptomatic hypoglycaemia was more frequent with lixisenatide (42.9%) vs placebo (23.6%), but was similar between groups not taking sulphonylureas (32.6% vs 28.3%, respectively). No severe hypoglycaemia was reported. The greater benefits found in GetGoal-L Asia than in GetGoal-L may be explained by the fact that incretin-based therapies might be more effective in Asian T2D patients due to pathophysiological considerations, such as more pronounced insulin and GLP-1 deficiencies in such patients.

5 P. Darmon, D. Raccah / Diabetes & Metabolism 41 (2015) 6S21-6S27 6S25 Table 3 Main results of randomized controlled trials evaluating the addition of a short-acting GLP-1 RA to basal insulin HbA 1c Weight Hypoglycemia Buse et al. (2011) -1.74% (EXE) vs -1.04% (PBO), Δ -0.69% (95% CI to -0.46, p < 0.001) -1.8 kg (EXE) vs +1 kg (PBO), Δ -2.7 kg (95% CI -3.7 to -1.7, p < 0.001) Minor hypoglycemia (% patients) : 25% (EXE) vs 29% (PBO) Minor nocturnal hypoglycemia (% patients) : 17% (EXE) vs 26% (PBO) Major hypoglycemia (% patients) : 0% (EXE) vs 1% (PBO) Seino et al. (2012) (GetGoal-L Asia) -0.77% (LIXI) vs +0.11% (PBO), Δ -0.88% (95% CI to -0.65, p < ) kg (LIXI) vs kg (PBO) (95% CI -0,925 to +0,061, p = 0.09) 42.9% (LIXI) vs 23.6% (PBO) Symptomatic hypoglycemia (% patients not receiving SU) : 32.6% (LIXI) vs 28.3% (PBO) Riddle et al. (2013) (GetGoal-L) -0.7% (LIXI) vs -0.4% (PBO), Δ -0.4% (95% CI -0.6 to -0.2, p = ) -1.8 kg (LIXI) vs -0.5 kg (PBO), Δ -1.3 kg (95% CI -1.8 to -0.7, p < ) 27.7% (LIXI) vs 21.6% (PBO) Severe hypoglycemia (% patients) : 1.2% (LIXI) vs 0% (PBO) Riddle et al. (2013) (GetGoal-Duo 1) -0.7% (LIXI) vs -0.4% (PBO), Δ -0.3% (95% CI -0.5 to -0.2, p < ) +0.3 kg (LIXI) vs +1.2 kg (PBO), Δ -0.9 kg (95% CI -1.4 to -0.4, p = ) 39.9% (LIXI) vs 16.1% (PBO) Diamant et al. (2014) (4B) -1.13% (EXE) vs -1.10% (LIS TD), Δ -0.04% (95% CI to +0.11, ns) -2.5 kg (EXE) vs +2.1 kg (LIS TD), Δ -4.6 kg (95% CI -5.2 to -3.9, p < 0.001) Minor hypoglycemia (% patients) : 41% (EXE) vs 30% (LIS TD) Nocturnal hypoglycemia (% patients) : 25% (EXE) vs 27% (LIS TD) Major hypoglycemia (number of patients) : 2 (EXE) vs 7 (LIS TD) Shao et al. (2014) Rosenstock et al. (2015) (GetGoal-Duo 2) -1.42% (EXE) vs -1.31% (INS TD), ns -0.63% (LIXI) vs -0.58% (GLU QD) vs -0.84% (GLU TD) Δ -0.05% (95% CI to +0.06) vs GLU QD, Δ +0,21% (95% CI 0,10 to 0,33) vs GLU TD -7.7 kg (EXE) vs +3.3 kg (INS TID), p < ,7 kg (LIXI) vs +0.9 kg (GLU QD) vs +1.3 kg (GLU TD) Δ -1.7 kg (95% CI -2.3 to -1.1, p < ) vs GLU QD, Δ -2,0 kg (95% CI -2.6 to -1.4, p < ) vs GLU TD Symptomatic hypoglycemia : 0% (EXE) vs 10% (INS TD) 32,9% (LIXI) vs 38,9% (GLU QD) vs 44,9% (GLU TD) EXE: exenatide; LIXI: lixisenatide; PBO: placebo; MET: metformin; LIS: lispro; INS: insulin; GLU: glulisine; BD: twice daily; QD: once daily; TD, thrice daily GetGoal-Duo 1 was a randomized controlled trial examining the efficacy of lixisenatide 20 µg once daily vs placebo in 446 T2D patients previously treated by OADs whose HbA 1c levels remained elevated (7 9%) despite the use of intensified insulin glargine in combination with metformin with or without a thiazolidinedione [43]. At week 24 while insulin titration was continued, HbA 1c was reduced by 0.71% with lixisenatide vs 0.40% with placebo (mean difference: -0.32%; P < ). In addition, lixisenatide significantly reduced 2-h PPG after a standardized breakfast (placebo corrected: -3.2 mmol/l; P < ) and had a favourable effect on body weight (placebo corrected: kg; P = ). Symptomatic hypoglycaemic events were more common with lixisenatide. Presented during the 75 th ADA Scientific Sessions, GetGoal-Duo 2 was a randomized controlled trial to compare once-daily prandial GLP-1 RA (lixisenatide 20 µg) with a rapid-acting insulin analogue before the main meal (glulisine QD as basal-plus) or before each meal (glulisine TD as basal-bolus). In fact, they were add-ons to insulin glargine with or without metformin in 894 T2D patients who were previously treated with basal insulin and up to three OADs, yet whose glycaemic control remained poor (HbA 1c 7 9%) despite a 12-week period of glargine titration [44]. At week 26, lixisenatide QD was non-inferior to glulisine QD and glulisine TD in reducing HbA 1c (-0.6 ± 0.1% vs -0.6 ± 0.1% and -0.8 ± 0.1%, respectively), and statistically superior to both in reducing 2-h PPG from baseline after a standardized breakfast (-65.5 ± 10.8 mg/dl vs ± 10.7 mg/dl and ± 10.5 mg/dl, respectively) and in changing body weight (-0.6 ± 0.3 kg vs +1.0 ± 0.3 kg and +1.4 ± 0.3 kg, respectively). Changes from baseline to week 26 in glargine doses were comparable across the treatment arms (-0.7 ± 1.0 U/d vs -0.1 ± 1.0 U/d and -3.1 ± 1.0 U/d, respectively). The rate of symptomatic hypoglycaemia events was significantly reduced in the lixisenatide group vs glulisine QD (-25%) and glulisine TD (-51%) groups, as was the rate of nocturnal hypoglycaemias. As expected and as seen in all of the GetGoal studies, gastrointestinal adverse events, especially nausea and/or vomiting, were more frequent with lixisenatide.

6 S26 P. Darmon, D. Raccah / Diabetes & Metabolism 41 (2015) 6S21-6S27 4. Summary and future perspectives According to these encouraging results, adding a shortacting, prandial GLP-1 RA to the treatment of T2D patients still uncontrolled despite basal insulin and OADs now represents a promising strategy, and could constitute a worthwhile alternative to insulin intensification with basal-plus and basal-bolus regimens. Indirect comparisons derived from five randomized trials show that lixisenatide offers a safe and effective therapeutic alternative to enhanced basal insulin treatments, improving glucose control without weight gain and with less risk of hypoglycaemia than a once-daily meal-time rapid-acting insulin [45]. Although direct comparisons are as yet limited, the available data suggest that adding a prandial GLP-1 RA results in a similar or greater reduction of HbA 1c, weight loss instead of weight gain and a lower risk of hypoglycaemic episodes compared with the addition of once-daily up to thrice-daily meal-time rapid-acting insulin analogues [36,40,44]. This emerging strategy is becoming a more attractive albeit more expensive alternative in T2D patients insufficiently controlled with basal insulin when clinicians wish primarily to limit weight gain (or even induce weight loss) or reduce the risk of hypoglycaemia, or when patients are not able or not willing to deal with the complexities of multidose insulin regimens. Moreover, adding a GLP-1 RA to insulin may also be of value, as it can reduce basal insulin dose requirements, thereby reducing hypoglycaemia risk and facilitating weight management. In patients starting a GLP-1 RA in combination with basal insulin, some authors recommend an empirical 20% decrease in basal insulin doses in those with HbA 1c < 8%, as in the study by Buse et al. [37], with evaluation of the patient s responses over time to allow further insulin dose adjustments. In addition, the effect of GLP-1 RA on α-cell function would explain why this therapeutic option would be of value even in latestage T2D. Also, although insulin injections are generally perceived as worrisome by T2D patients, the subcutaneous route of GLP-1 RA administration does not appear to affect patient satisfaction [46]. In fact, a fixed-ratio formulation combining insulin glargine with lixisenatide (LixiLan) in a simple pen device is currently in development and has already shown promising results [47]. Finally, while gastrointestinal adverse events are common with GLP-1 RAs especially during initiation and titration, their combination with basal insulin is not expected to augment the expected side-effects. Nevertheless, long-term controlled studies are still needed to further examine the potential clinical benefits and costeffectiveness of such a combination, even though a low rate of hypoglycaemia may well reduce the economic burden. Also, in future, short-acting GLP-1 RAs will have to be compared with long-acting GLP-1 RAs for those indications, although once-weekly albiglutide has already been shown to be a simpler therapeutic option than thrice-daily lispro for enhancing basal insulin glargine therapy, resulting in comparable HbA 1c reductions with weight loss and lower hypoglycaemia risk [48]. The durability of the combination of GLP-1 RAs to intensify basal insulin and control PPG will also need to be assessed, given that short-acting GLP-1RAs act mainly via gastric-emptying delay whereas long-acting GLP-1 RAs work mainly through insulin secretion stimulation. Disclosure of interest Au cours des 5 dernières années, Patrice Darmon perçu des honoraires ou financements de la part des laboratoires Sanofi, Novo Nordisk, Lilly et AstraZeneca concernés par le thème de l article, pour des Participation à un congrès, Communications, Actions de formation, Travaux de recherche, Participation à des groupes d experts, Rédaction d articles ou documents, Conseil. Denis Raccah n a pas transmis ses liens d intérêts. References [1] Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care 2015;38: [2] Riddle M, Umpierrez G, DiGenio A, et al. Contributions of basal and postprandial hyperglycemia over a wide range of A1C levels before and after treatment intensification in type 2 diabetes. Diabetes Care 2011;34, [3] Holst JJ, Vilsbøll T. Combining GLP-1 receptor agonists with insulin: therapeutic rationales and clinical findings. Diabetes Obes Metab 2013;15:3-14. [4] Balena R, Hensley IE, Miller S, Barnett AH. Combination therapy with GLP-1 receptor agonists and basal insulin: a systematic review of the literature. Diabetes Obes Metab 2013;15: [5] Vora J. Combining incretin-based therapies with insulin. Realizing the potential in type 2 diabetes. Diabetes Care 2013;36 (Suppl2):S [6] Vora J, Bain SC, Damci T, et al. Incretin-based therapy in combination with basal insulin: a promising tactic for the treatment of type 2 diabetes. Diabetes Metab 2013;39:6-15. [7] Carris NW, Taylor JR, Gums JG. Combining a GLP-1 receptor agonist and basal insulin: study evidence and practical considerations. Drugs 2014;74: [8] Tibaldi J. Achieving glycemic goals with addition of incretin-based therapies to insulin in patients with type 2 diabetes mellitus. Am J Med Sci 2014;347: [9] Levin PA, Zhang Q, Mersey JH, et al. Glycemic control with insulin glargine plus insulin glulisine versus premixed insulin analogues in real-world practices: a cost-effectiveness study with a randomized pragmatic trial design. Clin Ther 2011;33: [10] Ligthelm RJ, Mouritzen U, Lynggaard H, et al. Biphasic insulin aspart given thrice daily is as efficacious as a basal-bolus insulin regimen with four daily injections: a randomised open-label parallel group four months comparison in patients with type 2 diabetes. Exp Clin Endocrinol Diabetes 2006;114: [11] Meneghini L, Mersebach H, Kumar S, Svendsen AL, Hermansen K. Comparison of 2 intensification regimens with rapid-acting insulin aspart in type 2 diabetes mellitus inadequately controlled by once-daily insulin detemir and oral antidiabetes drugs: the step-wise randomized study. Endocr Pract 2011;17: [12] Rodbard HW, Visco VE, Andersen H, Hiort LC, Shu DHW. Treatment intensification with stepwise addition of prandial insulin aspart boluses compared with full basal-bolus therapy (FullSTEP Study): A randomised, treat-to-target clinical trial. The Lancet Diabetes and Endocrinology 2014;2:30-7. [13] Miser WF, Arakaki R, Jiang H, Scism-Bacon J, Anderson PW, Fahrbach JL. Randomized, open-label, parallel-group evaluations of basal-bolus therapy versus insulin lispro premixed therapy in patients with type 2

7 P. Darmon, D. Raccah / Diabetes & Metabolism 41 (2015) 6S21-6S27 6S27 diabetes mellitus failing to achieve control with starter insulin treatment and continuing oral antihyperglycemic drugs: a noninferiority intensification substudy of the DURABLE trial. Clin Ther 2010;32: [14] Miao R, Wei W, Baser O, Xie L. Real world outcomes of adding rapid-acting insulin versus switching to analog premix insulin among US patients with type 2 diabetes treated with insulin glargine. Patient Prefer Adherence 2013;7: [15] Rayman G, Profozic V, Middle M. Insulin glulisine imparts effective glycaemic control in patients with Type 2 diabetes. Diabetes Res Clin Pract 2007;76: [16] Liebl A, Prager R, Binz K, Kaiser M, Bergenstal R, Gallwitz B. Comparison of insulin analogue regimens in people with type 2 diabetes mellitus in the PREFER Study: a randomized controlled trial. Diabetes Obes Metab 2009;11: [17] Blak BT, Smith HT, Hards M, Curtis BH, Ivanyi T. Optimization of insulin therapy in patients with type 2 diabetes mellitus: beyond basal insulin. Diabet Med 2012;29:e [18] Dailey G, Rosenstock J, Moses RG, Ways K. Insulin glulisine provides improved glycemic control in patients with type 2 diabetes. Diabetes Care 2004;27: [19] Edelman SV, Liu R, Johnson J, Glass LC. AUTONOMY: The First Randomized Trial Comparing Two Patient-Driven Approaches to Initiate and Titrate Prandial Insulin Lispro in Type 2 Diabetes. Diabetes Care 2014;37: [20] Rosenstock J, Ahmann AJ, Colon G, Scism-Bacon J, Jiang H, Martin S. Advancing insulin therapy in type 2 diabetes previously treated with glargine plus oral agents: prandial premixed (insulin lispro protamine suspension/lispro) versus basal/bolus (glargine/lispro) therapy. Diabetes Care 2008;31:20-5. [21] Raccah D, Haak TJ, Huet D, et al. Comparison of stepwise addition of prandial insulin to a basal-bolus regimen when basal insulin is insufficient for glycaemic control in type 2 diabetes: results of the OSIRIS study. Diabetes Metab 2012;38: [22] Dailey G, Lin J, Wang E. Efficacy and safety of a basal-bolus regimen of basal insulin plus insulin glulisine in patients with type 2 diabetes: A pooled analysis of 5 clinical trials. Diabetes 2012;61:A284. [23] Raccah D, Bretzel RG, Owens D, Riddle M. When basal insulin therapy in type 2 diabetes mellitus is not enough: what next? Diabetes Metab Res Rev 2007;23: [24] Tinahones FJ, Gross JL, Onaca A, Cleall S, Rodriguez A. Insulin lispro low mixture twice daily versus basal insulin glargine once daily and prandial insulin lispro once daily in patients with type 2 diabetes requiring insulin intensification: a randomized phase IV trial. Diabetes Obes Metab 2014;16: [25] Del Prato S, Lankisch M, Lin J, Wang E, Owens DR. Efficacy and safety of a basal-plus regimen of insulin glargine plus insulin glulisine in patients with type 2 diabetes: A pooled analysis of 4 clinical trials. Diabetes 2012;61:A297. [26] Lankisch MR, Owens DR, Wang E, Lin J, Del Prato S. Basal-plus regimen of insulins glargine and glulisine is weight neutral across BMI groups. Diabetes 2012;61:A [27] Owens DR, Luzio SD, Sert-Langeron C, Riddle MC. Effects of initiation and titration of a single pre-prandial dose of insulin glulisine while continuing titrated insulin glargine in type 2 diabetes: a 6-month proof-of-concept study. Diabetes Obes Metab 2011;13: [28] Owens DR, Del Prato S, Lin J, Wang E, Lankisch M. Impact of diabetes duration on efficacy and safety of a basal-plus regimen of insulin glargine plus once daily insulin glulisine: A Pooled analysis from 4 clinical trials. Diabetes 2012;61:A599. [29] Riddle MC, Rosenstock J, Vlajnic A, Gao L. 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: [30] Raccah D, Lin J, Bahloul A, Wang E, Penfornis A. Meta-analysis of efficacy and safety outcomes associated with the basal-plus regimen of insulin glargine and insulin glulisine among type 2 diabetes patients with HbA1c >8%. Diabetes 2012;61:A620. [31] Lankisch MR, Ferlinz KC, Leahy JL, Scherbaum WA. Introducing a simplified approach to insulin therapy in type 2 diabetes: a comparison of two single-dose regimens of insulin glulisine plus insulin glargine and oral antidiabetic drugs. Diabetes Obes Metab 2008;10: [32] Peyrot M, Barnett AH, Meneghini LF, Schumm-Draeger PM. Factors associated with injection omission/non-adherence in the Global Attitudes of Patients and Physicians in Insulin Therapy study. Diabetes Obes Metab 2012;14: [33] Ceriello A, Novials A, Canivell S et al. Simultaneous GLP-1 and insulin administration acutely enhances their vasodilatory, antiinflammatory, and antioxidant action in type 2 diabetes. Diabetes Care 2014;37: [34] Eng C, Kramer CK, Zinman B, Retnakaran R. Glucagon-like peptide-1 receptor agonist and basal insulin combination treatment for the management of type 2 diabetes: a systematic review and meta-analysis. Lancet 2014;384: [35] Meier JJ, Rosenstock J, Hincelin-Méry A et al. Contrasting effects of lixisenatide and liraglutide on postprandial glycemic control, gastric emptying, and safety parameters in patients with type 2 diabetes on optimized insulin glargine with or without metformin: a randomized, open-label trial. Diabetes Care 2015;38: [36] Shao N, Kuang HY, Hao M, Gao XY, Lin WJ, Zou W. Effects of exenatide on obesity and NAFLD with elevated liver enzymes in patients with type 2 diabetes. Diabetes Metab Res Rev 2014;30: [37] Buse JB, Bergenstal RM, Glass LC et al. Use of twice-daily exenatide in basal insulin-treated patients with type 2 diabetes. A randomized, controlled trial. Ann Intern Med 2011;154: [38] Rosenstock J, Shenouda SK, Bergenstal RM et al. Baseline factors associated with glycemic control and weight loss when exenatide twice daily is added to optimized insulin glargine in patients with type 2 diabetes. Diabetes Care 2012;35: [39] Buse JB, Han J, Miller S et al. Addition of exenatide BID to insulin glargine: a post-hoc analysis of the effect on glycemia and weight across a range of insulin titration. Curr Med Res Opin 2014;30: [40] Diamant M, Nauck MA, Shaginian R, et al; 4B Study Group. Glucagonlike peptide 1 receptor agonist or bolus insulin with optimized basal insulin in type 2 diabetes. Diabetes Care 2014;37: [41] Riddle MC, Aronson R, Home P 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: [42] Seino Y, Min KW, Niemoeller E, Takami A. 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: [43] Riddle MC, Forst T, Aronson R 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: [44] Rosenstock J, Hanefeld M, Gentile S et al, on behalf of the GetGoal- Duo2 Study investigators. Advancing basal insulin glargine with prandial lixisenatide QD vs insulin glulisine QD or TID in T2DM: the GetGoal-Duo2 evidence-based trial (NCT ). Diabetes 2015;64 (Suppl1A):LB27-8. [45] Raccah D, Lin J, Wang E et al. Once-daily prandial lixisenatide versus once-daily rapid-acting insulin in patients with type 2 diabetes mellitus insufficiently controlled with basal insulin: analysis of data from five randomized, controlled trials. J Diabetes Complications 2014;28:40-4. [46] Davies M, Pratley R, Hammer M et al. Liraglutide improves treatment satisfaction in people with type 2 diabetes compared with sitagliptin, each as an add on to metformin. Diabet Med 2011;28: [47] Rosenstock J DM, Silvestre L, Souhami E, et al. Benefits of a fixedratio formulation of once-daily insulin glargine/lixenatide (LixiLan) vs glargine in type 2 diabetes (T2DM) inadequately controlled on metformin. Diabetes 2014;63 (Suppl 1):A87-8. [48] Rosenstock J, Fonseca VA, Gross JL et al. Advancing basal insulin replacement in type 2 diabetes inadequately controlled with insulin glargine plus oral agents: a comparison of adding albiglutide, a weekly GLP-1 receptor agonist, versus thrice-daily prandial insulin lispro. Diabetes Care 2014;37: