DIABETICMedicine DOI: 10.1111/dme.12814

Research: Treatment Empagliflozin as add-on to metformin in people with Type 2 diabetes L. Merker1*, H-U. H€ aring2*, A. V. Christiansen3, F. Roux4, A. Salsali5, G. Kim6, T. Meinicke7, H. J. Woerle6 and U. C. Broedl6 on behalf of the EMPA-REG EXTENDTM MET investigators €bingen, Tu €bingen, Germany, 3Boehringer Ingelheim Danmark A/S, Copenhagen, Denmark, Diabetes- und Nierenzentrum, Dormagen, Germany, 2University of Tu Boehringer Ingelheim France S.A.S., Reims, France, 5Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT, USA, 6Boehringer Ingelheim Pharma GmbH & Co. KG, Ingelheim, Germany and 7Boehringer Ingelheim Pharma GmbH & Co. KG, Biberach, Germany 1 4

Accepted 26 May 2015

Abstract Aims To investigate the long-term efficacy and safety of empagliflozin as add-on to metformin in people with Type 2 diabetes. Methods Of 637 participants treated with empagliflozin 10 mg, empagliflozin 25 mg, or placebo once daily for 24 weeks, 463 (72.7%) were treated in a double-blind extension trial for ≥ 52 weeks. Prespecified exploratory endpoints included changes from baseline in HbA1c, weight and blood pressure at week 76.

Compared with placebo, adjusted mean changes from baseline in HbA1c (overall baseline mean  SD 63  9 mmol/mol [7.9  0.9%]) were 7 mmol/mol [( 0.6%) 95% CI 8, 5 mmol/mol ( 0.8, 0.5%); P < 0.001] and 8 mmol/mol [( 0.7%) 95% CI 10, 6 mmol/mol ( 0.9, 0.6%); P < 0.001], for empagliflozin 10 mg and 25 mg, respectively. Compared with placebo, adjusted mean changes from baseline in weight were 1.9 kg (95% CI 2.5, 1.3; P < 0.001) and 2.2 kg (95% CI 2.8, 1.6; P < 0.001) for empagliflozin 10 mg and 25 mg, respectively. Empagliflozin led to sustained reductions in systolic blood pressure vs. placebo. Adverse events were reported in 77.7, 80.2 and 72.0% of participants on placebo, empagliflozin 10 mg and empagliflozin 25 mg, respectively. Confirmed hypoglycaemic adverse events (glucose ≤ 3.9 mmol/l and/or event requiring assistance) were reported in 3.4, 4.1 and 4.2% of participants in these groups, respectively. Results

Conclusions In people with Type 2 diabetes, empagliflozin 10 mg and 25 mg given as add-on to metformin for 76 weeks were well tolerated and led to sustained reductions in HbA1c, weight and systolic blood pressure.

Diabet. Med. 32, 1555–1567 (2015)

Introduction Empagliflozin is a potent and selective inhibitor of sodiumglucose cotransporter 2 (SGLT2) [1]. SGLT2 is the transporter that is primarily responsible for the reabsorption of the glucose filtered by the kidney [2]. In people with Type 2 diabetes, empagliflozin reduces renal glucose reabsorption and thus increases urinary glucose excretion, leading to a reduction in plasma glucose with a low risk of hypoglycaemia [3]. Correspondence to: Ludwig Merker E-mail: [email protected] *These authors contributed equally to this work. An abstract reporting some of the data presented in this manuscript was presented at the annual meeting of the American Diabetes Association (ADA) in San Francisco in June 2014 (Merker L, et al. Empagliflozin (EMPA) for ≥ 76 weeks as add-on to metformin in patients with Type 2 diabetes (T2DM). Diabetes 2014; 63 (suppl 1): A278 [1074-P]).

ª 2015 The Authors. Diabetic Medicine ª 2015 Diabetes UK

In phase III trials in people with Type 2 diabetes, empagliflozin given as monotherapy or as add-on therapy has been shown to improve glycaemic control and reduce body weight and blood pressure [4–10]. In addition, in these trials, empagliflozin was well tolerated, although a higher proportion of people receiving empagliflozin than placebo reported events consistent with genital infection. Metformin is recommended as first-line pharmacological therapy for people with Type 2 diabetes in whom lifestyle intervention has not achieved, or is unlikely to achieve, glycaemic control [11]; however, as diabetes progresses, metformin often fails to maintain glycaemic control and additional antidiabetes agents are required [11,12]. In the phase III EMPA-REG METTM trial, empagliflozin 10 mg and 25 mg as add-on to metformin for 24 weeks improved glycaemic control, body weight and blood pressure vs. placebo [6]. The present paper describes a 52-week extension

1555

DIABETICMedicine

Empagliflozin as add-on to metformin in people with Type 2 diabetes  L. Merker et al.

What’s new? • The data in this study show that the improvements in glycaemic control, weight and systolic blood pressure achieved with empagliflozin as add-on to metformin after 24 weeks’ treatment were sustained at week 76 and that empagliflozin was well tolerated over 76 weeks of treatment. • Empagliflozin plus metformin may provide an attractive treatment option for people with Type 2 diabetes unable to achieve glycaemic control with metformin alone. of this study, EMPA-REG EXTENDTM MET, which was undertaken to evaluate the long-term safety, tolerability and efficacy of empagliflozin 10 mg and 25 mg compared with placebo as add-on to metformin in people with Type 2 diabetes.

Participants and methods In the initial randomized, placebo-controlled, double-blind phase III trial, adults with Type 2 diabetes and insufficient glycaemic control [HbA1c ≥ 53 mmol/mol (≥ 7%) and ≤ 86 mmol/mol (≤ 10%)], despite being on a diet and exercise programme and receiving a stable regimen of immediaterelease metformin (≥ 1500 mg/day or maximum dose according to the local label; unchanged for ≥ 12 weeks before randomization), who had a body mass index (BMI) ≤ 45 kg/m2 were enrolled [6]. Exclusion criteria included uncontrolled hyperglycaemia [glucose level > 13.3 mmol/l after an overnight fast, confirmed by a second measurement], indication of liver disease and impaired renal function [estimated glomerular filtration rate (GFR) < 30 ml/min/1.73 m2] [6]. Participants were randomized (1:1:1) to receive empagliflozin 10 mg, empagliflozin 25 mg or placebo once daily for 24 weeks. Participants who completed 24 weeks’ treatment, who still did not contravene the exclusion criteria for the initial study and who had not developed contraindications to metformin according to the local label during the initial study, could elect to continue double-blind treatment for ≥ 52 weeks in an extension trial. Participants were required to re-confirm their consent before starting the extension trial. Participants remained on the treatments they received in the initial 24-week study, without re-randomization, and remained in the trial until the last participant entered had been treated for 76 weeks. The initial trial and the extension trial were registered with clinicaltrials.gov (NCT01159600 and NCT01289990, respectively) and were approved by the institutional review boards and independent ethics committees and competent authorities of the participating centres, and complied with the Declaration of Helsinki in accordance with the International Conference

1556

on Harmonisation Harmonised Tripartite Guideline for Good Clinical Practice. Participants who received antidiabetes rescue medication during the initial 24-week trial, and were still taking it at enrolment in the extension trial, continued to take it. Rescue medication could be initiated during the extension trial if a participant had a glucose level > 10.0 mmol/l after an overnight fast or HbA1c > 64 mmol/mol (> 8%). The choice and dosage of rescue medication were at the discretion of the investigator, according to local prescribing information. In cases of hypoglycaemia, dose reduction or discontinuation of rescue medication was to be initiated. Where hyper- or hypoglycaemia could not be controlled, the participant was to be discontinued from the trial.

Endpoints and analyses

No primary endpoint was defined for the present study, as the primary efficacy endpoint, change from baseline in HbA1c, was analysed at week 24 in the initial trial [6]. Prespecified exploratory efficacy endpoints in the extension trial were change from baseline at week 76 in HbA1c, fasting plasma glucose, body weight, systolic blood pressure, diastolic blood pressure and use of rescue therapy over 76 weeks. The proportion of participants who were treated in the extension trial and had HbA1c ≥ 53 mmol/mol (≥ 7%) at baseline who achieved HbA1c < 53 mmol/mol (< 7%) at week 76 was also analysed as an exploratory endpoint. These endpoints were also analysed at week 52. Baseline was defined as baseline in the initial trial, i.e. before the administration of the study drug. No formal sample size calculation was performed as the extension trial was open to all participants of the initial trial who were eligible to participate. The full analysis set was defined as participants who received ≥ 1 dose of study drug and had a baseline HbA1c measurement in the initial study. Treatment differences versus placebo were assessed using an ANCOVA model in the full analysis set with baseline HbA1c and the baseline value of the endpoint in question as linear covariates, and baseline estimated GFR, region and treatment as fixed effects. All values observed after a participant started antidiabetes rescue therapy were set to missing. A last observation carried forward approach was used to impute missing data. Sensitivity analyses were undertaken for changes over time in HbA1c, fasting plasma glucose, body weight and blood pressure. A restricted maximum likelihood-based mixed model repeated measures approach was used in the full analysis set and in full analysis set-completers using observed cases. Full analysis set-completers comprised participants who completed 76  1 weeks’ treatment and had an HbA1c value during this period. These mixed model repeated measures analyses included baseline HbA1c and the baseline value of the endpoint in question as linear covariates, and baseline estimated GFR, region, treatment, visit and visit by

ª 2015 The Authors. Diabetic Medicine ª 2015 Diabetes UK

Research article

DIABETICMedicine

treatment interaction as fixed effects. The use of rescue therapy was assessed using logistic regression, including treatment as a factor and baseline HbA1c as a covariate, in the full analysis set. Categorical change in HbA1c was

analysed using logistic regression in participants from the full analysis set who were treated in the extension trial with noncompleters considered failure imputation (participants from the initial trial who did not enter the extension trial were not

Table 1 Demographics and baseline characteristics Placebo n = 207 Male, n (%) 116 (56.0) Age, years 56.0 (9.7) Race, n (%) White 113 (54.6) Asian 92 (44.4) Black/African-American 2 (1.0) American-Indian/Alaska Native 0 Time since diagnosis of Type 2 diabetes, n (%) ≤ 1 year 19 (9.2) > 1 to 5 years 83 (40.1) > 5 to 10 years 65 (31.4) > 10 years 40 (19.3) Body weight, kg 79.7 (18.6) 2 BMI, kg/m 28.7 (5.2) HbA1c, mmol/mol 63 (10) HbA1c, % 7.9 (0.9) Fasting plasma glucose, mmol/l 8.7 (1.8) Systolic blood pressure, mmHg 128.6 (14.7) Diastolic blood pressure, mmHg 78.1 (7.9) Estimated GFR, ml/min/1.73 m2 89.7 (21.4) (Modification of Diet in Renal Disease)

Empagliflozin 10 mg n = 217

Empagliflozin 25 mg n = 213

Total N = 637

125 (57.6) 55.5 (9.9)

120 (56.3) 55.6 (10.2)

361 (56.7) 55.7 (9.9)

112 99 4 2

(51.6) (45.6) (1.8) (0.9)

113 (53.1) 98 (46.0) 0 2 (0.9)

338 289 6 4

(53.1) (45.4) (0.9) (0.6)

20 78 68 51 81.6 29.1 63 7.9 8.6 129.6 79.6 89.5

(9.2) (35.9) (31.3) (23.5) (18.5) (5.5) (9) (0.8) (2.0) (14.1) (8.0) (19.6)

58 230 207 142 81.2 29.2 63 7.9 8.5 129.4 78.7 89.0

(9.1) (36.1) (32.5) (22.3) (18.8) (5.5) (9) (0.9) (1.8) (14.6) (8.1) (20.1)

19 69 74 51 82.2 29.7 62 7.9 8.3 130.0 78.4 87.7

(8.9) (32.4) (34.7) (23.9) (19.3) (5.7) (10) (0.9) (1.7) (15.1) (8.4) (19.3)

Data are mean (SD) unless otherwise indicated. Data are for the full analysis set. BMI, body mass index; GFR, glomerular filtration rate.

FIGURE 1 Study flow. The full analysis set comprised participants who received ≥ 1 dose of study drug and had a baseline HbA1c measurement in the initial study (n = 207 in the placebo group, n = 217 in the empagliflozin 10 mg group, n = 213 in the empagliflozin 25 mg group).

ª 2015 The Authors. Diabetic Medicine ª 2015 Diabetes UK

1557

DIABETICMedicine

Empagliflozin as add-on to metformin in people with Type 2 diabetes  L. Merker et al.

Table 2 Summary of efficacy results at week 76 Placebo n = 207 HbA1c at week 76, mmol/mol HbA1c at week 76, % Change from baseline, mmol/mol Change from baseline, % Difference vs. placebo, mmol/mol (95% CI) Difference vs. placebo, % (95% CI) P Participants with HbA1c ≥ 53 mmol/mol (≥ 7%) at baseline who reached HbA1c < 53 mmol/mol (< 7%) at week 76*, n (%) Odds ratio vs. placebo (95% CI) P Fasting plasma glucose at week 76, mmol/l Change from baseline Difference vs. placebo (95% CI) P Body weight at week 76, kg Change from baseline Difference vs. placebo (95% CI) P Systolic blood pressure at week 76, mmHg Change from baseline Difference vs. placebo (95% CI) P Diastolic blood pressure at week 76, mmHg Change from baseline Difference vs. placebo (95% CI) P

63 7.9 0 0.0

(1) (0.1) (1) (0.1)

Empagliflozin 10 mg n = 217 56 7.3 7 0.6 7

(1) (0.1) (1) (0.1) ( 8,

55 7.2 8 0.7 8 5)

0.6

17 (14.3)

( 0.8, < 0.001 44 (27.8)

0.5)

128.6 (0.8) 0.8 (0.8)

78.2 (0.5) 0.5 (0.5)

( 10,

6)

( 0.9, < 0.001 58 (42.3)

0.6)

4.62 (1.29, 4.56)

80.7 (0.2) 0.5 (0.2)

(1) (0.1) (1) (0.1)

0.7

2.42

9.1 (0.1) 0.6 (0.1)

Empagliflozin 25 mg n = 213

0.006 7.7 (0.1) 0.8 (0.1) 1.4 ( 1.7, < 0.001 78.8 (0.2) 2.4 (0.2) 1.9 ( 2.5, < 0.001 124.1 (0.8) 5.2 (0.8) 4.4 ( 6.6, < 0.001 76.2 (0.5) 2.5 (0.5) 2.0 ( 3.4, 0.008

1.1)

1.3)

2.3)

0.5)

(2.46, 8.67) < 0.001 7.3 (0.1) 1.2 (0.1) 1.7 ( 2.0, 1.4) < 0.001 78.5 (0.2) 2.7 (0.2) 2.2 ( 2.8, 1.6) < 0.001 124.8 (0.8) 4.5 (0.8) 3.7 ( 5.9, 1.5) < 0.001 76.8 (0.5) 1.9 (0.5) 1.4 ( 2.8, 0.1) 0.068

Data are adjusted mean (SE) based on ANCOVA in the full analysis set (last observation carried forward) unless otherwise indicated. Statistical analyses were performed on HbA1c% data only. * Logistic regression analysis in participants from the full analysis set who were treated in the extension trial and had HbA1c ≥ 53 mmol/mol (≥ 7%) at baseline (n = 119 for placebo, n = 158 for empagliflozin 10 mg, n = 137 for empagliflozin 25 mg). Non-completers were considered failures.

considered non-completers). The model included treatment, baseline estimated GFR, region and HbA1c. Safety was assessed via reporting of adverse events (preferred terms coded according to the Medical Dictionary for Drug Regulatory Activities version 16.0) over the whole study period. Confirmed hypoglycaemic adverse events were defined as hypoglycaemic adverse events with plasma glucose ≤ 3.9 mmol/l and/or requiring assistance. Events consistent with urinary tract infection, genital infection and

volume depletion were identified using prospectively defined search categories based on 77, 89 and 8 preferred terms, respectively. Changes from baseline in clinical laboratory values were assessed. Safety analyses were performed on the treated set (participants treated with ≥ 1 dose of study drug in the initial 24-week study) and were descriptive in nature, except for the analysis of lipid measurements, which was conducted using an ANCOVA with the respective baseline value and baseline HbA1c as linear

FIGURE 2 Effect of empagliflozin on glycaemic measurements. (a) Change from baseline in HbA1c at week 76 (ANCOVA, full analysis set using last observation carried forward to impute missing data; statistical analyses were performed on HbA1c% data only); (b) HbA1c over time (analysed using mixed model repeated measures in the full analysis set using observed cases); (c) percentage of participants with HbA1c ≥ 53 mmol/mol (≥ 7%) at baseline who reached HbA1c < 53 mmol/mol (< 7%) at week 76 (logistic regression in participants from the full analysis set treated in the extension trial using non-completers considered failures approach; P-value for odds ratio); (d) change from baseline in fasting plasma glucose at week 24 (ANCOVA, full analysis set, last observation carried forward) and (e) change in fasting plasma glucose over time (mixed model repeated measures, full analysis set, observed cases). Data are adjusted mean (SE) or % participants. Full analysis set comprised participants who received ≥ 1 dose of study drug and had a baseline HbA1c measurement in the initial study. *P < 0.001 vs. placebo; †P = 0.006 vs. placebo. FPG, fasting plasma glucose.

1558

ª 2015 The Authors. Diabetic Medicine ª 2015 Diabetes UK

Research article

DIABETICMedicine

Baseline mean:

(a)

63 mmol/mol (7.9%)

63 mmol/mol (7.9%)

62 mmol/mol (7.9%)

0.2

Adjusted mean (SE) change from baseline in HbA1c (%)

0.0

0 0 mmol/mol (0.0%)

–0.2

–2

–0.4

–4

–0.6

–7 –7 mmol/mol (–0.6%)

–0.8

* –1.0

–9

–8 mmol/mol (–0.7%)

*

Placebo (n=207) Empagliflozin 10 mg (n=217) Empagliflozin 25 mg (n=213)

Adjusted mean (SE) change from baseline in HbA1c (mmol/mol)

2

–11

Week

Adjusted mean (SE) HbA1c (%)

0

6

12

18

24

52

41

64

76

8.2

66

8.0

64

7.8

62 60

7.6 Placebo Empagliflozin 10 mg

7.4

57

Empagliflozin 25 mg 7.2

55

7.0

53

0.0

0

Adjusted mean (SE) HbA1c (mmol/mol)

(b)

Number of participants analyzed Placebo

199

195

190

173

156

96

86

74

70

Empagliflozin 10 mg

216

213

212

201

198

153

144

138

130

Empagliflozin 25 mg

205

204

197

191

186

135

130

122

118

*

42.3

45

8.7

8.6

8.3

1.0 Placebo (n=207)

0.6

35



27.8

30

Placebo (n=119)

25

Empagliflozin 10 mg (n=158)

20

Empagliflozin 25 mg (n=137) 14.3

15 10

Adjusted mean (SE) change from baseline in FPG (mmol/l)

40

1c

1c

Baseline mean:

(d) who had HbA

Empagliflozin as add-on to metformin in people with Type 2 diabetes.

To investigate the long-term efficacy and safety of empagliflozin as add-on to metformin in people with Type 2 diabetes...
482KB Sizes 0 Downloads 6 Views