14 CLINICAL STUDIES

14.1 Overview of Clinical Studies

A total of 3908 patients with type 2 diabetes mellitus participated in 6 randomized, parallel and active or placebo-controlled phase 3 trials of 26 weeks duration.

Three studies were conducted in patients inadequately controlled on one or more OADs (e.g. metformin, pioglitazone, sulfonylurea, or sodium-glucose cotransporter-2 inhibitors (SGLT2i) (Tables 5-7). Three studies were conducted in patients converting from liraglutide or basal insulin: one study was conducted in patients converting from liraglutide (with doses up to 1.8 mg), (Table 8), one study was conducted in patients converting from any basal insulin (Table 9), and one study was conducted in patients converting from insulin glargine U-100 (Table 10).

In all trials, XULTOPHY 100/3.6 was titrated twice weekly by increments or decrements of 2 units (2 units insulin degludec/0.072 mg liraglutide), towards a pre-specified fasting blood glucose target. The same titration algorithm was applied for basal insulin comparators.

14.2 Patients with Type 2 Diabetes Mellitus Uncontrolled on OAD Treatment

NCT01336023: The efficacy and safety of XULTOPHY 100/3.6 compared to insulin degludec and liraglutide, all administered once-daily, was studied in a 26-week randomized, open-label, three-arm parallel trial in 1660 patients with type 2 diabetes mellitus inadequately controlled on 1-2 OADs (metformin or metformin with or without pioglitazone).

The mean age of the trial population was 55 years and mean duration of diabetes was 7 years. 51% were male. 62% were White, 7% were Black or African American and 15% were Hispanic or Latino ethnicity, 5% of patients had eGFR < 60mL/min/1.73m2; no patients had eGFR < 30mL/min/1.73m2. The mean BMI was 31.2 kg/m2.

The starting dose of XULTOPHY 100/3.6 was 10 units (10 units insulin degludec/0.36 mg liraglutide). The starting dose of insulin degludec was 10 units. XULTOPHY 100/3.6 and insulin degludec were titrated twice weekly towards a target fasting blood glucose goal of 72-90 mg/dL. Patients in the liraglutide arm followed a fixed dose escalation scheme with a starting dose of 0.6 mg and a dose increase of 0.6 mg weekly until a daily dose of 1.8 mg was reached. Patients continued on pre-trial treatment with metformin or metformin and pioglitazone throughout the trial.

At the end of 26 weeks, treatment with XULTOPHY 100/3.6, insulin degludec, and liraglutide resulted in a reduction in HbA1c from baseline of 1.81%, 1.35% and 1.21%, respectively, (see Table 5). The end of trial dose of XULTOPHY 100/3.6 was 38 units (38 units insulin degludec/1.37 mg liraglutide).

Table 5: Results of a 26-Week Trial with XULTOPHY 100/3.6 in Patients with Type 2 Diabetes Mellitus Inadequately Controlled on Metformin Alone or in Combination with Pioglitazone

	XULTOPHY 100/3.6 + metformin ± pioglitazone	Insulin degludec + metformin ± pioglitazone	Liraglutide + metformin ± pioglitazone
Total (N)	833	413	414
HbA1c (%)
Baseline	8.3	8.3	8.3
End of Trial (LS Mean)#	6.5	6.9	7.1
Change from baseline (LS Mean)#	-1.81	-1.35	-1.21
Estimated treatment difference [95% CI]#	--	-0.46% [-0.59; -0.34]A	-0.60% [-0.72; -0.47]A
Percentage of patients achieving HbA1c <7%##	74.1%	60.5%	56.0%
Fasting Plasma Glucose (FPG) (mg/dL)
Baseline	166	169	163
End of Trial (LS Mean)#	104	107	133
Change from baseline (LS Mean)#	-61.8	-59.2	-32.4

Ap<0.01. Primary endpoint was tested for non-inferiority of XULTOPHY 100/3.6 to insulin degludec based on pre-specified non-inferiority margin of 0.3% and for superiority of XULTOPHY 100/3.6 to liraglutide.

#Estimated using an ANCOVA with treatment, baseline HbA1c stratum, sub-study, concomitant diabetes treatment and country as factors and baseline response as covariate. Multiple imputation modelled “return to baseline” of the treatment effect for subjects having missing week 26 data.

##Patients with missing HbA1c value at week 26 data were considered non-responders.

There were 11.8% of subjects in the XULTOPHY 100/3.6 arm, 12.3% in the insulin degludec arm and 15.5% in the liraglutide arm for whom HbA1c data was missing at week 26.

NCT01618162: The efficacy and safety of XULTOPHY 100/3.6 compared to placebo were studied in a 26-week randomized, double-blind, treat-to-target trial in 435 patients with type 2 diabetes mellitus inadequately controlled on sulfonylurea alone or in combination with metformin.

The mean age of the trial population was 60 years, and mean duration of diabetes was 9 years. 52% were male. 75% were White, 7% were Black or African American and 9% were Hispanic or Latino ethnicity, 11% of patients had eGFR < 60mL/min/1.73m2; no patients had eGFR < 30mL/min/1.73m2. The mean BMI was 31.5 kg/m2.

XULTOPHY 100/3.6 was started at 10 units (10 units insulin degludec/0.36 mg liraglutide) and titrated twice weekly towards a target fasting blood glucose goal of 72-108 mg/dL. Patients continued on pre-trial treatment with sulfonylurea, with or without metformin throughout the trial.

Treatment with XULTOPHY 100/3.6 for 26 weeks resulted in a statistically significant reduction in mean HbA1c compared to placebo (see Table 6). The end of trial dose of XULTOPHY 100/3.6 was 28 units (28 units insulin degludec/1.01 mg liraglutide).

Table 6: Results of a 26-Week Trial with XULTOPHY 100/3.6 in Patients with Type 2 Diabetes Mellitus Inadequately Controlled on Sulfonylurea Alone or in Combination with Metformin

	XULTOPHY 100/3.6 + sulfonylurea### ± metformin	Placebo + sulfonylurea### ± metformin
Total (N)	289	146
HbA1c (%)
Baseline	7.9	7.9
End of Trial (LS Mean)#	6.5	7.3
Change from baseline (LS Mean)#	-1.42	-0.62
Estimated treatment difference [95% CI]#	-0.81% [-0.98; -0.63]A
Percentage of patients achieving HbA1c <7%##	70.9%	26.7%
FPG (mg/dL)
Baseline	164	165
End of Trial (LS Mean)#	118	152
Change from baseline (LS Mean)#	-46.2	-12.1

Ap<0.01. Primary endpoint was tested for superiority of XULTOPHY 100/3.6 to placebo.

#Estimated using an ANCOVA with treatment, region and pre-trial medication as fixed factors and baseline response as covariate. Multiple imputation modelled “jump-to-control” of the treatment effect for subjects having missing week 26 data.

##Patients with missing HbA1c value at week 26 data were considered non-responders.

There were 12.8% of subjects in the XULTOPHY 100/3.6 arm and 24.7% in the placebo arm for whom HbA1c data was missing at week 26.

###Dose of sulfonylurea was ≥ half of the maximum approved dose.

NCT02773368: The efficacy and safety of XULTOPHY 100/3.6 compared to insulin glargine U-100, both administered once-daily, were studied in a 26-week randomized, open-label, two-arm parallel trial in 420 patients with type 2 diabetes mellitus inadequately controlled on a SGLT2i alone or in combination with other OADs (with or without metformin, pioglitazone, and/or dipeptidyl peptidase-4 [DPP4] inhibitor). At randomization, the DPP4 inhibitor was discontinued.

The mean age of the trial population was 57 years and mean duration of diabetes was 10 years. 59% were male. 82% were White, 1% were Black or African American and 16% were Hispanic or Latino ethnicity, 3% of patients had eGFR < 60mL/min/1.73m2; none of the patients had eGFR < 30mL/min/1.73m2. The mean BMI was 31.2 kg/m2.

The starting dose of XULTOPHY 100/3.6 was 10 units (10 units insulin degludec/0.36 mg liraglutide). The starting dose of insulin glargine U-100 was 10 units. XULTOPHY 100/3.6 and insulin glargine U-100 were titrated twice weekly to target a fasting blood glucose goal of 72-90 mg/dL. Patients could not increase the dose of XULTOPHY 100/3.6 and insulin glargine U-100 by more than 4 units per week, and there was no maximum allowed dose of insulin glargine. The patients continued on pre-trial treatment with SGLT2i, with or without other OADs throughout the entire trial. The targeted fasting blood glucose goal was achieved by 49.0% of patients randomized to XULTOPHY 100/3.6 and 41.9% of patients randomized to insulin glargine at 26 weeks.

At the end of 26 weeks, XULTOPHY 100/3.6 resulted in a reduction in HbA1c from baseline of 1.97% and insulin glargine U-100 resulted in a reduction of 1.59% (see Table 7). At the end of trial, the average dose of XULTOPHY 100/3.6 was 36 units (36 units insulin degludec/1.3 mg liraglutide) and the dose of insulin glargine was 54 units; it is unclear that these observed differences in insulin doses are clinically important. The difference in HbA1c effect observed at 26 weeks may not necessarily reflect the effect in the care setting where insulin glargine may be more rapidly titrated.

Table 7: Results of a 26-Week Trial in Patients with Type 2 Diabetes Mellitus Inadequately

Controlled on SGLT2i Alone or in Combination with Metformin, Pioglitazone and/or DPP4 Inhibitor

	XULTOPHY 100/3.6 + SGLT2i ± metformin± pioglitazone	Insulin Glargine U-100 + SGLT2i ± metformin± pioglitazone
Total (N)	210	210
HbA1c (%)
Baseline	8.2	8.4
End of Trial (LS Mean)#	6.3	6.7
Change from baseline (LS Mean)#	-1.97	-1.59
Estimated treatment difference [95% CI]#	-0.38% [-0.54; -0.23]A
Percentage of patients achieving HbA1c <7%##	79.5%	68.6%
FPG (mg/dL)
Baseline	171	172
End of Trial (LS Mean)###	108	112
Change from baseline (LS Mean)###	-63.8	-59.9

APrimary endpoint was tested for non-inferiority of XULTOPHY 100/3.6 to insulin glargine U-100 on a non-inferiority margin of 0.3%.

#Estimated using an ANCOVA with treatment, pre-trial OAD group and region as factors and corresponding baseline value as covariate. For HbA1c (%), missing week 26 measurements from subjects who discontinued early were multiply-imputed using information from subjects who also discontinued early but still had a week 26 measurement (retrieved dropouts).

##Patients with missing HbA1c value at week 26 data were considered non-responders.

There were 6.2% of subjects in the XULTOPHY 100/3.6 arm and 2.4% in the insulin glargine U-100 arm for whom HbA1c data was missing at week 26.

###Missing week 26 FPG measurements from subjects who discontinued early were imputed using multiple imputation with a mean for each subject equal to their respective baseline value.

14.3 Patients currently on basal insulin or GLP-1 receptor agonist

Converting to XULTOPHY 100/3.6 from GLP-1 receptor agonist

NCT01676116: The efficacy and safety of XULTOPHY 100/3.6 (once-daily) compared to unchanged pre-trial liraglutide up to a dose of 1.8 mg daily, were studied in a 26-week randomized, open-label, treat-to-target (fasting blood glucose goal of 72-90 mg/dL) trial. The trial included 348 patients with type 2 diabetes mellitus inadequately controlled on liraglutide and metformin alone or in combination with pioglitazone, sulfonylurea or both. Oral anti-diabetic drugs (OADs) were continued at pre-trial doses throughout the trial, and 22% of subjects were treated with sulfonylureas (SU) in combination with metformin with or without pioglitazone. The mean age of the population was 58 years, and mean duration of diabetes was 10 years. 49% were male. 91% were White, 8% Black or African American. 11% were Hispanic or Latino ethnicity, 6% of patients had eGFR < 60mL/min/1.73m2; no patient had eGFR < 30mL/min/1.73m2. The mean BMI was 33.1 kg/m2.

The starting dose of XULTOPHY 100/3.6 was 16 units (16 units insulin degludec/0.58 mg liraglutide), and the average starting dose of liraglutide was 1.7 mg. XULTOPHY 100/3.6 was titrated twice weekly to target a fasting blood glucose goal of 72-90 mg/dL. The end of trial dose of XULTOPHY 100/3.6 was 44 units (44 units insulin degludec/1.58 mg liraglutide). The primary endpoint, change in HbA1c, was tested for superiority of XULTOPHY 100/3.6 to unchanged liraglutide therapy.

At the end of 26 weeks, there was a reduction in HbA1c from baseline of 1.31% for XULTOPHY 100/3.6 and 0.36% for liraglutide (see Table 8).

Table 8: Results of a 26-Week Trial with XULTOPHY 100/3.6 in Patients with Type 2 Diabetes Mellitus Inadequately Controlled on Liraglutide up to 1.8 mg Daily

	XULTOPHY 100/3.6 + metformin±pioglitazone±SU	liraglutide 1.8 mg+ metformin±pioglitazone±SU
Total (N)	232	116
HbA1c (%)
Baseline	7.8	7.8
End of Trial (LS Mean)#	6.4	7.4
Change from baseline (LS Mean)#	-1.31	-0.36
Estimated treatment difference [95% CI]	-0.95 [-1.15; -0.75]A
Percentage of patients achieving HbA1c <7%##	74.6%	30.2%
FPG (mg/dL)
Baseline	161	169
End of Trial (LS Mean)#	112	153
Change from baseline (LS Mean)#	-51.1	-10.9

: ATest for superiority evaluated at 5.0% level for significance, (p<0.0001)
: #Estimated using an ANCOVA with treatment, pre-trial liraglutide and region as fixed factors and baseline response as covariate. Multiple imputation modelled “return to baseline” of the treatment effect for subjects having missing week 26 data.
: ##Patients with missing HbA1c data at week 26 were considered as non-responders.
: There were 5.2% of subjects in the XULTOPHY 100/3.6 arm and 19.0% in the liraglutide arm for whom HbA1c data was missing at week 26.

Figure 3: Mean HbA1c (%) By Treatment Week in Patients with Type 2 Diabetes Mellitus Inadequately Controlled on Liraglutide

IDegLira=XULTOPHY 100/3.6

Converting from basal insulin

NCT01392573: The efficacy and safety of XULTOPHY 100/3.6 compared to insulin degludec, both once daily and added on to metformin, were studied in a 26-week randomized, double-blind, trial in 398 patients with type 2 diabetes mellitus inadequately controlled on basal insulin and metformin alone or in combination with sulfonylurea/glinides. Basal insulin and sulfonylurea/glinides were discontinued at randomization.

The mean age of the trial population was 57 years, and mean duration of diabetes was 11 years. 55% were male. 77% were White, 5% Black or African American. 10% were Hispanic or Latino ethnicity, 7% of patients had eGFR < 60mL/min/1.73m2; no patient had eGFR < 30mL/min/1.73m2. The mean BMI was 33.7 kg/m2. The mean dose of metformin and basal insulin in patients entering the trial was approximately 2 g and 29 units, respectively.

The starting dose of XULTOPHY 100/3.6 and insulin degludec was 16 units (16 units insulin degludec/0.58 mg liraglutide) and 16 units, respectively. XULTOPHY 100/3.6 and degludec were to be titrated twice weekly to target a fasting blood glucose goal of 72-90 mg/dL. Patients could not increase their dose by more than 4 units per week, and the maximum dose of insulin degludec was limited to 50 units. The targeted fasting blood glucose goal was achieved in 24.0% of patients randomized to insulin degludec and in 31.6% of the patients randomized to XULTOPHY 100/3.6 at 26 weeks.

At the end of 26 weeks, reductions in HbA1c from baseline of 1.94% for XULTOPHY 100/3.6 and 1.05% for insulin degludec limited to 50 units daily were observed (see Table 9). The mean difference (95% CI) in HbA1c reduction between XULTOPHY 100/3.6 and insulin degludec was -0.89 [-1.10; -0.68] and statistically significant. The trial was designed to show the contribution of the liraglutide component to glycemic lowering and the insulin degludec dosing algorithm was selected to isolate the effect of the GLP-1 component. At the end of the trial, the doses of insulin degludec were equivalent between treatment groups. The mean final dose of XULTOPHY 100/3.6 and insulin degludec was 46 units (for XULTOPHY 100/3.6: 46 units insulin degludec/1.66 mg liraglutide). The difference in glucose lowering effect observed in the trial may not necessarily reflect the effect that will be observed in the care setting where insulin degludec dosage can be different than that used in the trial.

Table 9: Results of a 26-Week Trial in Patients with Type 2 Diabetes Mellitus Inadequately Controlled on Basal Insulin

	XULTOPHY 100/3.6 + metformin	Insulin degludec* + metformin
Total (N)	199	199
HbA1c (%)
Baseline	8.7	8.8
End of Trial (LS Mean)#	6.9	7.7
Change from baseline (LS Mean)#	-1.94	-1.05
Estimated treatment difference [95% CI]#	-0.89 [-1.10; -0.68]A
Percentage of patients achieving HbA1c <7%##	57.3%	22.6%
FPG (mg/dL)
Baseline	175	172
End of Trial (LS Mean)#	110	118
Change from baseline (LS Mean)#	-63.5	-55.5

Ap<0.01. The trial was designed to show the contribution of the liraglutide component to glycemic lowering and the insulin degludec dosing algorithm was selected to isolate the effect of the GLP-1 component. At the end of the trial, the doses of insulin degludec were equivalent between treatment groups. The mean final dose of XULTOPHY 100/3.6 and insulin degludec was 46 units (for XULTOPHY 100/3.6: 46 units insulin degludec/1.66 mg liraglutide). The difference in glucose lowering effect observed in the trial may not necessarily reflect the effect that will be observed in the care setting where alternative insulin degludec dosage can be used.

*Maximum dose 50 units

: #Estimated using an ANCOVA with treatment, country, and previous antidiabetic treatment as fixed factors and baseline response as covariate. Multiple imputation modelled “jump to control” of the treatment effect for subjects having missing week 26 data.
: ##Patients with missing HbA1c data at week 26 were considered non-responders.
: There were 11.1% of subjects in the XULTOPHY 100/3.6 arm and 13.1% in the insulin degludec arm for whom HbA1c data was
: missing at week 26.

NCT01952145: The efficacy and safety of XULTOPHY 100/3.6 compared to insulin glargine U-100, both once daily and added on to metformin, were studied in a 26-week randomized, open-label, two-arm parallel trial in 557 patients with type 2 diabetes mellitus inadequately controlled on insulin glargine U-100 and metformin.

The mean age of the trial population was 59 years and mean duration of diabetes was 12 years. 50% were male. 95% were White, 2% Black or African American. 43% were Hispanic or Latino ethnicity, 6% of patients had eGFR < 60mL/min/1.73m2; one patient had eGFR < 30mL/min/1.73m2. The mean BMI was 32 kg/m2. The mean dose of insulin glargine U-100 in patients entering the trial was 32 units.

XULTOPHY 100/3.6 and insulin glargine were to be titrated twice weekly to target a fasting blood glucose goal of 72-90 mg/dL. The starting dose of XULTOPHY 100/3.6 was 16 units (16 units insulin degludec/0.58 mg liraglutide). The average starting dose of insulin glargine U-100 was 32 units. Patients could not increase the dose of the two products by more than 4 units per week and there was no maximum allowed dose of insulin glargine. The targeted fasting plasma blood glucose goal was achieved in 39.6% of patients randomized to insulin glargine and 32.9% of the patients randomized to XULTOPHY 100/3.6 at 26 weeks.

At the end of 26 weeks, treatment with XULTOPHY 100/3.6 resulted in a reduction in HbA1c from baseline of 1.67% and was 1.16% for insulin glargine U-100 (see Table 10) and excluded the pre-specified non-inferiority margin of 0.3%. At the end of the trial, the average dose of XULTOPHY 100/3.6 was 41 units (41 units insulin degludec/1.48 mg liraglutide) and the dose of glargine was 66 units, it is unclear that these observed differences in insulin doses are clinically important. The difference in HbA1c effect observed at 26 weeks may not necessarily reflect the effect in the care setting where insulin glargine may be more rapidly titrated.

Table 10: Results of a 26-Week Trial in Patients with Type 2 Diabetes Mellitus Inadequately Controlled on Insulin Glargine U-100

	XULTOPHY 100/3.6 + metformin	Insulin glargine U-100 + metformin
Total (N)	278	279
HbA1c (%)
Baseline	8.4	8.2
End of Trial (LS Mean)#	6.6	7.1
Change from baseline (LS Mean)#	-1.67	-1.16
Estimated treatment difference [95% CI]	-0.51 [-0.67; -0.34] A
Percentage of patients achieving HbA1c <7%##	68.3%	46.2%
FPG (mg/dL)
Baseline	161	160
End of Trial (LS Mean)#	110	110
Change from baseline (LS Mean)#	-49.9	-49.6

: Ap<0.01. Primary endpoint was tested for noninferiority of XULTOPHY 100/3.6 to insulin glargine U-100. The difference in glucose lowering effect observed in the trial may not necessarily reflect the effect that will be observed in the care setting where alternative insulin glargine dosage can be used.
: #Estimated using an ANCOVA with treatment and region as fixed factors and baseline response as covariate. Multiple imputation modelled “return to baseline” of the treatment effect for subjects having missing week 26 data.
: ##Patients with missing HbA1c value at week 26 data were considered non-responder.
: There were 10.1% of subjects in the XULTOPHY 100/3.6 arm and 4.7% in the insulin glargine U-100 arm for whom HbA1c data was missing at week 26.

14.4 Cardiovascular Outcomes Trials in Patients with Type 2 Diabetes Mellitus and Atherosclerotic Cardiovascular Disease Conducted with Liraglutide 1.8 mg and Insulin Degludec

The effect of XULTOPHY 100/3.6 on the risk of cardiovascular outcomes in patients with type 2 diabetes mellitus and atherosclerotic cardiovascular disease has not been established. The studies below were conducted with liraglutide 1.8 mg and insulin degludec, individually.

Liraglutide 1.8 mg

The LEADER trial (NCT01179048) randomized 9340 patients with inadequately controlled type 2 diabetes mellitus and cardiovascular disease to liraglutide 1.8 mg or placebo in addition to standard of care treatments for type 2 diabetes mellitus for a median follow up of 3.5 years.

Patients either were 50 years of age or older with established, stable cardiovascular, cerebrovascular, peripheral artery disease, chronic kidney disease or chronic heart failure (80% of patients) or were 60 years of age or older and had other specified risk factors for cardiovascular disease (20% of patients). The population was 64% male, 78% White, 10% Asian, and 8% Black or African American; 12% of the population was Hispanic or Latino. The mean duration of type 2 diabetes mellitus was 13 years, the mean HbA1c was 8.7% and the mean BMI was 33 kg/m2; the mean eGFR at baseline was 79 mL/min/1.73 m2.

In total, 96.8% of the patients completed the trial; vital status was available for 99.7%. The primary endpoint was the time from randomization to first occurrence of a major adverse cardiovascular event (MACE) defined as: cardiovascular death, non-fatal myocardial infarction, or non-fatal stroke. No increased risk for MACE was observed with liraglutide 1.8 mg. The total number of primary component MACE endpoints was 1302 (608[13.0%] with liraglutide 1.8 mg and 694 [14.9%] with placebo).

Insulin degludec

The DEVOTE trial (NCT01959529) randomized 7,637 patients with inadequately controlled type 2 diabetes mellitus and cardiovascular disease to either insulin degludec or insulin glargine U-100. Each was administered once-daily in addition to standard of care treatments for diabetes for a median duration of follow up of 2 years.

Patients either were 50 years of age or older and had established, stable cardiovascular, cerebrovascular, peripheral artery disease, chronic kidney disease or chronic heart failure (85% of patients) or were 60 years of age or older and had other specified risk factors for cardiovascular disease (15% of patients). The population was 63% male, 76% White 11% Black or African American, and 10% Asian; 15% of the population was Hispanic or Latino. The mean HbA1c was 8.4% and the mean BMI was 33.6 kg/m2. The baseline mean eGFR was 68 mL/min/1.73m2.

In total, 98% of the patients completed the trial; vital status was known at the end of the trial for 99%. The primary endpoint was the time from randomization to the first occurrence of MACE, defined as: cardiovascular death, non-fatal myocardial infarction, or non-fatal stroke. No increased risk for MACE was observed with insulin degludec when compared to insulin glargine U-100. The total number of primary MACE endpoints was 681 (325 [8.5%] with insulin degludec and 356 [9.3%] with insulin glargine).