Efficacy and safety of insulin degludec U100 and insulin glargine U100 in combination with meal-time bolus insulin in hospitalized patients with type 2 diabetes: an open-label, randomized controlled study

Abstract

The short-term efficacy and safety of insulin degludec U100 (IDeg) in patients with type 2 diabetes have not been reported widely. We compared insulin IDeg and insulin glargine U100 (IGla) for glycemic control and glucose variability in hospitalized patients with type 2 diabetes. In an open-label, multicenter, randomized controlled trial, 74 patients were randomly assigned to either the IDeg (36 patients) or IGla (38 patients) group and were administered with basal-bolus therapy during hospitalization. Following the start of the treatment, on day 11, glucose variability was assessed by continuous glucose monitoring. A fasting blood glucose level of 110 mg/dL and 2-hour postprandial blood glucose level of 180 mg/dL throughout at least one day during the observation period were achieved in 31.3% (10/32) and 30.6% (11/36) of the patients in the IDeg and IGla groups, respectively. The 6-point self-monitoring of blood glucose profiles showed a significant difference between the two groups. On day 7, the intra-day variation was larger in the IDeg group than in the IGla group. The incidence of hypoglycemia or glucose variability was comparable in the two groups. This study suggests that short-term efficacy and safety of IDeg and IGla in patients with type 2 diabetes during the initial phase of basal-bolus therapy were comparable, and these results can help in deciding which treatment to opt for.

THE PRINCIPAL CAUSES OF TYPE 2 DIABETES are decreased insulin secretion from the pancreas and increased insulin resistance. Insulin resistance can be improved to a certain extent by diet [1, 2] and exercise [3-5], or use of an insulin resistance-improving agent. However, insulin secretion decreases with age [6]. In particular, lower pancreatic insulin secretion due to genetic factors has been reported much more among the Japanese population than among the North American and European populations [7]. Furthermore, it is difficult to improve insulin secretion with existing oral antidiabetic drugs.

There is also evidence that a persistent hyperglycemic state contributes to the induction of β cell dedifferentiation [8, 9]. However, β cell re-differentiation is possible after the normalization of blood glucose level and elimination of glucotoxicity [9, 10]. Moreover, multiple daily insulin injections (MDIs) are often given to patients with type 2 diabetes in order to achieve rapid normalization of glycemia and preservation of β-cell function.

Previously, neutral protamine Hagedorn (NPH) insulin was widely used to supplement basic secretion of insulin. This was subsequently replaced by insulin glargine U100 (IGla) and insulin detemir, which are long-acting soluble types of insulin. However, these products not only resulted in unstable insulin concentrations in the blood [11] but also had a short duration of action [12, 13], often making glycemic control difficult. Moreover, when long-acting insulin is administered before sleep, hypoglycemia frequently occurs between midnight and early morning. In addition, a hyperglycemic state occurs around the time of the evening meal, on the following day [11].

Insulin degludec U100 (IDeg), was developed as an ultra-long-acting basal insulin. After subcutaneous injection, it forms soluble multi-hexamers that slowly dissociate into monomers, resulting in a flat and consistent insulin action profile [14]. IDeg has a half-life of 24 hours, while IGla has a half-life of only 12 hours, and its glucose-lowering effects are stronger during the first 12 hours after injection than during the subsequent 12 hours [15, 16]. Compared to IGla, IDeg has a flat insulin action curve and a fourfold smaller pharmacodynamic variability [17], resulting in a similar efficacy but a lower risk of hypoglycemia [18-20].

In Japan, insulin therapy is often initiated during hospitalization for approximately two weeks to help with the management of dysglycemia and to assist elderly patients in learning the proper use of insulin devices. However, owing to its long half-life, titration of IDeg is time consuming. It is generally accepted that titration should allow the monitoring of glycemia over 2–3 days, so IDeg may be unsuited for titration every 1–2 days during hospitalization because of its long duration of action. Therefore, we performed this study to evaluate the short-term efficacy and safety of IDeg administered at the start of basal-bolus therapy in patients with type 2 diabetes who required hospitalization due to poor glycemic control.

Materials and Methods

Patients

This study involved subjects with type 2 diabetes, aged ≥20 but <80 years, hospitalized for control of glycemia at either the Yokohama City University Medical Center or the Fujisawa City Hospital. The inclusion criteria were (1) insulin naïve, drug-free, or treated only with an oral antidiabetic drugs, (2) HbA1c levels ≥8.0% and <14.0% on admission, and (3) fasting blood glucose levels ≥200 mg/dL or a 2-hour postprandial blood glucose of ≥300 mg/dL, 2 days after admission and requiring insulin therapy due to poor glycemic control. The exclusion criteria were: (1) history of diabetic ketoacidosis or diabetic coma within 6 months prior to selection for this study, (2) severe infection or severe trauma, (3) scheduled to undergo surgery during the observation period of this study, (4) severe liver dysfunction (Child-Pugh score class C), (5) renal insufficiency (serum creatinine >2.0 mg/dL or estimated glomerular filtration rate [eGFR] <30 mL/min), (6) pregnant or breast-feeding, and (7) history of hypersensitivity to IGla or IDeg.

This study was performed in accordance with the ethics and principles of the Declaration of Helsinki, and the study protocol was approved by the Ethics Committees of the Yokohama City University School of Medicine and the Fujisawa City Hospital. All patients gave written informed consent, and patient anonymity was preserved. The protocol was registered with the UMIN Clinical Trial Registry as UMIN000010447.

Study protocol

This open-label, multicenter, randomized controlled trial was designed to assess the efficacy and safety of IGla and IDeg, though comparing the two agents was not the primary objective. All patients were hospitalized and randomly assigned to an IGla or an IDeg group (1:1 ratio) by a minimization method stratified according to fasting plasma glucose (FPG), age, and sex. The allocation was performed at the data center for clinical trials of the Yokohama City University.

During hospitalization, participants were tested 6 times daily, before every meal and 2 hours after every meal using a finger-stick glucose monitoring kit (Glutest Every; Sanwa Kagaku Kenkyusho, Aichi, Japan). Diet therapy with a pre-determined energy content ({ideal body weight} × 25–27 kcal/day) was also initiated upon admission. Meal times were as follows: breakfast 8:00 am, lunch 12:00 pm, and dinner 6:30 pm. The first two days were used for observation, and the basal-bolus therapy was started on the third day. The first day of insulin treatment was considered as day 1, with day 0 being the baseline. All oral antidiabetic drugs were discontinued before the commencement of insulin treatment. Insulin titration was performed by an expert endocrinologist; basal insulin (IGla or IDeg) was administered before sleep, and insulin aspart, insulin lispro, or insulin glulisine was administered as bolus insulin. Insulin treatment was commenced with 4 units of bolus insulin before each meal, and 4 units of basal insulin. Basal insulin was adjusted based on the glycemia measured before breakfast. Bolus insulin was titrated in accordance with the algorithm presented in Table 1, based on blood glucose levels, before lunch, before dinner, and 2 hours after dinner for the morning, afternoon, and evening administrations, respectively. Considering the half-life of IDeg, the quantity of insulin was adjusted every 2 days. On Day 11, continuous glucose monitoring (CGM) was performed using iPro2 (Medtronic Minimed, Northbridge, CA, USA).

Table 1 The insulin adjustment algorithm

Adjustment of basal insulin (insulin glargine U100 or insulin degludec U100)
Blood glucose before breakfast	<80 mg/dL	Decrease by 2 units
	80–109 mg/dL	No adjustment
	110–140 mg/dL	Increase by 1 unit
	141–180 mg/dL	Increase by 2 units
	181–250 mg/dL	Increase by 3 units
	≥251 mg/dL	Increase by 4 units
Adjustment of bolus insulin
Bolus insulin adjustment before breakfast
Blood glucose before lunch	<56 mg/dL	Decrease by 4 units
	56–79 mg/dL	Decrease by 2 units
	80–109 mg/dL	No adjustment
	110–140 mg/dL	Increase by 1 unit
	141–180 mg/dL	Increase by 2 units
	181–250 mg/dL	Increase by 3 units
	≥251 mg/dL	Increase by 4 units
Bolus insulin adjustment before lunch
Blood glucose before dinner	<56 mg/dL	Decrease by 4 units
	56–79 mg/dL	Decrease by 2 units
	80–109 mg/dL	No adjustment
	110–140 mg/dL	Increase by 1 unit
	141–180 mg/dL	Increase by 2 units
	181–250 mg/dL	Increase by 3 units
	≥251 mg/dL	Increase by 4 units
Bolus insulin adjustment before dinner
Blood glucose 2 hours after dinner	<56 mg/dL	Decrease by 2 units
	56–79 mg/dL	Decrease by 1 unit
	80–180 mg/dL	No adjustment
	181–250 mg/dL	Increase by 1 unit
	≥251 mg/dL	Increase by 2 units

Basal insulin (insulin glargine U100 or insulin degludec U100) was adjusted every 2 days according to blood glucose levels before breakfast. Bolus insulin was adjusted according to blood glucose levels before lunch (morning administration), before dinner (afternoon administration), and 2 hours after dinner (evening administration).

The primary endpoint was the achievement of target glycemic control during the first 12 days and the time (days) taken to achieve it. Target glycemic control included fasting blood glucose below 110 mg/dL and 2-hour postprandial blood glucose below 180 mg/dL, throughout a day. Glycemic control was assessed by a 6-point self-monitoring of blood glucose (SMBG). This was based on the blood glucose levels reported to be efficacious for inhibiting the development of microvascular complications, in the Kumamoto Study [21]. The secondary endpoints were average fasting and 2-hour-postprandial blood glucose levels, and the parameters of glucose variability as standard deviation (SD), the coefficient of variation (CV), and mean amplitude of glucose excursions (MAGE) in the SMBG data at baseline (day 0), day 7, and day 12. The changes from baseline were also evaluated on day 7 and day 12. The frequency of hypoglycemia was considered as the endpoint for assessing the safety of IDeg and IGla.

Statistical methods

The primary analysis calculated the proportion of patients who achieved the target glycemic control in the IDeg and IGla groups. These proportions were tested against the null hypothesis value of 20% (this value was pre-determined clinically). Each drug was considered effective if the proportion of patients achieving target glycemic control exceeded 20% with a one-sided p-value <0.05. Time to achievement of target glycemic control was summarized and compared using the t-test. A secondary analysis compared the proportions of patients who achieved glycemic control in the two groups, using the Chi-squared test. It also compared the incidence of hypoglycemia, 6-point SMBG profiles (adjusted mean and its error, change from baseline, SD, CV, and MAGE on day 7 and day 12), insulin units, CGM profiles (glucose mean, SD, CV, MAGE, and hypoglycemia duration in minutes). Comparisons were performed using the t-test and Wilcoxon test. We also performed post hoc analysis: analysis of covariance adjusting for the unbalanced baseline characteristics.

The sample size was calculated based on the assumption that the true probability of achieving the target glycemic control was 40% in each group [22-24]. The required sample size was at least 30 in each group to test against the null hypothesis probability of 20% with a statistical power of 80% at the one-sided alpha of 5%. Analyses were performed using SAS version 9.4 (SAS institute, Cary, NC, USA).

Results

Patient characteristics

A total of 74 patients were randomized between April 2013 and March 2015. Of these, 2/38 patients in the IGla group and 4/36 patients in the IDeg group dropped out before completion of the study, due to withdrawal of consent (n = 2), emergency surgery (n = 2), violation of the rules on starting medication (n = 1), and missing data (n = 1) (Fig. 1). These patients were excluded from the final analysis. After obtaining consent, 24-hour CGM was performed in 50/68 subjects who completed this study (28 in the IGla group and 22 in the IDeg group), starting at 6:00 pm on Day 11.

Fig. 1

Flow chart of the study

This study included 74 patients, allocated randomly to the insulin glargine U100 (IGla) group (n = 38) or the insulin degludec U100 (IDeg) group (n = 36). Of these, 36 and 32 patients from the IGla and IDeg groups, respectively completed the trial. Continuous glucose monitoring (CGM) was performed in 50 patients (IGla group, 28 patients; IDeg group, 22 patients).

The baseline characteristics of the 74 subjects are shown in Table 2. Significant differences between the IGla and IDeg groups in the levels of HbA1c (10.4 ± 1.9 vs. 11.3 ± 1.4, p = 0.019) and eGFR (72.8 ± 17.3 vs. 82.8 ± 17.6, p = 0.024) were noted. Furthermore, there were more patients on sulfonylurea drugs in the IGla group compared to the IDeg group (20 patients [57.1%] vs. 8 patients [25.8%]). No other significant differences were noted between the two groups.

Table 2 Baseline characteristics of the patients

Total	Insulin glargine U100 group (n = 38)	Insulin degludec U100 group (n = 36)	p value
Male	19 (52.8%)	19 (59.4%)	0.631
Age (years)	61.8 ± 9.4	58.9 ± 10.5	0.248
BMI (kg/m2)	24.8 ± 3.5	25.7 ± 5.0	0.382
Disease duration (years)	6.6 ± 8.2	3.9 ± 4.6	0.110
FPG (mg/dL)	207.0 ± 89.4	185.8 ± 43.2	0.226
HbA1c (%)	10.4 ± 1.9	11.3 ± 1.4	0.019
eGFR (mL/min/1.73 m2)	72.8 ± 17.3	82.8 ± 17.6	0.024
Fasting CPR (ng/mL)	2.5 ± 0.8	2.5 ± 1.1	0.815
CPR index	1.35 ± 0.50	1.36 ± 0.59	0.751
Urine albumin (mg/gCr)	53.8 ± 176.1	19.8 ± 43.1	0.333
Urine albumin, median [range]	8.7 [0–983.4]	7.2 [0–215.9]	0.694
Oral antidiabetic drugs before the start of the study; n (%)
Metformin	8 (22.9%)	6 (18.8%)	0.769
Sulfonylurea	20 (57.1%)	8 (25.8%)	0.013
DPP-4 inhibitor	15 (42.9%)	13 (40.6%)	1.000
α-Glucosidase inhibitor	7 (20.6%)	4 (12.9%)	0.516
Pioglitazone	2 (5.9)	4 (12.5)	0.420
Glinide	1 (2.9%)	0 (0%)	1.000
Complications; n (%)
Diabetic retinopathy	7 (21.2%)	3 (10.7%)	0.319
Urinary albumin below 30 mg/gCr	22 (61.1%)	22 (68.8%)	0.199
30–299 mg/gCr	8 (22.2%)	9 (28.1%)
above 300 mg/gCr	6 (16.7%)	1 (3.1%)
CGM evaluated patients	Insulin glargine U100 group (n = 28)	Insulin degludec U100 group (n = 22)	p value
Male	14 (50%)	14 (63.6%)	0.398
Age (years)	61.1 ± 10.1	57.6 ± 12.1	0.273
BMI (kg/m2)	25.0 ± 3.2	25.9 ± 5.5	0.434
Disease duration (years)	6.0 ± 8.3	4.7 ± 5.0	0.526
HbA1c (%)	10.3 ± 1.9	11.4 ± 1.3	0.031
FPG (mg/dL)	202.0 ± 87.9	185.4 ± 44.7	0.423

Data are presented as mean ± SD

BMI, body mass index; FPG, fasting plasma glucose; HbA1c, glycosylated hemoglobin; CPR, C-peptide immunoreactivity; CGM, continuous glucose monitoring.

CPR index: C-peptide immunoreactivity index; serum C-peptide immunoreactivity/fasting plasma glucose × 100

All oral antidiabetic drugs were stopped before starting insulin treatment.

In the IGla group, the rapid-acting insulin types used were insulin aspart in 10 (26.3%) patients, insulin lispro in 28 (73.7%), and insulin glulisine in 0 (0%) patients in the. In the IDeg group, those used were insulin aspart in 11 (30.6%) patients, insulin lispro in 25 (69.4%) patients, and insulin glulisine in 0 (0%) patients.

Glycemic control and frequency of hypoglycemia

While 11/36 (30.6%) patients achieved target glycemic control in the IGla group [90% confidence interval (CI) 18.2%–45.5%, one-sided p-value = 0.057], 10/32 (31.3%) patients achieved the same in the IDeg group [90% CI, 18.0%–47.2%, one sided p-value = 0.056] (Table 3a). The difference between the two groups was not statistically significant (Chi-squared p-value = 0.951). The average number of days ± SD to achieve target glycemic control in the IGla and IDeg groups was 8.0 ± 3.4 and 8.7 ± 1.6, respectively (p = 0.559) (Table 3a).

Table 3 Data on study endpoints (a) Proportion of patients achieving target glycemic control and frequency of hypoglycemia

	Insulin glargine U100 group (n = 36)	p value	Insulin degludec U100 group (n = 32)	p value
Proportion of patients achieving target glycemic control
Achieved (n, %)	11 (30.6%)	0.057*	10 (31.3%)	0.056*
Not achieved (n, %)	25 (69.4%)		22 (68.7%)
Days to achievement (mean ± SD)	8.0 ± 3.4		8.7 ± 1.6	0.559
Frequency of hypoglycemia
Patients (n, %)
No hypoglycemia	17 (47.2%)		16 (50.0%)	0.782
Hypoglycemia (from 54 to 70 mg/dL)	15 (41.7%)		13 (40.6%)
Hypoglycemia (lower than 54 mg/dL)	4 (11.1%)		3 (9.4%)
Events per patient (median [min–max])
70 mg/dL or below†	0.5 [0–7]		0.5 [0–5]	0.931
lower than 54 mg/dL	0 [0–1]		0 [0–1]	0.825

*p value for the null hypothesis: proportion of patients achieving target glycemic control 20%

^†includes less than 54 mg/dL

The frequency of hypoglycemia (blood glucose ≤70 mg/dL) during the insulin titration period (day 1 to day 12) was calculated. While 15 (41.7%) and 13 (40.6%) patients in the IGla and IDeg groups, respectively had hypoglycemia from 54 to 70 mg/dL, 4 (11.1%) and 3 (9.4%) patients in the IGla and IDeg groups, respectively had hypoglycemia lower than 54 mg/dL (p = 0.782, Mantel-Haenszel Chi-squared test) (Table 3a). The number of hypoglycemic events per patient was 0.5 (0–7) in the IGla group and 0.5 (0–5) in the IDeg group (p = 0.932), with no significant difference.

Six-point SMBG profile data

The 6-point SMBG profiles at baseline (the day before starting of insulin treatment, day 0), day 7, and day 12, with mean, SD, CV, and MAGE as indices of glucose variability on each day, are shown in Table 3b. Comparisons between the groups were adjusted with the baseline levels of HbA1c. No significant differences were noted between the IGla and IDeg groups on both days 7 and 12; but a significant difference was noted in the SD on day 7 (IGla group: 36.9 ± 2.7%, IDeg group: 45.3 ± 2.9%, p = 0.041) and in CV on day 7 (IGla group: 25.7 ± 1.6%, IDeg group: 30.8 ± 1.7%, p = 0.037) (Table 3b). Groups with and without adjustment for baseline HbA1c had comparable results.

Table 3 (b) 6-point Self-monitoring of blood glucose (SMBG) profile data

	Insulin glargine U100 group (n = 36)			Insulin degludec U100 group (n = 32)			p value for group comparison adjusted with baseline HbA1c
	Day 0	Day 7	Day 12	Day 0	Day 7	Day 12	Day 0	Day 7	Day 12
Before breakfast (mg/dL)	196.9 ± 8.2	127.3 ± 6.0	113.5 ± 4.3	180.6 ± 9.1	119.6 ± 6.4	108.1 ± 4.7	0.200	0.386	0.402
Change from Day 0 (%)		–30.9 ± 3.4	–36.7 ± 2.9		–30.7 ± 3.7	–39.4 ± 3.3		0.979	0.566
After breakfast (mg/dL)	282.9 ± 14.5	162.0 ± 9.3	150.7 ± 8.9	278.5 ± 16.1	168.3 ± 9.9	169.2 ± 9.2	0.844	0.648	0.154
Change from Day 0 (%)		–38.3 ± 4.6	–41.5 ± 5.3		–35.9 ± 5.1	–35.0 ± 5.7		0.740	0.425
Before lunch (mg/dL)	243.7 ± 18.3	142.5 ± 7.6	114.8 ± 9.8	222.5 ± 20.3	131.3 ± 7.9	131.6 ± 9.8	0.453	0.319	0.235
Change from Day 0 (%)		–31.3 ± 4.3	–42.8 ± 5.7		–42.4 ± 4.6	–41.5 ± 6.0		0.089	0.882
After lunch (mg/dL)	269 ± 13.5	141.3 ± 10.0	135.5 ± 9.4	279.1 ± 14.9	168.1 ± 10.5	132 ± 9.8	0.629	0.069	0.805
Change from Day 0 (%)		–39.6 ± 4.6	–47.1 ± 4.7		–40.9 ± 5.0	–50.3 ± 5.3		0.852	0.665
Before dinner (mg/dL)	194.0 ± 12.3	118.4 ± 6.3	132.8 ± 10.9	180.5 ± 13.2	134.6 ± 6.6	127.4 ± 11.5	0.467	0.084	0.739
Change from Day 0 (%)		–29.0 ± 4.4	–27.9 ± 6.4		–24.2 ± 4.5	–29.9 ± 6.7		0.462	0.833
After dinner (mg/dL)	269.3 ± 11.2	158.9 ± 10.0	167.6 ± 10.8	281.0 ± 12.0	184.3 ± 10.5	145.6 ± 11.3	0.491	0.086	0.169
Change from Day 0 (%)		–38.5 ± 4.7	–36.5 ± 4.7		–32.7 ± 4.8	–48.6 ± 5.0		0.398	0.092
6-point SMBG mean (mg/dL)	242.8 ± 11.6	141.9 ± 5.2	136.7 ± 5.6	237 ± 12.9	148.0 ± 5.5	136.6 ± 6.0	0.746	0.433	0.996
SD (mg/dL)	56.4 ± 3.7	36.9 ± 2.7	38.3 ± 4.0	57.7 ± 4.1	45.3 ± 2.9	41.3 ± 4.3	0.814	0.041	0.605
CV (%)	23.2 ± 1.1	25.7 ± 1.6	27.5 ± 2.2	25.0 ± 1.2	30.8 ± 1.7	29.4 ± 2.3	0.273	0.037	0.557
MAGE (mg/dL)	78.7 ± 5.9	53.9 ± 4.8	58.0 ± 6.6	74.1 ± 6.6	67.5 ± 5.2	57.1 ± 7.1	0.618	0.059	0.929

Data are presented as LS-Mean ± SE. LS, least square; CV, coefficient of variation; MAGE, mean amplitude of glucose excursions

Insulin units

The number of insulin units administered on day 1, 7, and 12 is shown in Table 3c. No significant differences were noted between the two groups in the total daily insulin dose. There was also no significant difference in the basal and bolus insulin doses.

Table 3 (c) Insulin units administered

	Insulin glargine U100 group (n = 36)	Insulin degludec U100 group (n = 32)	p value
Total
Day 7 (units)	27.2 ± 9.5	26.1 ± 9.1	0.631
Day 12 (units)	28.5 ± 13.3	28.3 ± 13.9	0.940
Basal insulin (IGla or IDeg)
Day 7 (units)	7.3 ± 3.1	6.3 ± 3.0	0.167
Day 12 (units)	7.9 ± 4.6	6.4 ± 4.7	0.181
Bolus insulin-Breakfast
Day 7 (units)	7.0 ± 3.0	7.2 ± 2.9	0.798
Day 12 (units)	7.3 ± 4.3	8.0 ± 3.8	0.510
Bolus insulin-Lunch
Day 7 (units)	5.6 ± 2.1	5.7 ± 2.5	0.936
Day 12 (units)	5.6 ± 2.5	5.9 ± 3.6	0.639
Bolus insulin-Dinner
Day 7 (units)	7.3 ± 2.6	7.0 ± 2.5	0.624
Day 12 (units)	7.7 ± 3.6	8.0 ± 3.6	0.754

Data are presented as mean ± SD. IGIa, insulin glargine U100; IDeg, insulin degludec U100

CGM glucose data

Data on glucose variability parameters (mean, SD, CV, and MAGE) for the 50 patients (28 and 22 in the IGla and IDeg groups, respectively) in whom CGM was performed for 24 hours starting at 6 pm on day 11 are shown in Table 4. The CGM profiles are shown in Fig. 2 No differences were seen in glucose variability between the two groups, throughout the period of either hypoglycemia or hyperglycemia, even during the night (0 am–6 am).

Table 4 Data on glucose variability derived from CGM

	Insulin glargine U100 group (n = 28)	Insulin degludec U100 group (n = 22)	p value
Day 11, 6 pm–Day 12, 6 pm, mean ± SD
CGM glucose mean (mg/dL)	122.3 ± 17.5	127.2 ± 23.9	0.407
SD (mg/dL)	31.9 ± 9.9	33.2 ± 15.1	0.721
CV (%)	26.4 ± 8.7	26.7 ± 13.2	0.940
MAGE (mg/dL)	47.3 ± 13.9	49.3 ± 23.1	0.713
Day 12, 0 am–Day 12, 6 am, mean ± SD
CGM glucose mean (mg/dL)	93.5 ± 21.8	101.2 ± 32.2	0.316
SD (mg/dL)	8.8 ± 5.4	11.7 ± 6.9	0.101
CV (%)	9.6 ± 6.2	11.8 ± 6.2	0.206
MAGE (mg/dL)	13.0 ± 7.9	18.2 ± 10.7	0.053
Hypoglycemia duration (minutes)
≤70 mg/dL
median [min–max]	0 [0–495]	0 [0–605]	*0.509
median [25% tile–75% tile]	0 [0–35]	0 [0–175]
<54 mg/dL
median [min–max]	0 [0–365]	0 [0–330]	*0.486
median [25% tile–75% tile]	0 [0–0]	0 [0–0]
Hyperglycemia duration (minutes)
>180 mg/dL
median [min–max]	80 [0–440]	65 [0–640]	*0.890
median [25% tile–75% tile]	80 [32.5–152.5]	65 [0–220]

* Wilcoxon test

CGM, continuous glucose monitoring; SD, standard deviation; CV, coefficient of variation; MAGE, mean amplitude of glucose excursions

Fig. 2

Comparison of 24-hour continuous glucose monitoring (CGM) profiles. Shown are the glucose levels for patients in the insulin degludec U100 (IDeg) and insulin glargine U100 (IGla) groups on day 11, 6 pm to day 12, 6 pm.

Discussion

In this open-label randomized controlled trial, the primary endpoint was the proportion of patients who achieved target glycemic control, which exceeded the pre-specified threshold for both groups, although it was not statistically significant. Furthermore, there was no significant difference in the frequency of glycemic control between the two groups. The 6-point SMBG profiles showed no significant differences in the mean blood glucose levels at baseline (day 0) and on days 7 or 12. However, the glucose variability parameters such as SD and CV on day 7 were significantly greater in the IDeg group than in the IGla group. No significant differences were noted in (a) the number of insulin units used on day 7 or day 12, (b) incidence of hypoglycemia, (c) any of the glucose variability parameters (mean, SD, CV, and MAGE), (d) duration of hypoglycemia, and (e) the results of CGM performed between day 11 and day 12.

In Japan, insulin therapy is often started during hospitalization. For all patients, this helps with the management of dysglycemia. For elderly people, it also helps in learning how to use the insulin devices properly. To date, IGla has been used in many cases when introducing basal-bolus treatment in hospitals. IGla is one of the standard long-acting soluble insulins, and its safety has been established [25]. It is frequently used owing to its comparatively short half-life, which makes insulin adjustment every 1–2 days easy. In recent years, however, IDeg has appeared on the market as the new, long-acting, soluble insulin, with a better stable concentration in the blood compared to IGla [14-17]. However, most of the evidence is based on comparatively long-term studies.

IDeg has a long duration of action of at least 42 hours, [26] and a steady state is achieved after 2–3 days of once a day administration [27]. However, this increases the risk of erratic glycemic variations, thereby enhancing the risk of hypoglycemia and requiring titration every 1–2 days. There are concerns that this could be a disadvantage for the short-term use of IDeg.

In this study, the proportion of patients who achieved target glycemic control during insulin titration in the hospital was not statistically significant because our expected value of 40% was higher than the results observed. Glycemic control was similar in the two groups, and there was no significant difference in the number of days taken to achieve this target as shown in Table 3a.

Regarding safety, previous studies have shown that the risk of hypoglycemia particularly nocturnal, is lower with IDeg than with IGla [18-20]. This finding has also been reported in Phase III studies on IDeg [22, 28, 29]. Moreover, IDeg is safer than IGla with respect to frequency of severe hypoglycemia in a long-term study in patients with type 2 diabetes [30]. In this study, the incidence of hypoglycemia was low and showed no difference between the two groups. This could be due to the short trial period and controlled titration, preventing hypoglycemic episodes from developing.

Increased glucose fluctuation is known to be associated with elevated risks of mortality [31] and cardiovascular disorders [32]. Although several reports have compared daily glucose variability with the use of IDeg or IGla, almost all of them involved patients with type 1 diabetes. Furthermore, the reported variability was significantly smaller with IDeg than with IGla [33-35]. Following a treat-to-target trial in insulin-naïve Japanese patients with type 2 diabetes, Aso et al. have reported that the CV values 24 weeks after starting insulin treatment were significantly smaller in the IDeg group than in the IGla group [36].

A short-term comparison of IGla and IDeg, in a randomized study of insulin naïve patients with type 2 diabetes has shown that IDeg provided glycemic control comparable to that by IGla, without increasing the frequency of hypoglycemia [37], although the study period was only 16 weeks. In a prospective observational study by Yamamoto et al., on in-patients with type 1 diabetes, CGM revealed that MAGE significantly decreased when switching from IGla to IDeg [35]. There have not been many similar reports for type 2 diabetes [38], and this is the first randomized clinical trial report, comparing the efficacy and safety of IGla and IDeg over a short term in patients with type 2 diabetes.

The SD and CV on day 7 were higher in the IDeg group, indicative of greater glucose fluctuation, some of which might be explained by higher baseline HbA1c, to begin with. However, the differences continued to be significant even after adjustment for baseline HbA1c. Additionally, no significant differences were observed in the SD and CV on day 12. Therefore, when administering IDeg, it might be necessary to check for hypoglycemia in the ultra-short period, which is within 7 days.

In the present study, IGla and IDeg were administered before sleep. However, since IDeg is long acting, administration before sleep is not imperative, and instead can be timed according to the convenience of the patients or their family members (in cases where self-administration is not possible). It has been reported that somewhat irregular administration does not adversely affect safety and efficacy [39-41]. Furthermore, the subjects in the present study were patients with type 2 diabetes, who retained some degree of insulin secretion capability. However, a report involving patients with type 1 diabetes showed that HbA1c levels, glucose profiles, the risk of hypoglycemia, and the degree of satisfaction with treatment, improved by switching from IGla or insulin detemir twice daily to IDeg once daily [42]. These findings indicate that the efficacy and safety of IDeg are comparable to those of IGla in the short term when insulin is introduced and that the use of IDeg is likely to improve the patients’ quality of life.

This study has some limitations. First, differences in the baseline HbA1c or use of oral antidiabetic drugs might have influenced the results. Further research using a larger sample size is required to validate these differences. Second, some of the improvement in glycemic control could be attributed to dietary changes, although diet therapy was started to match these conditions. Third, although CGM was carried out in this study, it could not be done for all cases, raising the possibility that insufficient data were obtained for comparing the two groups. Moreover, IGla U300 is presently commercially available in Japan, thereby increasing the options for long-acting insulin.

In conclusion, the present study showed that the efficacy and safety of IDeg and IGla are comparable in a 14-day time frame following the initial introduction of insulin. These short-term findings are meaningful because they show that an appropriate long-acting agent can be selected based on the existing reports regarding its long-term efficacy and safety [22, 43].

Acknowledgements

This work was conducted independently; no company or institution supported it financially. We thank Ms. Yamagiwa, Ms. Seki, and Ms. Morimoto for secretarial assistance. We thank all physicians who participated in this study. We would like to thank Editage (www.editage.jp) for English language editing.

Disclosure

Yasuo Terauchi received honoraria for lectures from MSD K.K.; Ono Pharmaceutical Co., Ltd.; Nippon Boehringer Ingelheim Co., Ltd.; Takeda Pharmaceutical Co., Ltd.; Mitsubishi Tanabe Pharma Corp.; Daiichi Sankyo Co., Ltd.; Sanwa Kagaku Kenkyusho Co., Ltd.; Kowa Pharmaceutical Co., Ltd.; Novo Nordisk Pharma Ltd.; Eli Lilly Japan K.K.; Sanofi K.K.; Shionogi & Co., Ltd.; Bayer Yakuhin, Ltd.; and AstraZeneca K.K. and obtained research support from MSD K.K.; Ono Pharmaceutical Co., Ltd.; Nippon Boehringer Ingelheim Co., Ltd.; Takeda Pharmaceutical Co., Ltd.; Mitsubishi Tanabe Pharma Corp.; Daiichi Sankyo Co., Ltd.; Sanwa Kagaku Kenkyusho Co., Ltd.; Novo Nordisk Pharma Ltd.; Eli Lilly Japan K.K.; Sanofi K.K.; Dainippon Sumitomo Pharma Co., Ltd.; Shionogi & Co., Ltd.; Bayer Yakuhin, Ltd.; and Astellas Pharma, Inc.

Tadashi Yamakawa received honoraria for lectures from MSD K.K.; Kowa Pharmaceutical Co., Ltd.; Novo Nordisk Pharma Ltd.; Sanofi K.K. and obtained research support from AstraZeneca K.K.

Jun Suzuki, Mari Oba, Jo Nagakura, Erina Shigematsu, Haruka Tamura, Kenichiro Takahashi, Yoshihisa Okamoto, Rika Sakamoto, Minori Shinoda/Matsuura, Tatsuro Takano, and Kazuaki Kadonosono declare that they have no potential conflicts of interest associated with this research.

References

• 1   Pastors  JG,  Warsaw  H,  Daly  A,  Franz  M,  Kulkarni  K (2002) The evidence for the effectiveness of medical nutrition therapy in diabetes management. Diabetes Care 25: 608–613.
• 2   Lindstrom  J,  Peltonen  M,  Eriksson  JG,  Ilanne-Parikka  P,  Aunola  S, et al. (2013) Improved lifestyle and decreased diabetes risk over 13 years: long-term follow-up of the randomised Finnish Diabetes Prevention Study (DPS). Diabetologia 56: 284–293.
• 3   Umpierre  D,  Ribeiro  PA,  Kramer  CK,  Leitao  CB,  Zucatti  AT, et al. (2011) Physical activity advice only or structured exercise training and association with HbA1c levels in type 2 diabetes: a systematic review and meta-analysis. JAMA 305: 1790–1799.
• 4   Dunstan  DW,  Zimmet  PZ,  Welborn  TA,  Cameron  AJ,  Shaw  J, et al. (2002) The Australian Diabetes, Obesity and Lifestyle Study (AusDiab)—methods and response rates. Diabetes Res Clin Pract 57: 119–129.
• 5   Mavros  Y,  Kay  S,  Anderberg  KA,  Baker  MK,  Wang  Y, et al. (2013) Changes in insulin resistance and HbA1c are related to exercise-mediated changes in body composition in older adults with type 2 diabetes: interim outcomes from the GREAT2DO trial. Diabetes Care 36: 2372–2379.
• 6   UK prospective diabetes  study (1995) UK Prospective Diabetes Study Group 16: overview of 6 years’ therapy of type II diabetes: a progressive disease. Diabetes 44: 1249–1258.
• 7   Bo  S,  Gentile  L,  Castiglione  A,  Prandi  V,  Canil  S, et al. (2012) C-peptide and the risk for incident complications and mortality in type 2 diabetic patients: a retrospective cohort study after a 14-year follow-up. Eur J Endocrinol 167: 173–180.
• 8   Talchai  C,  Xuan  S,  Lin  HV,  Sussel  L,  Accili  D (2012) Pancreatic beta cell dedifferentiation as a mechanism of diabetic beta cell failure. Cell 150: 1223–1234.
• 9   Wang  Z,  York  NW,  Nichols  CG,  Remedi  MS (2014) Pancreatic beta cell dedifferentiation in diabetes and redifferentiation following insulin therapy. Cell Metab 19: 872–882.
• 10   Weir  GC,  Aguayo-Mazzucato  C,  Bonner-Weir  S (2013) β-cell dedifferentiation in diabetes is important, but what is it? Islets 5: 233–237.
• 11   Rosenstock  J,  Dailey  G,  Massi-Benedetti  M,  Fritsche  A,  Lin  Z, et al. (2005) Reduced hypoglycemia risk with insulin glargine: a meta-analysis comparing insulin glargine with human NPH insulin in type 2 diabetes. Diabetes Care 28: 950–955.
• 12   Peyrot  M,  Rubin  RR,  Kruger  DF,  Travis  LB (2010) Correlates of insulin injection omission. Diabetes Care 33: 240–245.
• 13   Lajara  R,  Cengiz  E,  Tanenberg  RJ (2017) The role of the new basal insulin analogs in addressing unmet clinical needs in people with type 1 and type 2 diabetes. Curr Med Res Opin 33: 1045–1055.
• 14   Jonassen  I,  Havelund  S,  Hoeg-Jensen  T,  Steensgaard  DB,  Wahlund  PO, et al. (2012) Design of the novel protraction mechanism of insulin degludec, an ultra-long-acting basal insulin. Pharm Res 29: 2104–2114.
• 15   Heise  T,  Nosek  L,  Bottcher  SG,  Hastrup  H,  Haahr  H (2012) Ultra-long-acting insulin degludec has a flat and stable glucose-lowering effect in type 2 diabetes. Diabetes Obes Metab 14: 944–950.
• 16   Heise  T,  Hovelmann  U,  Nosek  L,  Hermanski  L,  Bottcher  SG, et al. (2015) Comparison of the pharmacokinetic and pharmacodynamic profiles of insulin degludec and insulin glargine. Expert Opin Drug Metab Toxicol 11: 1193–1201.
• 17   Heise  T,  Hermanski  L,  Nosek  L,  Feldman  A,  Rasmussen  S, et al. (2012) Insulin degludec: four times lower pharmacodynamic variability than insulin glargine under steady-state conditions in type 1 diabetes. Diabetes Obes Metab 14: 859–864.
• 18   Ratner  RE,  Gough  SC,  Mathieu  C,  Del Prato  S,  Bode  B, et al. (2013) Hypoglycaemia risk with insulin degludec compared with insulin glargine in type 2 and type 1 diabetes: a pre-planned meta-analysis of phase 3 trials. Diabetes Obes Metab 15: 175–184.
• 19   Lane  W,  Bailey  TS,  Gerety  G,  Gumprecht  J,  Philis-Tsimikas  A, et al. (2017) Effect of insulin degludec vs insulin glargine U100 on hypoglycemia in patients with type 1 diabetes: the SWITCH 1 randomized clinical trial. JAMA 318: 33–44.
• 20   Wysham  C,  Bhargava  A,  Chaykin  L,  de la Rosa  R,  Handelsman  Y, et al. (2017) Effect of insulin degludec vs insulin glargine U100 on hypoglycemia in patients with type 2 diabetes: the SWITCH 2 randomized clinical trial. JAMA 318: 45–56.
• 21   Ohkubo  Y,  Kishikawa  H,  Araki  E,  Miyata  T,  Isami  S, et al. (1995) Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: a randomized prospective 6-year study. Diabetes Res Clin Pract 28: 103–117.
• 22   Garber  AJ,  King  AB,  Del Prato  S,  Sreenan  S,  Balci  MK, et al. (2012) Insulin degludec, an ultra-long acting basal insulin, versus insulin glargine in basal-bolus treatment with mealtime insulin aspart in type 2 diabetes (BEGIN Basal-Bolus Type 2): a phase 3, randomised, open-label, treat-to-target non-inferiority trial. Lancet 379: 1498–1507.
• 23   Onishi  Y,  Iwamoto  Y,  Yoo  SJ,  Clauson  P,  Tamer  SC, et al. (2013) Insulin degludec compared with insulin glargine in insulin-naive patients with type 2 diabetes: a 26-week, randomized, controlled, Pan-Asian, treat-to-target trial. J Diabetes Investig 4: 605–612.
• 24   Giugliano  D,  Maiorino  MI,  Bellastella  G,  Chiodini  P,  Esposito  K (2011) Multiple HbA1c targets and insulin analogues in type 2 diabetes: a systematic review. J Diabetes Complications 25: 275–281.
• 25   Investigators  OT,  Gerstein  HC,  Bosch  J,  Dagenais  GR,  Diaz  R, et al. (2012) Basal insulin and cardiovascular and other outcomes in dysglycemia. N Engl J Med 367: 319–328.
• 26   Heise  T,  Korsatko  S,  Nosek  L,  Coester  HV,  Deller  S, et al. (2016) Steady state is reached within 2–3 days of once-daily administration of degludec, a basal insulin with an ultralong duration of action. J Diabetes 8: 132–138.
• 27   Tambascia  MA,  Eliaschewitz  FG (2015) Degludec: the new ultra-long insulin analogue. Diabetol Metab Syndr 7: 57.
• 28   Zinman  B,  Philis-Tsimikas  A,  Cariou  B,  Handelsman  Y,  Rodbard  HW, et al. (2012) Insulin degludec versus insulin glargine in insulin-naive patients with type 2 diabetes: a 1-year, randomized, treat-to-target trial (BEGIN Once Long). Diabetes Care 35: 2464–2471.
• 29   Rodbard  HW,  Gough  S,  Lane  W,  Korsholm  L,  Bretler  DM, et al. (2014) Reduced risk of hypoglycemia with insulin degludec versus insulin glargine in patients with type 2 diabetes requiring high doses of basal insulin: a meta-analysis of 5 randomized begin trials. Endocr Pract 20: 285–292.
• 30   Marso  SP,  McGuire  DK,  Zinman  B,  Poulter  NR,  Emerson  SS, et al. (2017) Efficacy and safety of degludec versus glargine in type 2 diabetes. N Engl J Med 377: 723–732.
• 31   Akirov  A,  Diker-Cohen  T,  Masri-Iraqi  H,  Shimon  I (2017) High glucose variability increases mortality risk in hospitalized patients. J Clin Endocrinol Metab 102: 2230–2241.
• 32   Tang  X,  Li  S,  Wang  Y,  Wang  M,  Yin  Q, et al. (2016) Glycemic variability evaluated by continuous glucose monitoring system is associated with the 10-y cardiovascular risk of diabetic patients with well-controlled HbA1c. Clin Chim Acta 461: 146–150.
• 33   Iga  R,  Uchino  H,  Kanazawa  K,  Usui  S,  Miyagi  M, et al. (2017) Glycemic variability in type 1 diabetes compared with degludec and glargine on the morning injection: an open-label randomized controlled trial. Diabetes Ther 8: 783–792.
• 34   Nakamura  T,  Sakaguchi  K,  So  A,  Nakajima  S,  Takabe  M, et al. (2015) Effects of insulin degludec and insulin glargine on day-to-day fasting plasma glucose variability in individuals with type 1 diabetes: a multicentre, randomised, crossover study. Diabetologia 58: 2013–2019.
• 35   Yamamoto  C,  Miyoshi  H,  Fujiwara  Y,  Kameda  R,  Ichiyama  M, et al. (2016) Degludec is superior to glargine in terms of daily glycemic variability in people with type 1 diabetes mellitus. Endocr J 63: 53–60.
• 36   Aso  Y,  Suzuki  K,  Chiba  Y,  Sato  M,  Fujita  N, et al. (2017) Effect of insulin degludec versus insulin glargine on glycemic control and daily fasting blood glucose variability in insulin-naive Japanese patients with type 2 diabetes: I’D GOT trial. Diabetes Res Clin Pract 130: 237–243.
• 37   Zinman  B,  Fulcher  G,  Rao  PV,  Thomas  N,  Endahl  LA, et al. (2011) Insulin degludec, an ultra-long-acting basal insulin, once a day or three times a week versus insulin glargine once a day in patients with type 2 diabetes: a 16-week, randomised, open-label, phase 2 trial. Lancet 377: 924–931.
• 38   Simioni  N,  Filippi  A,  Scardapane  M,  Nicolucci  A,  Rossi  MC, et al. (2017) Efficacy and safety of insulin degludec for hyperglycemia management in noncritical hospitalized patients with diabetes: an observational study. Diabetes Ther 8: 941–946.
• 39   Mathieu  C,  Hollander  P,  Miranda-Palma  B,  Cooper  J,  Franek  E, et al. (2013) Efficacy and safety of insulin degludec in a flexible dosing regimen vs insulin glargine in patients with type 1 diabetes (BEGIN: Flex T1): a 26-week randomized, treat-to-target trial with a 26-week extension. J Clin Endocrinol Metab 98: 1154–1162.
• 40   Meneghini  L,  Atkin  SL,  Gough  SC,  Raz  I,  Blonde  L, et al. (2013) The efficacy and safety of insulin degludec given in variable once-daily dosing intervals compared with insulin glargine and insulin degludec dosed at the same time daily: a 26-week, randomized, open-label, parallel-group, treat-to-target trial in individuals with type 2 diabetes. Diabetes Care 36: 858–864.
• 41   Nishimura  A,  Harashima  SI,  Fukushige  H,  Wang  Y,  Liu  Y, et al. (2017) A large difference in dose timing of basal insulin introduces risk of hypoglycemia and overweight: a cross-sectional study. Diabetes Ther 8: 385–399.
• 42   Galasso  S,  Facchinetti  A,  Bonora  BM,  Mariano  V,  Boscari  F, et al. (2016) Switching from twice-daily glargine or detemir to once-daily degludec improves glucose control in type 1 diabetes. An observational study. Nutr Metab Cardiovasc Dis 26: 1112–1119.
• 43   Zinman  B,  Marso  SP,  Poulter  NR,  Emerson  SS,  Pieber  TR, et al. (2018) Day-to-day fasting glycaemic variability in DEVOTE: associations with severe hypoglycaemia and cardiovascular outcomes (DEVOTE 2). Diabetologia 61: 48–57.