Insulin‐reducing effect of adding imeglimin to ongoing dipeptidyl peptidase‐4 inhibitor therapy with multiple daily insulin injections (INSPIRE study)

ABSTRACT

Introduction

Simplifying multiple daily insulin injections (MDI) has attracted great interest in clinical practice. Although imeglimin shows good compatibility with dipeptidyl peptidase‐4 inhibitors (DPP‐4i), its effectiveness in patients treated with MDI remains unclear. This study aimed to investigate the real‐world characteristics of such patients and the insulin‐reducing effects of adding imeglimin to ongoing DPP‐4i therapy.

Materials and Methods

Medical records of inpatients with type 2 diabetes (T2D) treated with imeglimin (2,000 mg/day) plus DPP‐4i and MDI were reviewed. Insulin dosage and self‐monitoring blood glucose (SMBG) values were compared between baseline and 5 days post‐imeglimin initiation. Patients were divided into bolus‐discontinuation (bolus insulin discontinued) and bolus‐continuation (bolus insulin continued) groups, and baseline characteristics were compared. The correlation between the reduction in bolus insulin (Δ bolus) and baseline parameters was assessed.

Results

Thirty patients (18 males and 12 females) were enrolled. Mean values were: age, 77.7 years; duration of T2D, 14.2 years; BMI, 21.6 kg/m², and C‐peptide reactivity index, 0.78. After imeglimin initiation, total insulin dosage decreased, with bolus insulin reduced by 10.6 ± 8.0 units/day. Pre‐breakfast and pre‐dinner SMBG values improved. Although insulin secretory capacity tended to be lower in the bolus‐continuation group, Δ insulin and Δ bolus did not differ between groups. Δ bolus showed no correlation with baseline BMI, duration of T2D, or insulin secretory capacity.

Conclusion

Adding imeglimin to DPP‐4i significantly reduced insulin dosage, even in patients with low insulin secretory capacity. This combination may represent a useful therapeutic option for Japanese patients with T2D treated with MDI.

INTRODUCTION

With the promotion of a patient‐centered approach and a rapidly aging population, treatment simplification has piqued huge interest in diabetes care ¹ . In particular, multiple daily insulin injections (MDI) require complicated procedures and treatment schedules and carry a high risk of hypoglycemia, making it difficult for older adults with impaired cognitive function and busy patients to continue. Moreover, it markedly decreases patients' quality of life owing to its huge treatment burden. For these reasons, simplifying treatment by reducing or discontinuing bolus insulin from MDI has emerged as an urgent need for both patients and healthcare professionals ² .

In the simplification of MDI, glinide and/or alpha‐glucosidase inhibitors (α‐GI) have frequently been used as a replacement for bolus insulin; however, their use is complex, requiring administration just before meals ³ , ⁴ . Conversely, adding an SGLT‐2 inhibitor (SGLT‐2i) or GLP‐1 receptor agonist (GLP‐1RA) to MDI significantly reduces insulin dosage, and their use with the aim of simplifying MDI is becoming more widespread ⁵ , ⁶ , ⁷ . However, these agents also lead to marked weight loss, which raises concerns regarding adverse events such as gastrointestinal symptoms, dehydration, and ketosis. Accordingly, a considerable proportion of patients, particularly older adults, face difficulties in simplifying MDI owing to the limited appropriate therapeutic options.

In Japan, dipeptidyl peptidase‐4 inhibitors (DPP‐4i) are the most commonly prescribed first‐line treatment for patients with type 2 diabetes (T2D) owing to their high efficacy, tolerability, and association with a low incidence of hypoglycemia and weight gain ⁸ , ⁹ . It significantly improves blood glucose levels and has good tolerability when combined with insulin ¹⁰ . For these reasons, DPP‐4i is frequently used as an additional treatment for MDI, especially in patients with restrictions on the use of other agents ⁹ . However, only a few cases involve bolus insulin discontinuation with DPP‐4i alone, with additional medication required in many cases.

Imeglimin, an oral anti‐diabetic agent, improves blood glucose levels through a mechanism different from that of existing drugs ¹¹ . In clinical trials, imeglimin in combination with DPP‐4 inhibitors has been shown to be safe and effective, demonstrating favorable efficacy and tolerability compared with other combinations ¹² . Additionally, combination therapy with these agents significantly improves postprandial blood glucose levels and blood glucose fluctuations as assessed by continuous glucose monitoring and meal tolerance test ¹³ . Furthermore, imeglimin has shown significant improvement in blood glucose levels and high tolerability even when used in combination with basal insulin ¹⁴ . However, no reports have shown the effects of combining imeglimin with MDI, and it is unclear whether imeglimin contributes to reducing insulin (basal and bolus) dosage. Moreover, it is unclear in what kind of cases it is being used or would be more effective.

In this study, through a review of medical records, we investigated the characteristics of patients treated by adding imeglimin to MDI already combined with DPP‐4i and evaluated the insulin‐reducing effect at a single institute.

MATERIALS AND METHODS

Study design and population

This single‐center, retrospective, observational study was conducted using an opt‐out approach and was registered in the University Hospital Medical Information Network Clinical Trial Registry (UMIN000058691), a non‐profit organization in Japan that complies with the requirements of the International Committee of Medical Journal Editors. The study was approved by the Medical Ethics Committee of Toho University Omori Medical Center (Approval No. M24143, October 30, 2024) and conducted in accordance with the principles of the Declaration of Helsinki and the current legal regulations in Japan. The requirement for informed consent was waived by the institutional review board because of the retrospective nature of the study, and an opt‐out approach was used. Participants were recruited from among inpatients of Toho University Omori Medical Center between September 2021 and January 2025. They comprised Japanese patients with T2D who were treated with imeglimin (2000 mg/day) for 5 days or more in addition to combination therapy with MDI and DPP‐4i. Patients who used oral hypoglycemic agents other than DPP‐4 inhibitors prior to starting imeglimin were excluded from the screening. Imeglimin was started when the attending physician determined that a certain degree of blood glucose control had been achieved through MDI and DPP‐4i treatment. Those who experienced changes in their condition after the addition of imeglimin that could affect glycemic control, those who had been treated with steroids or other medications influencing glycemic control, and those who were otherwise considered unsuitable for inclusion in the study were excluded. All inpatients who met the inclusion criteria during the study period were systematically screened and their data extracted for analysis.

Their glucose values were monitored using self‐monitoring blood glucose (SMBG) 4 times daily. After initiating imeglimin, insulin dosage was adjusted at the discretion of the attending physician. Discontinuation of bolus insulin was also determined by the attending physicians. We collected the data below from medical records: biochemical and anthropometric data recorded at the time of admission, insulin dosage, and SMBG values of the day before and 5 days after the initiation of imeglimin.

Study outcomes

The characteristics of the patients treated with additional imeglimin in combination with DPP‐4i and MDI were evaluated. Insulin dosage and SMBG values were compared at baseline and 5 days after imeglimin initiation. As an exploratory analysis to identify cases suitable for imeglimin, the following assessments were performed. (1) all patients were divided into two groups—a bolus‐discontinuation group (bolus insulin could be discontinued during the period from the introduction of imeglimin until discharge) and a bolus‐continuation group (bolus insulin could not be discontinued within discharge, or additional medication was required)—and their baseline characteristics were compared; and (2) correlation analyses were performed between the magnitude of change in insulin dosage (Δ insulin) after adding imeglimin and baseline data.

Statistical analysis

To our knowledge, no studies have shown the magnitude of insulin dosage reduction achieved by adding imeglimin on MDI, with or without other agents. To address this concern, the sample size calculation for this study was based on the previously reported magnitude of insulin dosage reduction achieved with other agents in Japanese patients with T2D (7). In that study, the mean reduction in insulin dosage was −13.3 ± 7.7 units with SGLT‐2i and −14.7 ± 7.5 units with GLP‐1RA. Therefore, the required sample size for this study was determined at 21 participants to achieve a 5% significance level and 80% statistical power. The maximum number of patients who met the eligibility criteria was then enrolled based on their medical records. Data analyses were conducted using Statcel (OMS, Saitama, Japan). Values obtained at baseline and 5 days after imeglimin initiation were compared using a paired t‐test. Comparisons between groups were performed using Mann–Whitney's U test. Correlation analyses were performed using Pearson's correlation coefficient. All data are presented as mean ± standard deviation, and statistical significance was set at P < 0.05.

RESULTS

Patients' characteristics

Figure 1 shows the flow chart of patient enrollment. Fifty‐one patients were screened, and those who did not meet the inclusion criteria or whose imeglimin was discontinued before completing the 5‐day course by their attending physician were excluded. In particular, among the patients who discontinued imeglimin within a short period, three were found to have discontinued it due to insufficient efficacy. Therefore, 30 patients met the inclusion criteria. The reasons for hospitalization among the patients were blood glucose control (10 cases) and treatment of other diseases (20 cases). Baseline characteristics are shown in Table 1. Most of the patients were older adults (mean age 77.7 ± 9.2 years) and were not obese (average BMI 21.6 ± 4.20 kg/m²). The average duration of T2D was 14.2 ± 13.6 years, with an average glycated hemoglobin (HbA1c) level of 9.03 ± 1.89%. Insulin secretory capacity tended to be low (average serum C‐peptide reactivity (CPR) index: 0.78 ± 0.43, urinary CPR: 31.1 ± 33.2 μg/day). Imeglimin was initiated 16.1 ± 9.5 days after admission.

Figure 1.

Flowchart of participant selection.

Table 1.

Baseline characteristics of patients in this study

N = 30
Male/Female	18/12
Age (years)	77.7 ± 9.2
Duration of T2D (years)	14.2 ± 13.6
Height (cm)	157.2 ± 8.17
Body weight (kg)	53.6 ± 11.8
BMI (kg/m2)	21.6 ± 4.20
Systolic blood pressure (mmHg)	128.2 ± 19.0
Diastolic blood pressure (mmHg)	76.9 ± 12.5
HbA1c (%)	9.03 ± 1.89
Cr (mg/dL)	0.75 ± 0.25
eGFR (mL/min/1.73)	70.6 ± 21.6
AST (U/L)	22.6 ± 10.0
ALT (U/L)	20.5 ± 15.0
Triglycerides (mg/dL)	115.7 ± 82.0
Total cholesterol (mg/dL)	166.7 ± 36.8
LDL cholesterol (mg/dL)	97.6 ± 26.7
CPR index	0.78 ± 0.43
Urine CPR (μg/day)	31.1 ± 33.2
Urinary albumin‐to‐creatinine ratio (mg/gCr)	152.7 ± 388.3
Total daily insulin (U/day)	25.3 ± 10.7
Basal insulin (U/day)	7.6 ± 6.0
Bolus insulin (U/day)	18.1 ± 8.7
Diabetic retinopathy (−/+/unknown)	17/2/11

Data are presented as mean ± SD.

Insulin dosage and glycemic control

Table 2 shows the insulin dosage at baseline and 5 days after imeglimin initiation. Both basal and bolus insulin were significantly reduced after imeglimin initiation, with the bolus insulin dose markedly reduced by 10.6 ± 8.0 units per day. SMBG values are shown in Figure 2 and Table S2. Despite the significant reduction in basal and bolus insulin after imeglimin initiation, pre‐breakfast and pre‐dinner blood glucose levels improved significantly, while pre‐lunch and bedtime levels remained the same.

Table 2.

Changes in insulin dosage before and after adding imeglimin

N = 30	Before	5 days after	P
Total daily dosage (U/day)	25.3 ± 10.7	12.5 ± 11.2	<0.01*
Δ Total daily dosage (U/day)		−12.8 ± 10.0
Basal insulin (U/day)	7.57 ± 5.99	5.57 ± 5.57	<0.01*
Δ Basal insulin (U/day)		−1.93 ± 3.62
Bolus insulin (U/day)	18.1 ± 8.66	7.53 ± 9.39	<0.01*
Δ Bolus insulin (U/day)		−10.57 ± 8.02

Data are presented as mean ± SD. P‐values indicate differences from previous values.

^* P < 0.05 in Paired t‐test.

Figure 2.

Changes in SMBG values from baseline to 5 days after adding imeglimin The dotted and solid lines indicate the SMBG values before and after adding imeglimin, respectively. *P < 0.05 for differences from baseline in paired t‐test.

Comparison of the bolus‐discontinuation and bolus‐continuation groups

In the bolus‐discontinuation group, the mean duration to discontinuation of bolus insulin was 4.13 ± 2.22 days after initiating imeglimin. Tables 3 and 4 show baseline characteristics and insulin dosages before and after imeglimin initiation in both groups. In two patients in the bolus‐discontinuation group, because bolus insulin was discontinued after 6 days or more post‐imeglimin initiation, their bolus insulin dosage at day 5 was not zero. Patients in this group had a significantly shorter duration of T2D, tended to have higher insulin secretory capacity and BMI compared with the bolus‐continuation group. In contrast, the bolus‐continuation group had significantly higher total daily dosage (TDD) and bolus insulin dosage before imeglimin initiation, and the difference remained after imeglimin initiation. However, focusing on Δ insulin after adding imeglimin, no significant differences were observed in Δ TDD, basal, and bolus insulin. In particular, Δ bolus insulin decreased by approximately 10 units/day in both groups. These findings suggest that the discontinuation or continuation of bolus insulin depends on the amount of insulin administered.

Table 3.

Comparison of baseline characteristics between the bolus‐discontinuation and bolus‐continuation groups

	Bolus‐discontinuation	Bolus‐continuation	P
N (F)	15 (5)	15 (7)
Age (years)	76.4 ± 12.1	79.1 ± 5.1	0.92
Duration of T2D (years)	9.6 ± 13.7	20.0 ± 11.7	<0.01*
Prescribed caloric intake (kcal/kg/day)	28.36 ± 1.36	28.45 ± 1.89	0.69
Mean length of hospital stays (days)	35.3 ± 25.8	27.5 ± 14.1	0.88
Height (cm)	158.0 ± 6.95	156.4 ± 9.39	0.42
Body weight (kg)	56.2 ± 12.3	50.9 ± 10.96	0.24
BMI (kg/m2)	22.5 ± 4.78	20.7 ± 3.46	0.31
Systolic blood pressure (mmHg)	120.7 ± 15.3	135.7 ± 19.3	0.051
Diastolic blood pressure (mmHg)	74.3 ± 12.3	79.5 ± 12.1	0.289
HbA1c (%)	9.19 ± 2.23	8.86 ± 1.50	1.00
Cr (mg/dL)	0.82 ± 0.29	0.67 ± 0.19	0.28
eGFR (mL/min/1.73)	62.4 ± 22.6	79.3 ± 17.3	<0.01*
AST (U/L)	25.3 ± 12.4	19.9 ± 5.7	0.329
ALT (U/L)	25.7 ± 18.1	15.4 ± 8.2	0.061
Triglycerides (mg/dL)	140.2 ± 98.5	85.1 ± 36.3	0.155
Total cholesterol (mg/dL)	163.9 ± 41.9	170.2 ± 28.8	0.829
LDL cholesterol (mg/dL)	96.3 ± 27.6	97.6 ± 22.7	0.836
CPR (ng/mL)	1.36 ± 0.81	1.08 ± 0.69	0.34
CPR index	0.90 ± 0.42	0.65 ± 0.42	0.08
Urine CPR (μg/day)	45.3 ± 42.9	17.9 ± 11.5	0.06
Urinary albumin‐to‐creatinine ratio (mg/gCr)	229.5 ± 548.3	86.1 ± 107.1	0.678
Total daily insulin (U/day)	21.1 ± 10.1	29.5 ± 9.75	0.04*
Basal insulin (U/day)	7.87 ± 6.78	7.27 ± 5.30	0.93
Bolus insulin (U/day)	13.2 ± 6.79	22.9 ± 7.70	<0.01*
Diabetic retinopathy (−/+/unknown)	9/1/5	8/1/6

Data are presented as mean ± SD. P‐values indicate differences between two groups.

^* P < 0.05 in Mann–Whitney's U test. Missing data were excluded from the analysis.

Table 4.

Comparison of change in insulin dosage between the bolus‐discontinuation and bolus‐continuation groups

	Bolus‐discontinuation	Bolus‐continuation	P
Total daily dosage (U/day)
Before	21.1 ± 10.1	29.5 ± 9.75	0.04*
5 days after	7.00 ± 7.39	19.3 ± 11.0	<0.01*
Δ Total daily dosage	−14.5 ± 10.9	−10.2 ± 8.23	0.44
Basal insulin (U/day)
Before	7.87 ± 6.78	7.27 ± 5.30	0.93
5 days after	5.00 ± 6.31	5.80 ± 4.93	0.57
Δ Basal insulin	−2.40 ± 4.36	−1.47 ± 2.77	0.97
Bolus insulin (U/day)
Before	13.3 ± 6.79	22.9 ± 7.70	<0.01*
5 days after	1.33 ± 3.83	13.7 ± 9.25	<0.01*
Δ Bolus insulin	−11.9 ± 7.54	−9.20 ± 8.51	0.59

Data are presented as mean ± SD. P‐values indicate differences between two groups.

^* P < 0.05 in Mann–Whitney's U test.

Correlation between baseline characteristics and Δ bolus insulin

To assess whether baseline characteristics were associated with the high insulin reduction effect of imeglimin, correlation analyses were performed between baseline characteristics and Δ bolus insulin (Table S1). No significant correlations were observed between Δ bolus insulin and duration of T2D, BMI, kidney function, or insulin secretory capacity.

Adverse events

Adverse events were observed in four patients, leading to the discontinuation of imeglimin. In all cases where imeglimin was discontinued, the patients were using only imeglimin and DPP‐4i. These events were primarily gastrointestinal, including loss of appetite (n = 1) and vomiting (n = 2). One patient had decreased renal function, which did not improve after discontinuation of imeglimin, and the causal relationship was unclear.

DISCUSSION

In this retrospective observational study, we evaluated, for the first time, the actual situation regarding the use of imeglimin for simplifying MDI and its insulin reduction effect as an adjunct to DPP‐4i in Japanese patients with T2D. In Japan, the average age and BMI of patients with T2D have been reported as 67.4 years and 24.9 kg/m², respectively ¹⁵ . Compared with these general characteristics, our patients were older and had lower BMI. Moreover, their insulin secretory capacity tended to be low (CPR index <1.0). After initiating imeglimin, both basal and bolus insulin were significantly reduced, resulting in TDD being reduced by approximately 12.8 ± 10.0 units per day. Particularly, bolus insulin was markedly reduced by approximately 10.6 ± 8.0 units per day, causing half of the patients to discontinue bolus insulin. After initiating imeglimin, the blood glucose levels pre‐breakfast and pre‐dinner significantly improved, but those pre‐lunch and at bedtime remained at the same level as before initiation (Figure 2 and Table S2). Imeglimin was initiated when the attending physician determined that blood glucose control was stable. The insulin requirement was carefully adjusted by the attending physician to prevent worsening of blood glucose levels while minimizing the use of bolus insulin. Therefore, bolus insulin was significantly reduced, and pre‐lunch and bedtime blood glucose levels returned to the same level as before imeglimin initiation. As an adjunct therapy to MDI, adding GLP‐1RA significantly reduces TDD by approximately 12 units per day and bolus insulin by 7 units per day ⁵ , ⁷ , whereas SGLT‐2i reduces TDD by approximately 9–11 units per day ⁵ , ⁶ . The use of GLP‐1RA and/or SGLT2i to MDI has recently become popular for simplifying treatment. However, despite their excellent insulin‐reducing properties, they are also associated with significant weight loss and potential adverse events such as gastrointestinal symptoms, dehydration, and ketosis ⁵ , ¹⁶ . In contrast, combining glinide and α‐GI with MDI can significantly reduce bolus insulin without increasing the risk of hypoglycemia or body weight loss ³ . However, both agents require complicated administration methods, limiting their use in patients with difficulty managing their medications. For these reasons, imeglimin has often been selected when DPP‐4i monotherapy is insufficient to simplify MDI in patients with low BMI or a high risk of adverse events owing to low activities of daily living or dementia.

Notably, in this study, adding imeglimin to basal‐bolus therapy with DPP‐4i significantly reduced insulin dosage, comparable with that achieved with GLP‐1RA or SGLT‐2i. In particular, bolus insulin was reduced significantly, resulting in a decrease in the number of insulin injections required. To date, to our knowledge, no studies have shown the effects of imeglimin on insulin dosage or patient characteristics, leaving its effectiveness unknown. However, previous clinical trials have shown that imeglimin improves not only fasting glucose but also postprandial glucose and glucose fluctuation ¹¹ , ¹⁷ , ¹⁸ , ¹⁹ . Furthermore, combining imeglimin with a DPP‐4i produces a synergistic effect on the circulation of insulin and GLP‐1 through different mechanisms, resulting in a greater glucose‐lowering effect ¹² , ¹³ . These findings suggest that the postprandial glucose‐lowering effect of imeglimin is enhanced by combination with DPP‐4i, contributing to a significant reduction and consequent discontinuation of bolus insulin. In addition, imeglimin contributed to a decrease in basal insulin requirements, indicating potential beneficial effects not only on postprandial glucose but also on fasting glucose control, further supporting its potential as an agent to simplify insulin therapy.

While glinides have been widely used as an adjunct to MDI in older patients owing to their low risk of hypoglycemia, body weight loss, dehydration, and ketosis, their glucose‐lowering effect depends on insulin secretory capacity ³ , making them less effective in patients with reduced capacity. Notably, in this study, addition of imeglimin to DPP‐4i resulted in a significant decrease in insulin dosage despite low insulin secretory capacity. Although the bolus‐continuation group had lower insulin secretory capacity, the magnitude of insulin reduction after adding imeglimin was comparable between both groups. Similarly, the TIMES 3 phase 3 imeglimin trial ¹⁴ showed that despite patients having a longer duration of T2D and lower BMI than those in other phase 3 imeglimin trials ¹² , ¹⁹ , comparable improvements in HbA1c were observed. These findings may indicate that imeglimin is effective even in patients with low insulin secretory capacity. Recent systematic reviews and meta‐analyses have reported that imeglimin improves β‐cell function and insulin sensitivity, resulting in sustained glycemic control ²⁰ , ²¹ , ²² . These findings are consistent with our results and support the clinical utility of imeglimin, even in patients with relatively low insulin secretory capacity.

To identify patients in whom imeglimin would be effective, a correlation analysis between Δ bolus insulin and baseline parameters was performed. No correlation was observed with insulin secretory capacity, BMI, or duration of T2D. In contrast, recent meta‐analyses have shown that the efficacy of imeglimin is increased in patients with lower BMI and lower insulin resistance ²³ , ²⁴ . The difference between our results and previous reports may be attributable to our small sample size. Nevertheless, our patients' characteristics were similar to those of populations in which imeglimin was effective in previous studies. Moreover, our finding of a marked reduction in insulin dosage with the addition of imeglimin may support previous results. In this regard, dulaglutide has often been selected in recent years as a GLP‐1 receptor agonist with minimal impact on body weight for older patients with T2D. Usui et al. ²⁵ reported a C‐peptide index of 1.103 as the cutoff for achieving an HbA1c of 7% in their study with liraglutide. Although our patients had even lower insulin secretion capacity, they still showed a significant insulin‐reducing effect. Overall, these results suggest that imeglimin may be more suitable for patients with lower insulin secretory capacity and insulin resistance. DPP‐4is have been well established to be more effective in Asian patients with lower BMI and insulin resistance compared with White patients (8), and the characteristics of suitable patients are similar to those for imeglimin. Therefore, combination therapy with imeglimin and DPP‐4i may represent a new option owing to good compatibility in glucose‐lowering action and similar suitable patient profiles. However, little is known about the efficacy and suitability of imeglimin, and further studies in larger populations should be conducted to gain further insights.

This study has some limitations. First, we screened inpatients at a single university hospital in Japan. Consequently, these patients had a relatively lower BMI and were older than the general population; therefore, our results may not be applicable to all patients with T2D. Second, this study was focused on hospitalized patients; thus, it is unclear whether similar results would be obtained in outpatient settings. Third, we confirmed only 5 days of imeglimin initiation; therefore, the longer‐term insulin‐reducing effects remain unclear. Fourth, we were unable to verify the insulin‐reducing effect of imeglimin alone. Therefore, large‐scale, prospective studies are needed to evaluate the insulin‐reducing effects of imeglimin. However, this study revealed that the addition of imeglimin to DPP‐4i reduced bolus insulin by approximately 10.6 ± 8.0 units per day, thereby simplifying MDI despite the short 5‐day observation period.

In conclusion, adding imeglimin to DPP4i significantly reduced insulin even in patients with low insulin secretory capacity, indicating its potential as a new therapeutic option for Japanese patients with T2D treated with MDI and DPP‐4i.

DISCLOSURE

Sumitomo Pharma is engaged in the manufacturing and sale of imeglimin in Japan. T. Hirose received lecture fees from Novo Nordisk Pharma Ltd., MSD K.K., Sumitomo Pharma Co., Ltd., Kowa Company, Ltd., Abbott Japan LLC, Eli Lilly Japan K.K., Sanofi K.K., and Mitsubishi Tanabe Pharma Corporation, and received research funds from Sumitomo Pharma Co., Ltd., Nippon Boehringer Ingelheim Co., Ltd., and Mitsubishi Tanabe Pharma Corporation. None of the funding agencies had any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. Fukumi Yoshikawa received lecture fees from Novo Nordisk Pharma Ltd. Atsushi Itsukaichi declares no conflicts of interest.

Approval of the research protocol: The study was approved by the Medical Ethics Committee of Toho University Omori Medical Center (Approval No. M24143, October 30, 2024) and conducted in accordance with the principles of the Declaration of Helsinki and the current legal regulations in Japan.

Informed consent: N/A.

Registry and the registration no. of the study/trial: This study was registered in the University Hospital Medical Information Network Clinical Trial Registry (UMIN000058691), a non‐profit organization in Japan that complies with the requirements of the International Committee of Medical Journal Editors.

Animal studies: N/A.

Supporting information

Table S1. Correlation analysis between Δ bolus insulin and baseline values.

Table S2. Changes in blood glucose levels from baseline to 5 days after imeglimin administration.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.