Dual action of imeglimin on insulin secretion and sensitivity in type 2 diabetes

Abstract

Aims

Imeglimin is a novel oral hypoglycemic agent approved for the treatment of type 2 diabetes in Japan, with dual actions to enhance insulin secretion and improve insulin sensitivity, suggested by preclinical evidence. However, the effect of imeglimin on tissue‐specific insulin sensitivity and glucose kinetics using glucose tracers is unclear.

Materials and Methods

In this single‐arm intervention study, 22 Japanese men with type 2 diabetes received imeglimin 2000 mg/day for 20 weeks. Glucose metabolism and insulin secretion were assessed by the 75‐g oral glucose tolerance test (OGTT) with double tracers at baseline, 1 week, and 20 weeks. Tissue‐specific insulin sensitivity and insulin clearance were also evaluated in 16 participants using a two‐step hyperinsulinemic–euglycemic clamp before and at 20 weeks. The primary endpoint was the change from baseline in the glucose area under the curve from 0 to 3 h (AUC_0–3h) during the OGTT at 20 weeks.

Results

The glucose AUC_0–3h during the OGTT was significantly decreased at 20 weeks of imeglimin administration (median change [interquartile range]: −108.63 [−148.25, −40.75] mg·h/dL, p = 0.0002). A similar reduction was already evident at 1 week. These reductions were mainly attributable to a decreased rate of oral glucose appearance. Insulin secretion indices were increased at 1 and 20 weeks. Clamp studies showed improved insulin sensitivity in skeletal muscle, liver, and adipose tissue, and increased insulin clearance.

Conclusions/Interpretation

These results provide the first comprehensive evidence in humans that imeglimin improves insulin secretion and insulin sensitivity in multiple tissues.

1. INTRODUCTION

Type 2 diabetes is a global health issue with a continuously increasing prevalence, and its prevention and improvement of glycemic control are urgent international priorities. ¹ The primary pathophysiology of type 2 diabetes involves impaired insulin secretion and reduced insulin sensitivity in insulin‐targeted tissues. ² , ³ Therapeutic agents that simultaneously act on both insulin secretion and insulin sensitivity have the potential to fundamentally improve the pathophysiology of diabetes.

Imeglimin is a new oral hypoglycemic agent approved in Japan for treatment of type 2 diabetes. Animal studies and in vitro research have suggested that imeglimin promotes insulin secretion and improves insulin sensitivity. ⁴ , ⁵ , ⁶ , ⁷ , ⁸ The efficacy and safety of imeglimin as a hypoglycemic agent for type 2 diabetes have been confirmed in Phase III clinical trials (TIMES 1–3). ⁹ , ¹⁰ , ¹¹ Regarding the mechanism of blood glucose‐lowering effects of imeglimin in humans, a hyperglycemic clamp test with type 2 diabetes showed insulin secretion‐promoting effects. ¹² However, no study has evaluated the effect of imeglimin on insulin sensitivity in humans using a reliable method. In particular, the hyperinsulinemic–euglycemic clamp, which is the gold standard assessment of tissue‐specific insulin sensitivity, has not yet been used with imeglimin.

Therefore, we conducted the Study to Investigate the Metabolic Action of Imeglimin on Patients with Type 2 Diabetes Mellitus (SISIMAI; jRCTs031210600). ¹³ This study aimed to determine the mechanism of the blood glucose‐lowering effect of imeglimin in patients with type 2 diabetes. We performed the 75‐g oral glucose tolerance test (OGTT) with double glucose tracers before the intervention and at 1 and 20 weeks after initiating the intervention. Quantitative evaluations of the rate of appearance of oral intake‐derived glucose (Ra oral), endogenous glucose production (EGP), and the rate of glucose disappearance (Rd during OGTT), as well as an analysis of insulin secretion responses, were performed. Additionally, a two‐step hyperinsulinemic–euglycemic clamp test with a glucose tracer was performed before and 20 weeks after initiating the intervention to evaluate tissue‐specific insulin sensitivity and insulin metabolism. Furthermore, we investigated changes in ectopic fat using ¹H magnetic resonance spectroscopy, as well as changes in body weight and body composition.

2. MATERIALS AND METHODS

2.1. Study design and ethical considerations

The study methods for the SISIMAI trial have been previously reported, ¹³ and a summary is provided below. This study was designed as a single‐arm intervention study. The study was conducted at Juntendo University Hospital, from February 2022 to March 2026. The protocol for this trial was approved by the Ethics Committee of Juntendo University (Approval Number: J21‐011) and registered in the Japan Registry of Clinical Trials (jRCTs031210600). This study was conducted in accordance with the principles of the Declaration of Helsinki and the Clinical Trials Act in Japan. Written informed consent was obtained from all participants after providing them with a thorough explanation of the study content.

2.2. Eligible patients

The study population consisted of patients with type 2 diabetes who were outpatients at Juntendo University Hospital. The main inclusion criteria were as follows: (1) patients with type 2 diabetes aged 20 years or older and younger than 65 years at the time of consent, (2) a BMI of 18.5 kg/m² or higher and less than 30.0 kg/m² at screening, (3) glycated haemoglobin A1c (HbA1c) value of 6.5% or higher and less than 10.0% at screening, and (4) not currently receiving treatment with any antidiabetic medications other than metformin. The main exclusion criteria included insulin‐dependent diabetes, severe liver, kidney, or heart disease, and malignant tumours.

2.3. Intervention

The participants were administered imeglimin 1000 mg orally twice daily for 20 weeks. Patients who were already taking metformin before the trial continued the same dose during the study period.

2.4. Efficacy endpoints

The primary endpoint was defined as the change in the AUC_0–3h of blood glucose concentrations during the 75‐g OGTT 20 weeks after starting imeglimin administration compared with before starting imeglimin administration.

Secondary endpoints included the change in the blood glucose AUC at 1 week after starting imeglimin administration, as well as changes in glucose kinetics, such as EGP, Ra oral, and glucose Rd, during the 75‐g OGTT. Further secondary endpoints were as follows: changes in insulin resistance and β‐cell function‐related indicators; tissue‐specific insulin sensitivity in the liver, skeletal muscle, and adipose tissue assessed by the hyperinsulinemic–euglycemic clamp; ectopic fat (intrahepatic lipid [IHL] and intramyocellular lipid [IMCL]) and visceral and subcutaneous fat areas; clinical indicators, including body weight, body composition, and blood pressure; and various other blood and urine tests.

2.5. Measurement and testing procedures

We conducted this study according to the research schedule shown in Figure S1. The main study procedures are summarised below. Detailed methodologies are described in the Supplementary Material.

2.6. Assessment of glucose kinetics and insulin sensitivity

Glucose kinetics were assessed using a 75‐g OGTT with double glucose tracers ([6,6‐²H₂]‐glucose and U‐[¹³C]‐glucose) at baseline and after 1 and 20 weeks of imeglimin treatment. Following an overnight fast, [6,6‐²H₂]‐glucose (Cambridge Isotope Laboratories, MA, USA) was infused intravenously as a primed (200 mg/m² body surface area [BSA])–continuous (2 mg/m² BSA/min) infusion for 3 h to achieve tracer equilibration (−180 to 0 min). Participants then ingested 75 g of glucose containing 1% U‐[¹³C]‐glucose (time 0). At the 1‐ and 20‐week visits, imeglimin (1000 mg) was administered orally 30 min before the OGTT. Glucose kinetics (EGP, oral Ra, and Rd) were calculated by standard methods. ¹⁴ , ¹⁵ , ¹⁶

A two‐step hyperinsulinemic–euglycemic clamp was performed at baseline and 20 weeks to evaluate tissue‐specific insulin sensitivity. After a 3‐h [6,6‐²H₂]‐glucose infusion to measure basal EGP, insulin was infused at 20 and 40 mU/m² BSA/min for 180 min each, following standard priming doses. ¹⁷ , ¹⁸ Plasma glucose was maintained at <95 mg/dL using variable 20% glucose infusion (<2.5% [6,6‐²H₂]‐glucose). Blood samples were obtained every 10 min during the final 30 min of the basal and each clamp period. Tissue glucose uptake (TGU) was normalised to fat‐free mass. Hepatic and adipose insulin sensitivities were calculated from the suppression of EGP and FFA during the first step (low‐dose insulin infusion), respectively, divided by steady‐state insulin concentrations; skeletal muscle insulin sensitivity was calculated as TGU during the second step (high‐dose insulin infusion) divided by steady‐state insulin concentrations. ¹⁹ The metabolic clearance rate of insulin (MCRI) was calculated as follows: MCRI = IIR/[SSSI − (BSI × SSSC/BSC)], ²⁰ where IIR is the insulin infusion rate, SSSI is the steady‐state serum insulin concentration, BSI is the basal serum insulin concentration, SSSC is the steady‐state serum C‐peptide concentration, and BSC is the basal serum C‐peptide concentration. Feedback inhibition of insulin secretion was estimated as follows: 1 − (SSSC/BSC). ²¹ , ²²

2.7. Assessment of body composition and ectopic f

IHL and IMCL content were quantified using ¹H‐magnetic resonance spectroscopy at baseline and 20 weeks. Visceral and subcutaneous fat areas were measured by magnetic resonance imaging at the same time points.

2.8. Other assessments

Maximal oxygen consumption (VO₂ peak) was estimated using a cycle ergometer at baseline and 20 weeks. Standard biochemical and hormonal analyses were performed on blood and urine samples collected at specified visits to evaluate safety and efficacy.

2.9. Statistical analyses

The primary efficacy analysis was conducted in the modified intention‐to‐treat (mITT) population. The safety analysis was conducted in the SS. The primary efficacy analysis (glucose AUC_0–3h during the OGTT) was performed using a paired t test analysis, as this endpoint was normally distributed and prespecified in the study protocol. For other comparisons before and after treatment, the Wilcoxon signed‐rank test was used. For comparisons involving three time points (week 0, week 1, and week 20), the Friedman test was applied. If this overall test was significant (p <0.05), further intra‐group comparisons were performed between week 0 and week 1, and between week 0 and week 20. Data are presented as median [25th–75th percentile]. The statistical significance level was set at p <0.05 (two‐sided). Statistical analyses were performed using SAS software (version 9.4; SAS Institute, Cary, NC, USA).

3. RESULTS

3.1. Patients and baseline characteristics

In this study, 39 patients were screened, and 26 enrolled (Figure S2). The safety set included 25 patients, excluding 1 who never took imeglimin. The mITT population comprised 22 patients after excluding 3 from the safety set: 1 with <75% adherence and missing primary data, 1 who failed to follow instructions, and 1 with pre‐existing muscle injury and unstable condition. These exclusions ensured data reliability and study validity. Among the mITT population, 1 patient discontinued due to facial nerve palsy, leaving 21 who completed the 20‐week evaluation.

The baseline characteristics of the mITT population (22 patients) are shown in Table 1. All participants were men, with a median (IQR) age of 57.5 (52.0–62.0) years and a median age at diagnosis of type 2 diabetes of 49.0 (41.0–58.0) years. The median baseline BMI was 24.8 (22.3–26.9) kg/m², and 12 (54.5%) participants were concomitantly taking metformin, with a mean dose of 1318.2 ± 385.7 mg/day.

TABLE 1.

Clinical and metabolic characteristics at baseline and at 20 weeks of imeglimin treatment.

Characteristics	Week 0	Week 20	p‐value
	n = 22	n = 21
Age at baseline, years	57.5 [52.0–62.0]	NA	NA
Age at T2DM diagnosis, years	49.0 [41.0–58.0]	NA	NA
Weight, kg	68.1 [63.0–77.9]	69.4 [63.3–81.1]	0.867
BMI, kg/m2	24.8 [22.3–26.9]	NA	NA
LBM, kg	50.9 [48.8–56.5]	51.0 [48.3–59.3]	0.212
Body fat percentage, %	24.2 [20.4–31.1]	25.6 [21.3–29.0]	0.789
SFA, cm2	162.61 [118.65–229.79]	170.86 [138.51–214.24]	0.408
VFA, cm2	111.02 [99.75–135.52]	105.87 [94.84–130.48]	0.602
IHL, %	9.86 [7.41–13.72]	9.01 [6.15–12.82]	0.269
IMCL (soleus)	10.65 [8.53–13.78]	11.82 [9.09–15.55]	0.019
IMCL (tibialis anterior)	2.89 [2.12–4.15]	4.20 [2.87–5.58]	0.154
RMR, kcal	1525.50 [1433.00–1637.00]	1500.00 [1432.00–1657.50]	0.133
Metabolic flexibility	0.12 [0.09–0.14]	0.12 [0.06–0.24]	0.637
VO2 peak, mL/kg/min	32.40 [26.40–35.90]	30.05 [26.95–33.55]	0.602
HbA1c, %	6.9 [6.7–7.5]	6.7 [6.2–7.3]	0.002
GA, %	17.9 [15.9–20.6]	15.9 [14.9–18.1]	<0.001
AST, U/L	21.0 [17.0–29.0]	20.0 [17.0–23.0]	0.003
ALT, U/L	21.0 [16.0–31.0]	19.0 [14.0–25.0]	<0.001
γ‐GTP, U/L	32.5 [25.0–51.0]	27.0 [21.0–37.0]	<0.001
TC, mg/dL	194.0 [170.0–224.0]	191.0 [171.0–212.0]	0.276
TGs, mg/dL	114.0 [86.0–158.0]	99.0 [78.0–156.0]	0.317
HDL‐C, mg/dL	58.5 [53.0–63.0]	56.0 [52.0–61.0]	0.360
LDL‐C, mg/dL	118.0 [94.0–140.0]	113.0 [94.0–141.0]	0.933
eGFR, mL/min/1.73 m2	78.3 [70.2–104.9]	79.7 [69.7–100.5]	0.960
Oxidised albumin, %	24.4 [21.3–26.1]	21.0 [20.2–24.0]	0.012
UACR	8.4 [4.4–15.3]	9.7 [7.8–23.5]	0.023

Note: Data are presented as the median [interquartile range]. p‐values were calculated using the Wilcoxon signed‐rank test to compare values between baseline and week 20.

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; eGFR, estimated glomerular filtration rate; GA, glycated albumin; γ‐GTP, gamma‐glutamyl transferase; HbA1c, glycated haemoglobin; HDL‐C, high‐density lipoprotein cholesterol; IHL, intrahepatic lipid; IMCL, intramyocellular lipid; LBM, lean body mass; LDL‐C, low‐density lipoprotein cholesterol; RMR, basal metabolic rate; SFA, subcutaneous fat area; T2DM, type 2 diabetes mellitus; TC, total cholesterol; TGs, triglycerides; UACR, urinary albumin‐to‐creatinine ratio; VFA, visceral fat area; VO₂ peak, maximal oxygen consumption.

3.2. Effects of imeglimin on glucose and insulin during the OGTT

The time course changes and AUC values in glucose, insulin, and C‐peptide concentrations, as well as other various hormones, during the 75‐g OGTT are shown in Figure 1. A summary of the various indicators obtained from the OGTT is shown in Table 2. The primary endpoint, glucose AUC_0–3h during the OGTT, significantly decreased at 20 weeks of imeglimin administration (paired t‐test p = 0.0002). The median glucose AUC_0–3h was also significantly decreased after 1 week of imeglimin administration. Additionally, a significant reduction in the glucose AUC from baseline was observed at 1 and 20 weeks over shorter (AUC_0–2h) and longer (AUC_0–4h) time frames.

FIGURE 1.

Time course and AUC of plasma concentrations during the 75‐g OGTT at week 0 and at 1 and 20 weeks of imeglimin treatment. Time course (A, C, E, G, I, K, M) and area under the curve (AUC) (B, D, F, H, J, L, N) of plasma glucose, insulin, C‐peptide, FFA, glucagon, GLP‐1, and GIP concentrations during the 75‐g OGTT in patients with type 2 diabetes mellitus. The tests were performed at week 0 (before the intervention, ●, grey), week 1 (1 week after initiating the intervention, ▲, blue), and week 20 (20 weeks after initiating the intervention, ◆, rose). Time course data are presented as the mean ± standard deviation. For the AUC panels, the bounds of the box represent the 25th and 75th percentiles, the line within the box marks the median, and the symbol (○ or +) indicates the mean. The whiskers extend to the minimum and maximum values, and outliers are plotted as individual points. Participants ingested 75 g of glucose at time 0. In the tests at weeks 1 and 20, the participants ingested 1000 mg of imeglimin 30 min before initiating the OGTT. Compared with week 0: *p <0.05, **p <0.01; NS, not significant. AUC, area under the curve; FFAs, free fatty acids; GLP‐1, glucagon‐like peptide‐1; GIP, glucose‐dependent insulinotropic polypeptide; OGTT, oral glucose tolerance test.

TABLE 2.

Parameters related to glucose metabolism during the 75‐g OGTT.

Parameter	Week 0	Week 1	Week 20	p‐value	p‐value	p‐value
	n = 22	n = 22	n = 21	Friedman test	Week 1 versus Week 0	Week 20 versus Week 0
Fasting glucose, mg/dL	148.5 [130.0–158.0]	133.5 [118.0–145.0]	134.0 [127.0–149.0]	0.011	0.003	0.078
Fasting insulin, μU/mL	6.2 [4.4–8.5]	5.8 [4.3–9.4]	6.3 [5.0–10.2]	0.648	NA	NA
Fasting C‐peptide, ng/mL	1.9 [1.5–2.7]	1.8 [1.5–2.3]	1.7 [1.5–2.4]	0.047	0.015	0.141
Fasting FFAs, μEq/L	635.5 [473.0–755.0]	624.5 [529.0–772.0]	693.0 [475.0–866.0]	0.566	NA	NA
Fasting glucagon, pg/mL	23.7 [17.7–37.6]	25.2 [20.2–31.0]	25.8 [18.3–34.9]	0.349	NA	NA
Fasting GLP‐1, pmol/L	1.6 [1.1–2.3]	2.1 [1.4–3.7]	2.1 [1.4–2.8]	0.072	NA	NA
Fasting GIP, pmol/L	5.1 [3.4–8.6]	4.6 [3.2–6.4]	6.2 [3.5–10.2]	0.653	NA	NA
AUC0–3h glucose, mg·h/dL	871.56 [792.88–935.75]	742.88 [665.00–836.13]	738.00 [691.00–776.38]	<0.001	<0.001	0.002
AUC0–3h insulin, μU·h/mL	67.76 [50.37–113.82]	94.73 [59.86–151.39]	86.43 [69.86–151.99]	<0.001	<0.001	0.001
AUC0–3h C‐peptide, ng·h/mL	15.34 [11.49–20.92]	17.58 [13.55–20.53]	18.48 [13.46–21.29]	0.035	0.043	0.016
AUC0–3h glucagon, pg.·h/mL	46.60 [32.83–68.30]	52.91 [44.39–71.84]	50.13 [41.05–75.51]	0.327	NA	NA
AUC0–3h FFAs, μEq·h/L	1009.75 [790.38–1313.25]	878.00 [738.50–979.00]	938.75 [831.50–1026.38]	0.389	NA	NA
AUC0–3h GLP‐1, pmol·h/L	11.46 [7.89–15.54]	17.61 [13.35–25.45]	14.98 [11.41–19.23]	0.010	0.003	0.046
AUC0–3h GIP, pmol·h/L	82.58 [72.44–98.034]	117.43 [83.89–131.89]	106.71 [87.98–118.14]	<0.001	<0.001	0.002
Insulinogenic index	0.13 [0.06–0.18]	0.13 [0.07–0.22]	0.16 [0.10–0.33]	0.011	0.038	0.004
AUC0–3h insulin/AUC0–3h glucose	0.08 [0.05–0.15]	0.13 [0.07–0.22]	0.13 [0.09–0.24]	<0.001	<0.001	<0.001
Matsuda ndex	4.63 [2.99–6.88]	4.65 [3.24–7.06]	4.88 [3.06–5.84]	0.925	NA	NA

Note: Data are presented as the median [interquartile range]. Fasting values represent samples taken at the −180‐min time point of the 75‐g OGTT. A closed testing procedure was used for statistical analysis. Overall changes across the three time points (week 0, week 1, and week 20) were assessed using the Friedman test. If this overall test was significant (p <0.05), further intra‐group comparisons were performed between week 0 and week 1, and between week 0 and week 20.

Abbreviations: GIP, glucose‐dependent insulinotropic polypeptide; GLP‐1, glucagon‐like peptide 1; FFAs, free fatty acids; NA, not applicable; OGTT, oral glucose tolerance test.

The insulinogenic index, an indicator of glucose‐responsive insulin secretion (GSIS), was increased at 1 and 20 weeks of imeglimin administration compared with baseline. Similarly, the AUC_0–3h insulin/glucose ratio, insulin AUC_0–3h, and C‐peptide AUC_0–3h were significantly increased at 1 and 20 weeks of imeglimin administration compared with baseline. The Matsuda index, an insulin sensitivity index calculated from the OGTT, showed no change. Additionally, the AUC_0–3h of glucagon‐like peptide‐1 (GLP‐1) and glucose‐dependent insulinotropic polypeptide (GIP) were significantly increased at 1 and 20 weeks compared with baseline. However, glucagon secretion (AUC_0–3h) showed no significant change at 1 or 20 weeks.

3.3. Effects of imeglimin on glucose metabolism during the OGTT

The time course changes and AUC values in glucose metabolism dynamics evaluated by the 75‐g OGTT with double glucose tracers are shown in Figure 2. Regarding Ra oral, the median AUC_0–4h at 1 week and that at 20 weeks were significantly reduced compared with the baseline median AUC_0–4h. Additionally, the time course of Ra oral showed a biphasic pattern, in which AUC_0–2h was significantly decreased at 1 and 20 weeks compared with baseline, while AUC_2–4h showed a significant increase at both time points compared with baseline. Therefore, the overall significant decrease in AUC_0–4hwas attributed to the marked decrease observed in the 0–2‐h period.

FIGURE 2.

Time course of glucose kinetics during the 75‐g OGTT at week 0 and at 1 and 20 weeks of imeglimin treatment. Time course (A, C, E) and area under the curve (AUC) (B, D, F) of the rate of appearance (Ra) of oral glucose, Ra of endogenous glucose production (EGP), and rate of disappearance (Rd) of glucose during the 75‐g OGTT in patients with type 2 diabetes mellitus. The tests were performed at week 0 (before the intervention, ●, grey), week 1 (1 week after initiating the intervention, ▲, blue), and week 20 (20 weeks after initiating the intervention, ◆, rose). Time course data are presented as the mean ± standard deviation. For the AUC panels, the bounds of the box represent the 25th and 75th percentiles, the line within the box marks the median, and the symbol (○ or +) indicates the mean. The whiskers extend to the minimum and maximum values, and outliers are plotted as individual points. Participants ingested 75 g of glucose at time 0. Compared with week 0: *p <0.05, **p <0.01; NS, not significant. AUC, area under the curve; EGP, endogenous glucose production; OGTT, oral glucose tolerance test; Ra, rate of appearance; Rd, rate of disappearance.

EGP showed no significant changes in AUC_0–4h at 1 and 20 weeks. However, when limited to the first 2 h, AUC_0–2h was significantly decreased at 1 week, but this effect was not observed at 20 weeks. Regarding Rd during the OGTT, which indicates total glucose disposal, no significant changes in AUC_0–4h were observed at 1 or 20 weeks. However, a time course analysis showed a biphasic response. Specifically, AUC_0–2h showed a significant decrease at both time points, while AUC_2–4h showed a significant increase compared with baseline.

3.4. Effect of imeglimin on tissue‐specific insulin sensitivity

To directly evaluate tissue‐specific insulin sensitivity, a two‐step hyperinsulinemic–euglycemic clamp test was performed in 16 participants before and at 20 weeks of imeglimin administration (Figure 3). At 20 weeks of imeglimin administration, TGU during the second step was significantly increased compared with baseline. The suppression rates of EGP and free fatty acids (FFAs) during the first step showed no significant improvement. In contrast, insulin sensitivity in skeletal muscle, liver, and adipose tissue, when each was normalised to the steady‐state serum insulin concentration during the corresponding insulin infusion step (i.e., low‐dose for liver and adipose tissue, high‐dose for skeletal muscle), was significantly increased at 20 weeks of imeglimin administration compared with baseline. Additionally, insulin clearance, represented by the MCRI, was significantly increased at 20 weeks of imeglimin administration compared with baseline. However, there were no significant changes in the feedback inhibition of insulin secretion by insulin, which indicates insulin sensitivity of pancreatic β‐cells.

FIGURE 3.

Effects of imeglimin on tissue‐specific insulin sensitivity and insulin kinetics assessed by a two‐step hyperinsulinemic–euglycemic clamp. (A) Tissue glucose uptake during the second step (high‐dose insulin infusion), (B) skeletal muscle insulin sensitivity, (C) suppression of EGP during the first step (low‐dose insulin infusion), (D) hepatic insulin sensitivity, (E) suppression of FFAs during the first step (low‐dose insulin infusion), and (F) adipose tissue insulin sensitivity, (G) insulin clearance during the first step, (H) insulin clearance during the second step, (I) feedback inhibition of insulin secretion during the first step, and (J) feedback inhibition of insulin secretion during the second step in patients with type 2 diabetes (n = 16) at week 0 (before the intervention) and week 20 (20 weeks after initiating the intervention). Data are presented as box‐and‐whisker plots. The bounds of the box represent the 25th and 75th percentiles, the line within the box marks the median, and the symbol (○ or +) indicates the mean. The whiskers extend to the minimum and maximum values, and outliers are plotted as individual points. Compared with week 0: *p <0.05, **p <0.01; NS, not significant. EGP, endogenous glucose production; FFM, fat‐free mass; FFAs, free fatty acids.

3.5. Effects of imeglimin on ectopic fat, body composition, and other metabolic markers

At 20 weeks of imeglimin administration, no significant change in IHL or IMCL in the tibialis anterior muscle, visceral fat area (VFA), or subcutaneous fat area. However, IMCL in the soleus muscle was significantly increased at 20 weeks of imeglimin administration compared with baseline. Additionally, no significant changes were observed in body composition indices, including body weight, skeletal muscle mass, and body fat percentage, as well as systemic energy metabolism‐related indices, including resting energy expenditure and VO₂ peak, following the 20‐week intervention (Table 1).

3.6. Effects of imeglimin on blood and urine test parameters

At 20 weeks of imeglimin administration, fasting blood tests showed that the median HbA1c value decreased from 6.9% (6.7%–7.5%) to 6.7% (6.2%–7.3%), and the median glycated albumin value decreased from 17.9% (15.9%–20.6%) to 15.9% (14.9%–18.1%). Additionally, liver enzymes were significantly decreased at 20 weeks of imeglimin administration compared with baseline. However, the urine albumin/creatinine ratio was significantly increased at 20 weeks of imeglimin administration compared with baseline. No significant changes were observed in concentrations of fasting blood glucose, C‐peptide, immunoreactive insulin, or glucagon, the lipid profile, or renal function (e.g., estimated glomerular filtration rate). Furthermore, the median oxidised albumin ratio, an indicator reflecting systemic oxidative stress, significantly decreased from 24.4% (21.3%–26.1%) to 21.0% (20.2%–24.0%) (Table 1).

3.7. Safety

In the safety analysis population (n = 25), 10 adverse events occurred in 8 participants (32.0%) (Table S1), none of which were serious. Eight events were mild and two were moderate. Gastrointestinal disorders (diarrhoea and nausea) were the most common, reported in 5 participants (20.0%), and were considered possibly related to imeglimin; however, no specific treatment was required. One participant (4.0%) experienced mild hypoglycemia, which also resolved without the need for intervention.

4. DISCUSSION

This SISIMAI trial investigated the mechanism of action of imeglimin in patients with type 2 diabetes by conducting the 75‐g OGTT with double tracers and a two‐step hyperinsulinemic–euglycemic clamp test with a glucose tracer. We found that imeglimin not only improved GSIS but also enhanced insulin sensitivity in insulin target tissues. Notably, the insulin secretagogue effect was observed as early as 1 week after starting imeglimin administration and was accompanied by increased incretin secretion. This enhancement of insulin secretion contributed to the suppression of hyperglycemia through suppression of Ra oral and EGP in the early postprandial phase, which was shown by a detailed analysis of glucose metabolism dynamics during the 75‐g OGTT. Additionally, the sustained blood glucose‐lowering effect observed at 20 weeks of imeglimin administration is likely to have been due, not only to these effects, but also to the improvement in insulin sensitivity. To the best of our knowledge, this is the first study to comprehensively examine the glucose‐lowering mechanism of imeglimin in humans. Although the observed reduction in HbA1c (from 6.9% to 6.7%) was statistically significant, it was modest in magnitude, suggesting a mild glycemic effect. However, given that imeglimin exerts dual actions by enhancing both insulin secretion and insulin sensitivity, it may be considered a suitable option for a wide range of patients with type 2 diabetes, consistent with current treatment guidelines that emphasise individualised therapy based on pathophysiology.

In this study, insulin secretion in response to blood glucose concentrations was significantly improved after 1 week of imeglimin treatment. A similar finding was reported in a previous hyperglycemic clamp study in patients with type 2 diabetes. ¹² Although our study did not investigate the underlying mechanisms directly, several preclinical studies have suggested possible pathways. For example, imeglimin was shown to increase NAD⁺ levels in rodent β‐cells via upregulation of nicotinamide phosphoribosyltransferase, thereby enhancing cyclic ADP‐ribose signalling and intracellular Ca²⁺ mobilisation, which may promote GSIS. ⁴ , ²³ Other animal studies have reported mitochondrial protective effects, such as reduced oxidative stress and β‐cell apoptosis. ²⁴ , ²⁵ As this latter mechanism requires time to develop, it is unlikely to explain the rapid improvement observed at 1 week. Consistent with previously reported, ²⁶ , ²⁷ GLP‐1 and GIP responses increased during glucose loading. Although we did not directly assess causality, this incretin response may have contributed to enhanced insulin secretion. Incretin hormones are also known to slow gastric emptying, which could explain the observed reduction in early‐phase oral glucose appearance. However, as gastric emptying was not directly measured, further studies are needed to clarify these mechanisms.

In this study, we demonstrated for the first time that imeglimin improved tissue‐specific insulin sensitivity using the hyperinsulinemic–euglycemic clamp method. In particular, the significant increase in TGU during the second step at 20 weeks, primarily in skeletal muscle, is noteworthy. No significant changes in body weight or body composition (e.g., skeletal muscle mass, IHL, and VFA) were observed in this study. Therefore, the improvement in insulin sensitivity was likely not due to a reduction in body fat and ectopic fat, but rather due to the direct action of imeglimin on insulin‐sensitive tissues. Previous preclinical studies have shown that imeglimin modulates mitochondrial function and enhances insulin signalling pathways through mechanisms such as reducing oxidative stress and activating AMPK in insulin target tissues, including the liver, muscle, and adipose tissue. ⁷ , ²⁸ Although such mechanisms were not directly assessed in the present study, they may help explain the observed multi‐organ insulin‐sensitising effects. The possibility cannot be ruled out that improved insulin sensitivity, resulting from improved blood glucose concentrations (i.e., improved glucotoxicity), contributed to the enhanced insulin sensitivity observed in this study. However, the extent of improvement of HbA1c values observed in this study was extremely small, suggesting that its contribution was likely small. Although insulin sensitivity was significantly improved based on clamp‐derived measurements, no significant change was observed in the Matsuda Index, a surrogate marker of insulin sensitivity derived from the OGTT. This discrepancy is likely due to the complex effects of imeglimin, which enhances insulin secretion and lowers blood glucose levels, thereby complicating the interpretation of OGTT‐based indices.

In this study, the administration of imeglimin resulted in improved insulin sensitivity and a significant increase in insulin clearance. The MCRI is an indicator primarily reflecting insulin degradation capacity in the liver. In states of impaired insulin sensitivity, the MCRI decreases as a compensatory mechanism, leading to hyperinsulinemia. ²⁹ Therefore, the increase in the MCRI observed in this study may have been due to improved insulin sensitivity in peripheral tissues, particularly in skeletal muscle, following imeglimin administration. Alternatively, imeglimin may directly affect insulin metabolic enzyme activity.

In this study, no significant changes were observed in indicators related to energy metabolism or fitness levels, such as resting energy expenditure, the respiratory quotient, or VO₂ peak. These results suggest that imeglimin may primarily influence the efficiency of insulin action rather than energy metabolism in the short term. In animal models, imeglimin has been reported to contribute to metabolic regulation through increased energy expenditure, maintenance of brown adipose tissue function, and improvement of the intestinal environment. ³⁰ Although this study did not evaluate brown adipose tissue activity or the gut microbiota, the absence of significant changes in indicators related to energy metabolism and physical performance suggests that this effect is limited in humans, or that it takes more time to manifest an effect than in animals.

In this study, a significant increase in IMCL was observed in the soleus muscle after 20 weeks of imeglimin administration, and a non‐significant but substantial increase was also seen in the tibialis anterior. Although the mechanisms remain unclear, one possible explanation is that enhanced glucose uptake in skeletal muscle may have shifted the preferred energy substrate towards lipid storage, thereby increasing IMCL as a physiological adaptation. Further investigation is warranted to clarify whether this change reflects beneficial metabolic flexibility or impaired lipid handling. Regarding UACR, although the median value remained within the normal range, a statistically significant increase was noted at 20 weeks. A previous report ³¹ showed that imeglimin tended to reduce UACR after 6 months of treatment, although the change did not reach statistical significance. In addition, an animal model demonstrated that longer‐term imeglimin treatment reduced renal fibrosis and improved kidney function, suggesting potential renoprotective effects, ³² although this finding may not fully translate to humans. In our study, no deterioration in renal function markers such as eGFR and serum creatinine was observed. Nevertheless, the increase in UACR highlights the need for careful monitoring of renal parameters in future long‐term studies of imeglimin to ensure safety.

In this study, serum oxidised albumin concentrations were significantly reduced 20 weeks after initiating imeglimin administration compared with baseline. Oxidised albumin increases under various pathological and physiological conditions, such as liver disease, renal failure, diabetes, hypertension, ageing, and after intense exercise, ³³ and has been considered an indicator reflecting systemic oxidative stress. Furthermore, recent ex vivo studies have suggested that oxidised albumin may be involved in endothelial dysfunction, inflammatory responses, and the induction of vascular damage. ³³ These findings have raised interest in the role of oxidised albumin not merely as a marker but also as a disease‐modifying factor. From this perspective, the reduction in the oxidised albumin ratio in this study suggests that imeglimin contributes to the prevention of the onset and progression of vascular complications in the long term. The mechanism underlying this effect may involve mitochondrial action, resulting in improved oxidative stress.

This study has several notable strengths and limitations. Its major strength is the combination of two precise methods, namely the two‐step hyperinsulinemic–euglycemic clamp and the 75‐g OGTT with double glucose tracers, allowing comprehensive and quantitative evaluation of imeglimin's mechanism of action. These methods allowed for the precise capture of the multifaceted effects of imeglimin. However, the study also has limitations. First, the single‐arm design without a placebo or active comparator group limits the ability to conclusively attribute observed effects solely to imeglimin, as improvements could partly reflect glycemic control or other confounding factors. Second, due to the invasiveness of the procedures, we faced challenges in recruiting a larger sample size, which may limit statistical power. Third, the study population comprised only relatively lean Japanese men with well‐controlled type 2 diabetes. In particular, the absence of female participants limits the ability to assess sex‐specific differences in the metabolic effects of imeglimin. Therefore, the generalizability of the findings to populations with different ethnic backgrounds, sex (especially women), BMI profiles, or poorer glycemic control remains uncertain. Fourth, the 20‐week duration may be insufficient to capture long‐term metabolic effects or safety outcomes. Future studies involving more diverse populations, longer observation periods, and randomised controlled designs are warranted to confirm and extend these findings.

In conclusion, this study suggests that imeglimin improves insulin secretion, insulin sensitivity, and insulin clearance in patients with type 2 diabetes, thereby improving metabolic homeostasis by restructuring systemic insulin dynamics and action. These findings provide strong evidence that imeglimin may be a useful option for individualised treatment in patients with type 2 diabetes with diverse pathologies.

AUTHOR CONTRIBUTIONS

T. T., H. K., and Y. T. contributed to the study design, data collection, interpretation of results, and manuscript preparation. N. I., T. K., H. N., and Sao. K. participated in data collection and contributed to the discussion. Sat. K., Y. N., and R. K. contributed to the discussion. H. W. contributed to the study design and edited the manuscript. H. K. and H. W. are the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

FUNDING INFORMATION

This study was supported by Sumitomo Pharma Co., Ltd., Tokyo, Japan.

CONFLICT OF INTEREST STATEMENT

This study was supported by Sumitomo Pharma Co., Ltd., Tokyo, Japan. Ryuzo Kawamori has received lecture fees from Sumitomo Pharma Co. and Novo Nordisk Pharma. Yoshifumi Tamura has received research funds from Astellas Pharma Inc., Kowa Pharmaceutical Co. Ltd., and Curves Holdings Co. Ltd., and Lotte Co. Ltd., and lecture fees from Boehringer Ingelheim, Sumitomo Pharma Co., Eli Lilly, Kowa Pharmaceutical Co., Mitsubishi Tanabe Pharma Co., Novo Nordisk Pharma, Ono Pharmaceutical Co., and Lion Co. Hirotaka Watada has received lecture fees from Novo Nordisk Pharma Ltd., Nippon Boehringer Ingelheim Co., Sumitomo Pharma Co., Ltd., Eli Lilly Japan K.K, Roche DC Japan, MSD K.K., Sanwa Kagaku Kenkyusho Co., Ltd., Daiichi Sankyo Ltd., Kyowa Kirin Co., Bayer Yakuhin, Ltd., Abbott Japan LLC, Taisho Pharmaceutical Co., LTD., Embecta, Teijin, GSK, Sanofi, and Mitsubishi Tanabe Pharma Corporation; research funding from Sanwa Kagaku Kenkyusho Co., Ltd., Kowa Company, Ltd., Nippon Boehringer Ingelheim Co., Ltd., Sumitomo Pharma Co., Ltd., Teijin Limited, LifeScan Japan, Sougouikagakukennkyusho, Taisho Pharmaceutical Co., LTD., Abbott Japan LLC, Takeda, and SBI Pharma Co., Ltd. Tsubasa Tajima, Hideyoshi Kaga, Naoaki Ito, Toshiki Kogai, Hitoshi Naito, Saori Kakehi, Satoshi Kadowaki and Yuya Nishida declare that they have no conflicts of interest.

Supporting information

Data S1. Supporting Information

DATA AVAILABILITY STATEMENT

The datasets generated during and/or analyzed during the current study are not publicly available due to sensitive nature of the human subject data and to protect the participants’ privacy but are available from the corresponding author upon reasonable request.