Search for a command to run...
Imeglimin is a novel class of therapeutic agents for type 2 diabetes which has a structure related to metformin. Based on basic research findings, imeglimin enhances glucose-induced insulin secretion and suppresses gluconeogenesis in rat hepatocyte, suggesting that imeglimin may improve both insulin secretion and insulin sensitivity [1]. Although previous clinical trials have demonstrated improvements in glucose tolerance and enhanced insulin secretion with imeglimin, HOMA-IR, a widely used index of insulin sensitivity, has generally remained unchanged [2, 3]. Consistently, our previous findings demonstrated that imeglimin significantly increased HOMA-β and the insulinogenic index; however, HOMA-IR and the Matsuda index showed trends toward improvement but the differences did not reach statistical significance in patients with type 2 diabetes [4]. Therefore, the evidence supporting its effect on insulin sensitivity remains limited. Furthermore, a previous report demonstrated that HOMA-IR and Matsuda index are not suitable for estimating insulin sensitivity in longitudinal settings [5], raising the need for alternative methods to assess insulin sensitivity. Oral glucose minimal model is the mathematical approach to estimate insulin sensitivity using the parameter, plasma glucose and insulin concentrations during oral glucose tolerance test (OGTT), and has been validated by the hyperinsulinemic-euglycemic clamp which considered the standard for evaluating insulin sensitivity [6-8] and a clinical trial using the oral minimal model demonstrated improvements in insulin sensitivity with liraglutide [9]. Oral C-peptide model can evaluate dynamic nature of β-cell function: total β-cell responsivity (φ), which integrates both dynamic and static components of the insulin response to glucose; the basal insulin secretion (φb), the static component (φs) reflecting sustained insulin secretion at steady state; and the dynamic component (φd), representing early-phase insulin secretion in response to the rate of increase in glucose [10-12]. In the present study, we conducted a sub-analysis of our previous study using the oral minimal model to further assess the effect of imeglimin and metformin on insulin sensitivity and insulin secretion. This was a sub-analysis of a previous single-centre, prospective, randomised, double-arm, open-label trial conducted in Kansai Electric Power Hospital. The protocol (jRCTs051220075) was approved by the ethics committee. The details of the study protocol are described in the previous article [4]. Briefly, eligible patients with type 2 diabetes who received no anti-diabetic drugs or have completed the 8-week washout period for anti-diabetic drugs were randomly allocated by the randomly permuted block method with stratification by BMI and HbA1c by the staff from outside the hospital to imeglimin treatment (IME, 2000 mg/day) or metformin treatment (MET, 1000 mg/day). OGTT was performed before, and 12 and 24 weeks after the initiation of imeglimin or metformin treatment. Blood samples were collected at 0, 10, 20, 30, 60, 90, 120, 150, 180 and 240 min after a 75 g glucose load. Written informed consent was obtained from all participants. Plasma glucose, insulin and C-peptide levels during OGTT were used to obtain SI (sensitivity index), φb (basal), φd (dynamic), and φs (static), and φ (total) by oral glucose minimal model and the oral C-peptide minimal model, respectively. The estimation of these parameters was conducted according to the previous studies [6, 10-12]. The detailed equations are presented in Table S1. All estimation procedures were carried out using MATLAB version R2024a (MathWorks). The sample size was determined to detect the change in the primary end point as previously reported [4]. Changes in SI and φ were defined as the secondary outcomes in the protocol. The results are written as the mean ± SEM. Clinical parameters were compared between the time points in each group using paired t-test or Wilcoxon signed-rank test with Bonferroni correction. For group comparison, repeated measures were analysed using a mixed effect model for repeated measures (MMRM). Statistical analysis was carried out using SPSS Statistics 28 software (IBM Corp.) and p values < 0.05 were considered statistically significant. Baseline characteristics were similar between the two groups (Table S2). As a main result of our previous report [4], area under the curve (AUC) for glucose during OGTT was significantly decreased at 12 and 24 weeks both in IME and MET to a similar extent. However, significant increase in insulin was observed only in IME, suggesting that enhancement of insulin secretion contribute to the mechanism of glucose-lowering effect of IME. Consistently, C-peptide levels numerically but not statistically increased (Table S3). When assessed with oral glucose minimal model, SI was gradually and significantly increased over time in IME (0 week; 30.22 ± 2.39, 12 weeks; 38.69 ± 4.61, 24 weeks; 55.78 ± 14.77* × 10−5dL/kg/min per μIU/mL; * p < 0.05 vs. 0 week). In MET, SI was also significantly increased at 12 weeks; however, the difference did not remain statistically significant at 24 weeks (0 week; 27.76 ± 3.75, 12 weeks; 36.50 ± 4.27*, 24 weeks; 36.29 ± 7.14 × 10−5dL/kg/min per μIU/mL; * p < 0.05 vs. 0 week) (Figure 1A). Changes in SI were numerically but not statistically greater in IME compared with MET (Figure S1A). Next, insulin secretion was evaluated using oral C-peptide minimal model; φb remained unchanged both in IME and MET (Figure 1B). φd numerically but not statistically increased both in IME and MET but the change ratio was more prominent in IME (Figure 1C). φs and φ were significantly increased in IME (φs, 0 week; 13.97 ± 2.80, 12 weeks; 18.55 ± 2.69*, 24 weeks; 22.63 ± 3.26* × 10−9 min−1, φ, 0 week; 12.96 ± 2.77, 12 weeks; 17.77 ± 2.67*, 24 weeks; 21.51 ± 3.22* × 10−9 min−1; * p < 0.05 vs. 0 week), whereas in MET, the changes in these parameters were minimal, with a small change observed at 24 weeks (φs, 0 week; 16.61 ± 3.58, 12 weeks; 19.31 ± 2.33, 24 weeks; 22.29 ± 4.12* × 10−9 min−1, φ, 0 week; 15.52 ± 3.49, 12 weeks; 18.06 ± 2.36, 24 weeks; 20.99 ± 3.99 × 10−9 min−1; * p < 0.05 vs. 0 week) (Figure 1D,E). Changes in φ were numerically but not statistically greater in IME compared with MET (Figure S1B). To elucidate the effects of the two drugs on insulin sensitivity and secretion, univariate analysis was conducted to examine associations between changes in SI or φ during 24 weeks and various clinical parameters. There was no correlation between change in SI and clinical parameters in IME and MET (Table 1A). Interestingly, change in φ was negatively correlated with baseline φ in MET, whereas no such correlation was observed in IME (Table 1B). In this study, using the oral minimal model, we observed that imeglimin, but not metformin, improved both insulin secretion and insulin sensitivity. Furthermore, imeglimin predominantly enhanced insulin secretion in response to glucose loading rather than at the basal state. Previous studies assessing the effect of imeglimin on insulin resistance using HOMA-IR and Matsuda index have failed to demonstrate significant improvements. Similarly, in our study, HOMA-IR and the Matsuda index showed trends toward improvement, consistent with the changes observed in SI, although they did not reach statical significance (Table S4). Considering the mechanism of action that imeglimin enhances insulin secretion, such effects may be difficult to detect with these parameters due to limitations inherent in its calculation formula that it includes insulin level in the numerator. When assessing insulin resistance in drugs that enhance insulin secretion, methods such as the hyperinsulinemic-euglycemic clamp or oral glucose minimal model used in this study, may be appropriate. SI and φb, φd, φs, and φ obtained in this study were largely consistent with previously reported values in Asian patients with type 2 diabetes [13, 14]. A previous report showed that 1.8 mg of liraglutide increased SI and φ by approximately 60% and 42%, respectively [9]. In this study, imeglimin improved SI and φ to a similar extent. Although direct comparisons across studies are not possible, these findings suggest that imeglimin provides a clinically meaningful improvement in insulin sensitivity and secretion. Based on previous studies, imeglimin enhances insulin secretion by improving mitochondrial function, and our previous study demonstrated that enhanced incretin secretion also contributes to this effect. Consistent with these findings, φs, and φ were significantly increased in IME; however, φd showed only a non-significant increase. φd may be less sensitive for capturing incretin-induced augmentation of the dynamic phase of insulin secretion with a limited sample size, as suggested in the previous report [15]. Imeglimin improved both insulin secretion and insulin resistance, regardless of patients' background. In contrast, although these findings should be considered hypothesis-generating because of the limited sample size, the negative correlation between the change in φ and baseline φ in MET suggests metformin improved insulin sensitivity while optimising overall insulin secretion by suppressing excessive secretion in individuals with high baseline levels and enhancing insulin secretion in those with low baseline levels (Figure S2). As a result, φ increased numerically but did not reach statistical significance overall. This finding may be due to the study population of Japanese participants with relatively low insulin secretory capacity. There are some limitations of this study. First, this study was designed as two parallel trials in which participants were randomised to ensure comparable baseline characteristics, and the sample size was primarily set to assess the effects of each drug separately rather than to directly compare the effects of imeglimin and metformin. Although our findings indicate that each drug was associated with distinct changes in insulin secretion, the limited sample size may have reduced the power to detect differences in direct comparison between IME and MET. The second is multiple testing of parameters. Regarding insulin secretion, we evaluated multiple values, each of which reflects different components of insulin secretion. Because this study was exploratory, adjustments for multiple comparisons were not applied to avoid excessively increasing the risk of false negatives. Nevertheless, in IME, the increases in φs and φ were robustly statistically significant even considering the possibility of false positives. Third, we set the dose of metformin to 1000 mg/day, the most commonly used dose in Japan in order to reflect the clinical effect of metformin. However, this dose is lower than the maximum clinical dose and may underestimate the full potential effect of metformin, even though the reduction in HbA1c was comparable between the imeglimin and metformin groups in this study. Moreover, unlike imeglimin, metformin may affect glucose absorption by promoting glucose excretion into the intestinal tract [16], which could potentially influence the calculation of the sensitivity index, although its impact is considered to be limited. Forth, this study is open-label and lacks a placebo group. Despite limitations, our current analysis shows that imeglimin improves both insulin secretion and sensitivity suggesting the potentially different profile between the two drugs. R.U. and Y.H. contributed to study design and analysis, collection and interpretation of data and writing of the manuscript. Y.O. contributed to analysis, collection and critical revisions of the manuscript for important intellectual content. H.T. and T.H. analysed the data and model assessment. H.K., K.S., K.M., Y.Y., and Y.S. contributed to analysis and critical revisions of the manuscript for important intellectual content. All authors approved the version to be published. The authors are deeply grateful to all past and present members of Center for Diabetes, Endocrinology and Metabolism who contributed to the work. The authors also thank the patients and colleagues who contributed to this study. This study was conducted with the research grants received from Sumitomo Pharma Co. Y.H. received grants from Sumitomo Pharma Co. and Nippon Boehringer Ingelheim Co. Ltd.; and honoraria for lectures from Novo Nordisk Pharma Ltd., Eli Lilly Japan K.K., Daiichi Sankyo Company, Mitsubishi Tanabe Pharma Corporation, Taisho Pharmaceutical Co., MSD K.K. and Sumitomo Pharma Co. H.K. received grants from Ono Pharmaceutical Co. Ltd., Taisho Pharmaceutical Co. and Novo Nordisk Pharma Ltd.; and honoraria for lectures from Sanofi K.K. and Taisho Pharmaceutical Co. Y.Y. has received honoraria for lectures from Ono Pharmaceutical Co. Ltd., Sumitomo Pharma Co., Novo Nordisk Pharma Ltd. and Teijin Pharma Limited. Y.S. has received grants from Nippon Boehringer Ingelheim Co. Ltd., ARKRAY Marketing Inc., Taisho Pharmaceutical Co., Novo Nordisk Pharma Ltd., Terumo Corporation and Sumitomo Pharma Co.; and honoraria for lectures from Taisho Pharmaceutical Co., Nippon Becton Dickinson Company Ltd., Novo Nordisk Pharma Ltd., Eli Lilly Japan K.K., Sumitomo Pharma Co., Ono Pharmaceutical Co. Ltd., and Kyowa Kirin Co. Ltd. The other authors declare no conflicts of interest. The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request. The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer-review/10.1111/dom.70693. Data S1: dom70693-sup-0001-Supinfo01.docx. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.