What this is
- This research evaluates the efficacy and safety of teneligliptin, a DPP-4 inhibitor, for treating type 2 diabetes mellitus (T2DM).
- A Bayesian network meta-analysis was conducted using data from 18 randomized controlled trials (RCTs) involving 3,290 participants.
- The study compares teneligliptin to other antidiabetic medications, including sitagliptin, vildagliptin, metformin, and bromocriptine.
Essence
- Teneligliptin, at doses of 20 mg and 40 mg, effectively reduces HbA1c and fasting plasma glucose in T2DM patients compared to several other treatments, with acceptable safety profiles.
Key takeaways
- Teneligliptin 40 mg showed the highest efficacy in reducing HbA1c, with a mean difference (MD) of -0.84 compared to placebo.
- The proportion of patients achieving HbA1c < 7% was significantly higher with 40 mg of teneligliptin (OR 11.97) compared to placebo.
- Teneligliptin had no significant difference in hypoglycemia or gastrointestinal adverse events compared to placebo, indicating a favorable safety profile.
Caveats
- The analysis included only two RCTs for the 40 mg dose, limiting the robustness of the findings for this higher dosage.
- Heterogeneity was noted in outcomes related to body weight and BMI, which may affect the reliability of those results.
Simplified
Introduction
Diabetes mellitus (DM) is a chronic metabolic disease mainly characterized by hyperglycemia, which seriously endangers human life and health. It is estimated that 537 million adults suffer from DM in the world at present, and this number is projected to increase to 783 million by 2045 (1, 2). Prevalence of DM is high, but its treatment rate and cure rate are low (3). Type 2 diabetes mellitus (T2DM) accounts for nearly 90% of DM in the world (4). The primary pathophysiology of T2DM is characterized by defective insulin secretion and insulin resistance (5). T2DM is mainly caused by a combination of genetic, metabolic, and environmental factors (6, 7). It is associated with increased risk of cardiovascular and renal outcomes under the long-term suboptimal glycemic control (8–10).
In recent years, some new antidiabetic drugs, such as dipeptidyl peptidase-4 inhibitors (DPP-4is), sodium-glucose cotransporter-2 inhibitors (SGLT-2is), and glucagon-like peptide-1 receptor agonists (GLP-1RAs), have been widely used in the treatment of T2DM (11). Of them, DPP-4is universally increase insulin secretion, and decrease levels of intact glucagon in patients with diabetes via potentiation of GLP-1 action (12). DPP-4is are widely welcomed by T2DM patients because of their excellent efficacy and safety.
As an oral DPP-4i launched in recent years, teneligliptin was approved as a treatment option for T2DM patients who have failed to control the blood glucose level under diet and exercise treatment in Japan (2012), Korea (2016), Thailand (2020), and China (2021) (13). It can significantly decrease the glycated hemoglobin A1c (HbA1c) and fasting plasma glucose (FPG) levels, and had a slight influence on body weight (BW). Twenty milligrams once daily is the current recommended dosage regimen for teneligliptin. In Japan, it is also being practiced with close monitoring to increase the dose of teneligliptin to 40 mg per day (14). Therefore, it needs to be further evaluated whether 40 mg of teneligliptin for T2DM patients is effective and safe or not.
It is crucial for rational application of antidiabetic drugs (e.g., teneligliptin) and precise drug treatment of DM by evaluating their efficacy and safety using some scientific methods. Network meta-analysis is a popular method to evaluate multiple treatments or interventions, and has usually been performed by the Bayesian approach (15, 16). Bayesian network meta-analysis is an effective method to simultaneously compare multiple treatments by combining the direct and indirect evidence, and it can provide results of relative rankings of different interventions (17, 18). The aim of this study was to evaluate the efficacy and safety of the DPP-4i teneligliptin in patients with T2DM by Bayesian network meta-analysis. The results of this study will provide important evidence-based basis for rational use of teneligliptin and clinical decision-making of T2DM medication.
Materials and methods
The Bayesian network meta-analysis was in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and its extension for Network Meta-Analysis (19, 20).
Search strategy
The literature search was conducted in PubMed, Web of Science, Embase, Cochrane Central Register of Controlled Trials, and ClinicalTrials.gov↗ from their inception to 22 March 2023, without language restriction. The databases were searched with the following MeSH (Medical Subject Headings) terms or keywords: (1) “Teneligliptin” OR “MP-513”; AND (2) “diabetes mellitus, type 2” OR “diabetes mellitus, type II” OR “noninsulin dependent diabetes” OR “non-insulin dependent diabetes” OR “NIDDM” OR “type II diabetes” OR “type 2 diabetes” OR “T2DM” OR “mature onset diabetes” OR “late onset diabetes” OR “adult onset diabetes”. Furthermore, the reference lists of identified trials were screened to further identify eligible trials.
Study selection
The concrete inclusion criteria were as follows: (1) Patients: any ethnicity, either gender, aged 18 years or older, and HbA1c ≥ 6.5%. (2) Interventions: any dose of teneligliptin used as monotherapy or combination therapy with duration of at least 12 weeks. (3) Comparison: placebo or active comparators with or without background therapy. (4) Outcomes: at least one of the following indicators was reported: HbA1c, the patients achieving HbA1c < 7%, FPG, 2 h postprandial plasma glucose (2h PPG), BW, body mass index (BMI), hypoglycemia, and gastrointestinal adverse events (GIAEs). (5) Study design: randomized controlled trials (RCTs) published without language restrictions.
The studies were excluded if they included patients of age over 75 years, with HbA1c>10%, and with a history of renal, hepatic failures, dyslipidemia, or cardiovascular disorders. Moreover, phase I studies and secondary analyses were excluded.
Data extraction
The following information from the included studies were extracted: study information, baseline characteristics of the patient, intervention measures, and prespecified outcomes. For all the outcomes, we extracted data for the intention-to-treat (ITT) population, which comprised all the randomly assigned patients who received at least one dose of the study medication.
Risk-of-bias assessment
Assessment of the risk of bias was conducted by Cochrane Review Manager (RevMan), which included selection, performance, detection, attrition, and reporting bias (21). The scores for each aspect of eligible studies were recorded as high, low, or unclear risk.
The study search and selection, data extraction, and risk of bias assessment were conducted independently by two reviewers (MZ and RFG). Any differences were resolved through discussion or consultation with a third independent reviewer (GM).
Statistical analysis
A network meta-analysis with Bayesian approach was performed using R version 4.2.3 (http://www.r-project.org/↗) with the packages GEMTC, RJAGS, EXPORT, and BUGSnet and Stata MP 17.0 (StataCrop LLC) in this study (22, 23).
Synthesis of treatments and outcomes
For each prespecified outcome, a network plot of all the interventions was made to identify possible direct and indirect comparisons. The width of the lines in the network plot is proportional to the number of studies, and the node sizes correspond to number of the participants achieving a certain treatment in the comparisons. Effect estimates included odds ratio (OR) for categorical outcomes and mean difference (MD) for continuous outcomes. If the standard deviation (SD) was not reported, it was calculated from the standard error (SE), probability (p) value, confidence interval (CI), or MD according to the guidance from the Cochrane Handbook for Systematic Reviews of Interventions (24).
Model fitting and consistency evaluation
The Markov chain Monte Carlo algorithm was used for each outcome based on 20,000 simulation iterations and 5,000 adaptation iterations. A thinning interval of 1 was applied, which collected one sample every one iteration. The selection between fixed and random model and the evaluation of model fit goodness were realized through deviance information criteria (DIC), and the model with a lower value was chosen. The difference value of DIC between inconsistency and consistency models under 3 indicates a good consistency of network meta-analysis.
Effectiveness evaluation
The difference between the comparisons was considered statistically significant when the 95% CI did not contain 0.00 for continuous outcomes or 1.00 for categorical outcomes. The point estimates with 95% CI for each treatment comparison were presented in a league table. The surface under the cumulative ranking curve (SUCRA), showing each intervention ranking with respective ranking possibility, was calculated to rank the efficacy of each treatment. The SUCRA values ranged from 0% to 100%. The higher value indicates that a particular treatment is more possible to be in a top rank; similarly, the lower value indicates that a particular treatment is more possible to be in a bottom rank (25).
Heterogeneity evaluation and publication bias
I2 statistic was used to assess the heterogeneity arising from differences between the studies within each direct comparison of treatments, and the value of >50% indicated significant heterogeneity between the studies. Publication bias was evaluated using funnel plots.
Results
Study selection and characteristics
The PRISMA flowchart of the included studies is shown in Figure 1. There were 681 publications initially identified in five databases. A total of 18 unique RCTs (26–43) met the inclusion criteria and were considered for the proposed study. The included studies were published from 2013 to 2023, which consisted of 3,290 participants with T2DM in total. The minimum number of participants was 40, and the maximum number was 447 in the 18 included RCTs.
The detailed characteristics of the included studies are provided in Table S1. The weighted means of age, BW, BMI, baseline HbA1c, and baseline FPG were 56.6 years, 73.5 kg, 26.7 kg/m2, 8.0%, and 156.0 mg/dL, respectively. Compared with the others, only one study (33) stood out in terms of BW and BMI (weighted means were 92.8 kg and 32.3 kg/m2, respectively). These differences did not have a significant impact on similarity of baseline characteristics of the included studies. The duration of treatment ranged from 12 to 24 weeks.
Included in the meta-analysis were the following doses: 5 mg, 10 mg, 20 mg, and 40 mg of teneligliptin (qd), 100 mg of sitagliptin (qd), 50 mg of vildagliptin (bid), 500 mg of metformin (qd), and 0.8 mg of bromocriptine (qd). Efficacy outcomes contained mean changes of HbA1c, FPG, 2h PPG, BW, and BMI as well as proportion of the patients achieving HbA1c < 7%. Safety outcomes included incidences of hypoglycemia and GIAEs. Outcomes of HbA1c and FPG were reported in all the 18 included studies. The other six outcomes were only reported in a part of the 18 included studies (Table S1). Network plots of all the efficacy and safety outcomes are illustrated in Figure 2.
PRISMA flowchart of the included studies.
Network plots of efficacy and safety outcomes. These network plots show comparisons of teneligliptin, sitagliptin, vildagliptin, metformin, bromocriptine, and placebo in the RCTs with respect to the sample sizes and number of studies according to eight different outcome measures. The doses of all the antidiabetic drugs are daily dose. Each node represents a certain intervention, and its size represents number of the participants given a certain intervention in the comparisons. The width of the lines represents the number of studies comparing every pair of interventions. HbA1c, glycated hemoglobin A1c; FPG, fasting plasma glucose; 2h PPG, 2 h postprandial plasma glucose; BW, body weight; BMI, body mass index; GIAEs, gastrointestinal adverse events.HbA1c;Patients achieving HbA1c < 7%;FPG;2h PPG;BW;BMI;Hypoglycemia;GIAEs. (A) (B) (C) (D) (E) (F) (G) (H)
Risk-of-bias analysis
The risk of bias in the 18 included RCTs is summarized in Figure 3. Among the 18 RCTs, all of them had low risk for bias of blinding of outcome assessment, incomplete outcome data, and selective reporting, 8 RCTs had low risk for bias in random sequence generation, and the other 10 RCTs had unclear risk of bias, 17 RCTs had low risk for bias in allocation concealment, 15 RCTs had low risk for bias in blinding of participants and personnel, 16 RCTs had low risk for bias in incomplete outcome data, and 1 RCT had unclear risk. Overall, these studies had a low or moderate level of risk.
Assessment of the risk of bias in the included studies.
Network meta-analysis
The statistical analysis of all indicators was performed using a random-effects model. SUCRA values of all the interventions and outcomes are shown in Figure 4 and Table 1. League tables of the efficacy outcomes (HbA1c, proportion of the patients achieving HbA1c < 7%, FPG, 2h PPG, BW, and BMI) are shown in Tables 2–4. League table of the safety outcomes (hypoglycemia, GIAEs) is shown in Table 5.
SUCRA plots of efficacy and safety outcomes. SUCRA, the surface under the cumulative ranking curve value; HbA1c, glycated hemoglobin A1c; FPG, fasting plasma glucose; 2h PPG, 2 h postprandial plasma glucose; BW, body weight; BMI, body mass index; GIAEs, gastrointestinal adverse events.HbA1c;Patients achieving HbA1c < 7%;FPG;2h PPG;BW;BMI;Hypoglycemia;GIAEs. (A) (B) (C) (D) (E) (F) (G) (H)
| Interventions | HbA1c | Patients achieving HbA1c < 7% | FPG | 2h PPG | BW | BMI | Hypoglycemia | GIAEs | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| SUCRA | Rank | SUCRA | Rank | SUCRA | Rank | SUCRA | Rank | SUCRA | Rank | SUCRA | Rank | SUCRA | Rank | SUCRA | Rank | |
| Placebo | 0.57 | 9 | 0.33 | 6 | 15.37 | 8 | 14.2 | 7 | 96.67 | 1 | 53.35 | 2 | 48.22 | 4 | 93.75 | 1 |
| Bromocriptine 0.8 mg | 16.18 | 8 | NA | NA | 0.38 | 9 | 1.93 | 8 | NA | NA | 70.87 | 1 | 80.82 | 1 | 0.15 | 3 |
| Metformin 500 mg | 57.48 | 5 | NA | NA | 57.45 | 4 | 48.07 | 6 | NA | NA | NA | NA | NA | NA | NA | NA |
| Vildagliptin 100 mg | 64.41 | 3 | NA | NA | 43.29 | 7 | 56.11 | 5 | NA | NA | NA | NA | NA | NA | NA | NA |
| Sitagliptin 100 mg | 61.3 | 4 | 77.85 | 2 | 73.5 | 2 | 79.06 | 1 | 18.25 | 3 | 31.48 | 4 | 42.7 | 5 | NA | NA |
| Teneligliptin 5 mg | 33.66 | 7 | 34.33 | 5 | 51.45 | 6 | NA | NA | NA | NA | NA | NA | 51.44 | 3 | NA | NA |
| Teneligliptin 10 mg | 55.21 | 6 | 43.6 | 4 | 54.8 | 5 | 60.25 | 4 | NA | NA | NA | NA | 75.66 | 2 | NA | NA |
| Teneligliptin 20 mg | 75.67 | 2 | 58.07 | 3 | 69.53 | 3 | 61.28 | 3 | 35.07 | 2 | 44.31 | 3 | 32.82 | 6 | 56.11 | 2 |
| Teneligliptin 40 mg | 85.51 | 1 | 85.81 | 1 | 84.24 | 1 | 79.06 | 2 | NA | NA | NA | NA | 18.35 | 7 | NA | NA |
| Placebo | −0.49 (−0.73, −0.26) | −0.66 (−0.86, −0.48) | −0.78 (−0.86, −0.70) | −0.84 (−1.03, −0.65) | −0.69 (−0.98, −0.40) | −0.74 (−1.38, −0.05) | −0.66 (−1.20, −0.12) | −0.27 (−0.53, −0.02) |
| 0.30 (0.07, 0.83) | Teneligliptin5 mg | −0.17 (−0.42, 0.08) | −0.28 (−0.51, −0.05) | −0.35 (−0.60, −0.10) | −0.20 (−0.56, 0.17) | −0.24 (−0.93, 0.48) | −0.17 (−0.75, 0.45) | 0.22 (−0.11, 0.55) |
| 0.22 (0.08, 0.49) | 0.99 (0.21, 2.92) | Teneligliptin10 mg | −0.11 (−0.30, 0.08) | −0.18 (−0.39, 0.04) | −0.03 (−0.36, 0.32) | −0.07 (−0.76, 0.63) | 0.00 (−0.57, 0.60) | 0.39 (0.09, 0.70) |
| 0.16 (0.09, 0.24) | 0.74 (0.18, 2.07) | 0.87 (0.30, 1.93) | Teneligliptin20 mg | −0.06 (−0.25, 0.12) | 0.08 (−0.19, 0.36) | 0.04 (−0.60, 0.72) | 0.12 (−0.42, 0.65) | 0.50 (0.26, 0.74) |
| 0.11 (0.04, 0.24) | 0.48 (0.10, 1.39) | 0.56 (0.17, 1.35) | 0.70 (0.24, 1.58) | Teneligliptin40 mg | 0.15 (−0.19, 0.49) | 0.10 (−0.57, 0.81) | 0.18 (−0.39, 0.76) | 0.56 (0.26, 0.88) |
| 0.13 (0.03, 0.33) | 0.61 (0.08, 2.15) | 0.71 (0.13, 2.20) | 0.81 (0.24, 2.01) | 1.47 (0.27, 4.68) | Sitagliptin100 mg | −0.04 (−0.78, 0.68) | 0.03 (−0.58, 0.63) | 0.42 (0.05, 0.78) |
| NA | NA | NA | NA | NA | NA | Vildagliptin100 mg | 0.08 (−0.80, 1.00) | 0.46 (−0.26, 1.14) |
| NA | NA | NA | NA | NA | NA | NA | Metformin500 mg | 0.38 (−0.20, 0.98) |
| NA | NA | NA | NA | NA | NA | NA | NA | Bromocriptine0.8 mg |
| Placebo | 0.58 (0.22, 0.92) | 0.82 (-0.29, 1.94) | NA |
| −0.11 (−0.94, 0.72) | Teneligliptin 20 mg | 0.25 (-0.82, 1.32) | NA |
| −0.20 (−1.25, 0.85) | −0.09 (−0.74, 0.57) | Sitagliptin 100 mg | NA |
| 0.39 (−1.48, 2.23) | 0.50 (−1.18, 2.16) | 0.59 (−1.21, 2.35) | Bromocriptine 0.8 mg |
| Placebo | 1.60 (0.03, 7.72) | 0.51 (0.01, 2.27) | 1.30 (0.70, 2.19) | 2.63 (0.46, 8.10) | 1.35 (0.30, 3.72) | 3,867,030.67 (0.00, 400,117.91) |
| NA | Teneligliptin5 mg | 2.13 (0.01, 16.00) | 9.39 (0.16, 39.22) | 21.77 (0.24, 75.01) | 9.84 (0.10, 42.35) | 5,377,343.72 (0.00, 947,506.86) |
| NA | NA | Teneligliptin10 mg | 15.44 (0.53, 87.30) | 26.63 (0.86, 149.16) | 16.56 (0.34, 89.05) | 7,695,257.57 (0.00, 1,838,505.19) |
| 0.76 (0.34, 1.28) | NA | NA | Teneligliptin20 mg | 2.13 (0.36, 6.88) | 1.03 (0.28, 2.69) | 2,438,052.60 (0.00, 340,812.34) |
| NA | NA | NA | NA | Teneligliptin40 mg | 0.85 (0.09, 3.32) | 2,299,670.23 (0.00, 224,503.51) |
| NA | NA | NA | NA | NA | Sitagliptin100 mg | 4,025,460.15 (0.00, 433,075.76) |
| 0.01 (0.00, 0.12) | NA | NA | 0.02 (0.00, 0.16) | NA | NA | Bromocriptine0.8 mg |
HbA1c
As shown in Table 2, compared to placebo, 5 mg, 10 mg, 20 mg, and 40 mg of teneligliptin showed significant efficacy in reducing HbA1c (MD [95% CI], −0.49 [−0.73 to −0.26], −0.66 [−0.86 to −0.48], −0.78 [−0.86 to −0.70], and −0.84 [−1.03 to −0.65], respectively). Compared to sitagliptin, vildagliptin, and metformin, 20 mg and 40 mg of teneligliptin showed better efficacy in reducing HbA1c (MD [95% CI], 20 mg: −0.08 [−0.36 to 0.19], −0.04 [−0.72 to 0.60], and −0.12 [−0.65 to 0.42]; 40 mg: −0.15 [−0.49 to 0.19], −0.10 [−0.81 to 0.57], and −0.18 [−0.76 to 0.39], respectively). Compared to sitagliptin, vildagliptin, and metformin, 5 mg and 10 mg of teneligliptin showed weaker efficacy in reducing HbA1c (MD [95% CI], 5 mg: 0.20 [−0.17 to 0.54], 0.24 [−0.48 to 0.93], and 0.17 [−0.45 to 0.75]; 10 mg: 0.03 [−0.32 to 0.36], 0.07 [−0.63 to 0.76], and 0.00 [−0.60 to 0.57], respectively). Compared to bromocriptine, 10 mg, 20 mg, and 40 mg of teneligliptin showed significant efficacy in reducing HbA1c (MD [95% CI], −0.39 [−0.70 to −0.09], −0.50 [−0.74 to −0.26], and −0.56 [−0.88 to −0.26], respectively).
The results of SUCRA indicated that 40 mg of teneligliptin was the best option in reducing HbA1c (85.51%), followed by 20 mg of teneligliptin (75.67%), 100 mg of vildagliptin (64.41%), 100 mg of sitagliptin (61.3%), 500 mg of metformin (57.48%), 10 mg of teneligliptin (55.21%), 5 mg of teneligliptin (33.66%), 0.8 mg of bromocriptine (16.18%), and placebo (0.57%) (Figure 4A and Table 1).
Proportion of the patients achieving HbA1c < 7%
Compared to placebo, 5 mg, 10 mg, 20 mg, and 40 mg of teneligliptin showed significant efficacy in increasing the proportion of the patients achieving HbA1c < 7% (OR [95% CI], 4.95 [1.20 to 14.09], 5.74 [2.03 to 13.22], 6.81 [4.13 to 11.03], and 11.97 [4.19 to 28.48], respectively). Compared to sitagliptin, 40 mg of teneligliptin showed better efficacy in increasing the proportion of the patients achieving HbA1c < 7% (OR [95% CI], 1.47 [0.27 to 4.68]). Compared to sitagliptin, 5 mg, 10 mg, and 20 mg of teneligliptin showed weaker efficacy (OR [95% CI], 0.61 [0.08 to 2.15], 0.71 [0.13 to 2.20], and 0.81 [0.24 to 2.01], respectively) (Table 2).
The results of SUCRA indicated that, superiority of increasing proportion of the patients achieving HbA1c < 7% ranked as follows: 40 mg of teneligliptin (85.81%), 100 mg of sitagliptin (77.85%), 20 mg of teneligliptin (58.07%), 10 mg of teneligliptin (43.6%), 5 mg of teneligliptin (34.33%), and placebo (0.33%) (Figure 4B and Table 1).
FPG
As shown in Table 3, compared to placebo, 5 mg, 10 mg, 20 mg, and 40 mg of teneligliptin showed significant efficacy in reducing FPG (MD [95% CI], −14.87 [−23.34 to −6.30], −15.87 [−21.42 to −9.88], −18.02 [−20.64 to −15.13], and −20.40 [−26.07 to −14.57], respectively). Compared to sitagliptin, 40 mg of teneligliptin showed better efficacy in reducing FPG (MD [95% CI], −1.20 [−13.21 to 10.38]). Compared to sitagliptin, 5 mg, 10 mg, and 20 mg of teneligliptin showed weaker efficacy (MD [95% CI], 4.32 [−9.34 to 17.68], 3.33 [−8.54 to 15.23], and 1.17 [−9.39 to 11.70], respectively). Compared to vildagliptin, 5 mg, 10 mg, 20 mg, and 40 mg of teneligliptin showed better efficacy in reducing FPG (MD [95% CI], −4.90 [−29.21 to 18.41], −5.90 [−29.54 to 17.06], −8.06 [−30.95 to 14.35], and −10.43 [−34.16 to 12.65], respectively). Compared to metformin, 10 mg, 20 mg, and 40 mg of teneligliptin showed better efficacy in reducing FPG (MD [95% CI], −0.60 [−17.68 to 16.16], −2.75 [−18.89 to 13.01], and −5.13 [−22.21 to 11.66], respectively). Compared to metformin, 5 mg of teneligliptin showed weaker efficacy (MD [95% CI], 0.40 [−17.98 to 18.00]). Moreover, compared to bromocriptine, 5 mg, 10 mg, 20 mg, and 40 mg of teneligliptin showed significant efficacy in reducing FPG (MD [95% CI], −31.08 [−45.38 to −17.22], −32.07 [−44.84 to −19.50], −34.23 [−45.93 to −22.96], and −36.61 [−49.33 to −24.01], respectively).
According to the SUCRA (Figure 4C and Table 1), 40 mg of teneligliptin (84.24%) seemed to be the most effective option in reducing FPG, followed by 100 mg of sitagliptin (73.5%), 20 mg of teneligliptin (69.53%), 500 mg of metformin (57.45%), 10 mg of teneligliptin (54.8%), 5 mg of teneligliptin (51.45%), 100 mg of vildagliptin (43.29%), placebo (15.37%), and 0.8 mg of bromocriptine (0.38%).
| Placebo | −14.87 (−23.34, −6.30) | −15.87 (−21.42, −9.88) | −18.02 (−20.64, −15.13) | −20.40 (−26.07, −14.57) | −19.20 (−29.90, −8.12) | −9.97 (−32.71, 13.08) | −15.27 (−31.23, 1.03) | 16.21 (4.77, 28.36) |
| NA | Teneligliptin5 mg | −0.99 (−9.87, 8.05) | −3.15 (−11.65, 5.37) | −5.53 (−14.81, 3.49) | −4.32 (−17.68, 9.34) | 4.90 (−18.41, 29.21) | −0.40 (−18.00, 17.98) | 31.08 (17.22, 45.38) |
| 45.09 (24.57, 64.93) | NA | Teneligliptin10 mg | −2.16 (−7.94, 3.47) | −4.54 (−11.12, 1.87) | −3.33 (−15.23, 8.54) | 5.90 (−17.06, 29.54) | 0.60 (−16.16, 17.68) | 32.07 (19.50, 44.84) |
| 45.61 (36.29, 53.97) | NA | 0.52 (−19.81, 20.53) | Teneligliptin20 mg | −2.38 (−8.15, 3.41) | −1.17 (−11.70, 9.39) | 8.06 (−14.35, 30.95) | 2.75 (−13.01, 18.89) | 34.23 (22.96, 45.93) |
| 53.00 (32.80, 73.09) | NA | 7.91 (−14.68, 30.84) | 7.39 (−12.56, 27.84) | Teneligliptin40 mg | 1.20 (−10.38, 13.21) | 10.43 (−12.65, 34.16) | 5.13 (−11.66, 22.21) | 36.61 (24.01, 49.33) |
| 58.79 (15.27, 104.65) | NA | 13.70 (−33.42, 62.96) | 13.18 (−29.91, 58.23) | 5.79 (−41.47, 55.43) | Sitagliptin100 mg | 9.23 (−16.00, 33.26) | 3.93 (−15.30, 23.23) | 35.40 (20.15, 51.02) |
| 41.41 (5.33, 76.86) | NA | −3.68 (−43.67, 35.64) | −4.20 (−38.77, 30.03) | −11.59 (−52.29, 27.73) | −17.38 (−74.25, 38.24) | Vildagliptin100 mg | −5.30 (−32.85, 22.49) | 26.17 (0.15, 51.36) |
| 34.04 (−12.59, 77.95) | NA | −11.05 (−60.74, 36.33) | −11.57 (−56.89, 31.81) | −18.95 (−68.78, 28.22) | −24.75 (−88.81, 35.62) | −7.37 (−63.51, 48.31) | Metformin500 mg | 31.48 (12.10, 50.80) |
| −15.31 (−41.06, 9.97) | NA | −60.40 (−91.62, −28.94) | −60.92 (−84.64, −36.67) | −68.31 (−99.66, −36.79) | −74.10 (−124.95, −24.55) | −56.72 (−98.61, −14.95) | −49.35 (−97.98, 1.49) | Bromocriptine0.8 mg |
2h PPG
Compared to placebo, 10 mg, 20 mg and 40 mg of teneligliptin showed significant efficacy in reducing 2h PPG (MD [95% CI], −45.09 [−64.93 to −24.57], −45.61 [−53.97 to −36.29], and −53.00 [−73.09 to −32.80], respectively). Compared to sitagliptin, 10 mg, 20 mg, and 40 mg of teneligliptin showed weaker efficacy in reducing 2h PPG (MD [95% CI], 13.70 [−33.42 to 62.96], 13.18 [−29.91 to 58.23], and 5.79 [−41.47 to 55.43], respectively). Compared to vildagliptin, 10 mg, 20 mg, and 40 mg of teneligliptin showed better efficacy in reducing 2h PPG (MD [95% CI], −3.68 [−43.67 to 35.64], −4.20 [−38.77 to 30.03], and −11.59 [−52.29 to 27.73], respectively). Compared to metformin, 10 mg, 20 mg, and 40 mg of teneligliptin also showed better efficacy in reducing 2h PPG (MD [95% CI], −11.05 [−60.74 to 36.33], −11.57 [−56.89 to 31.81], and −18.95 [−68.78 to 28.22], respectively). Compared to bromocriptine, 10 mg, 20 mg, and 40 mg of teneligliptin showed significant efficacy in reducing 2h PPG (MD [95% CI], −60.40 [−91.62 to −28.94], −60.92 [−84.64 to −36.67], and −68.31 [−99.66 to −36.79], respectively) (Table 3).
According to the SUCRA (Figure 4D and Table 1), both 100 mg of sitagliptin and 40 mg of teneligliptin seemed to be the best intervention in reducing 2h PPG (79.06%), followed by 20 mg of teneligliptin (61.28%), 10 mg of teneligliptin (60.25%), 100 mg of vildagliptin (56.11%), 500 mg of metformin (48.07%), placebo (14.2%), and 0.8 mg of bromocriptine (1.93%).
BW
As shown in Table 4, 20 mg of teneligliptin showed better efficacy in reducing BW than sitagliptin (MD [95% CI], −0.25 [−1.32 to 0.82]). However, 20 mg of teneligliptin showed significantly weaker efficacy than placebo (MD [95% CI], 0.58 [0.22 to 0.92]). According to the SUCRA (Figure 4E and Table 1), 20 mg of teneligliptin (35.07%) showed a better effect in reducing BW than sitagliptin (18.25%).
BMI
Compared to placebo, 20 mg of teneligliptin and 100 mg of sitagliptin showed weaker efficacy in reducing BMI (MD [95% CI], 0.11 [−0.72 to 0.94] and 0.20 [−0.85 to 1.25], respectively). Compared to placebo, 0.8 mg of bromocriptine showed better efficacy (MD [95% CI], −0.39 [−2.23 to 1.48]) (Table 4). According to the SUCRA, 0.8 mg of bromocriptine was the best option in reducing BMI (70.87%), followed by placebo (53.35%), 20 mg of teneligliptin (44.31%), and 100 mg of sitagliptin (31.48%) (Figure 4F and Table 1).
Hypoglycemia
As shown in Table 5, compared to placebo, 5 mg, 10 mg, 20 mg, and 40 mg of teneligliptin showed no-inferior risk of hypoglycemia (OR [95% CI], 1.60 [0.03 to 7.72], 0.51 [0.01 to 2.27], 1.30 [0.70 to 2.19], and 2.63 [0.46 to 8.10], respectively). In addition, compared to placebo, sitagliptin and bromocriptine also showed no significant difference in incidence of hypoglycemia. According to the SUCRA (Figure 4G and Table 1), 0.8 mg of bromocriptine (80.82%) was considered as the best intervention in avoiding hypoglycemia, followed by 10 mg of teneligliptin (75.66%), 5 mg of teneligliptin (51.44%), placebo (48.22%), 100 mg of sitagliptin (42.70%), 20 mg of teneligliptin (32.82%), and 40 mg of teneligliptin (18.35%).
GIAEs
Compared to placebo, 20 mg of teneligliptin showed no significant difference in incidence of GIAEs (OR [95% CI], 1.48 [0.78 to 2.98]). Compared to bromocriptine, 20 mg of teneligliptin had a significantly lower risk of GIAEs (OR [95% CI], 0.02 [0.00 to 0.16]) (Table 5). According to the SUCRA (Figure 4H and Table 1), 20 mg of teneligliptin (56.11%) had a lower incidence of GIAEs than 0.8 mg of bromocriptine (0.15%).
Consistency and heterogeneity tests
Difference value of DIC between inconsistency and consistency models for each outcome was less than three, indicating that inconsistency of this network analysis was not significant. For most outcome measures of this study, I2 statistic value was under 50%, and heterogeneity was not obvious. Only I2 values of two outcomes (i.e., BW and BMI) were over 50%. According to the sensitivity analysis (Table S2), the I2 value of network analysis of BW outcome decreased to 42.62% (<50%) after eliminating one (35) of the included studies. However, for network analysis of BMI outcome, the heterogeneity was obvious with a I2 value > 50% after eliminating one of the included studies (36, 38, 43) in turn.
Publication bias
The comparison-adjusted funnel plots for assessment of publication bias are shown in Figure 5. Visual inspections indicated that, distribution of the included studies was not asymmetric, and there was some angle between the adjusted auxiliary line and the horizontal zero line, suggesting that some publication bias may exist.
Comparison-adjusted funnel plot for outcomes. The funnel plots displayed publication bias of respective outcomes:HbA1c,Patients achieving HbA1c < 7%,FPG,2h PPG,BW,BMI,Hypoglycemia,GIAEs. The different colored nodes in the plots represent certain paired comparisons of respective interventions:Placebo,5 mg of teneligliptin,10 mg of teneligliptin,20 mg of teneligliptin,40 mg of teneligliptin,100 mg of sitagliptin,100 mg of vildagliptin,500mg of metformin,0.8 mg of bromocriptine. HbA1c, glycated hemoglobin A1c; FPG, fasting plasma glucose; 2h PPG, 2 h postprandial plasma glucose; BW, body weight; BMI, body mass index; GIAEs, gastrointestinal adverse events. (A) (B) (C) (D) (E) (F) (G) (H) (A) (B) (C) (D) (E) (F) (G) (H) (I)
Discussion
Compared to placebo, 5 mg, 10 mg, 20 mg, and 40 mg of teneligliptin were better in most efficacy outcomes except for reducing BW and BMI. This indicated that teneligliptin showed observable glucose-lowering and poor weight-loss effect. In all safety outcomes, there was no significant difference among placebo and the four doses of teneligliptin. This implied that teneligliptin showed acceptable safety. It should be noted that the heterogeneity in the network meta-analyses of BW and BMI were obvious, so their reliability was lower than the other outcome measures.
Efficacy of teneligliptin increased with its dose from 5 mg to 40 mg, but the risk of hypoglycemia also increased. In particular, 10 mg of teneligliptin showed the lowest risk of hypoglycemia among the four doses. Compared to 20 mg of teneligliptin, 40 mg of teneligliptin showed superior glucose-lowering efficacy and no-inferior safety. Therefore, it is a favorable option to increase the dose of teneligliptin from 20 mg to 40 mg per day when its antidiabetic effect is not satisfactory. It should be noted that the limitation of this study was that 40 mg of teneligliptin was only presented in two RCTs with 169 patients in this meta-analysis, and more clinical trials with more patients should be conducted to further confirm this result in the future.
Compared to the single dose of sitagliptin and vildagliptin, four doses of teneligliptin showed a different antidiabetic effect. Among these included DPP-4is, their antidiabetic effects ranked as follows: 40 mg of teneligliptin, 20 mg of teneligliptin, 100 mg of sitagliptin, 100 mg of vildagliptin, 10 mg of teneligliptin, and 5 mg of teneligliptin. It should be noted that a lower dose of teneligliptin showed better antidiabetic effect than sitagliptin and vildagliptin. Additionally, there was no significant difference among the four doses of teneligliptin and sitagliptin in all safety outcomes. The results can provide reference for the rational selection among these DPP-4is.
Compared to 500 mg of metformin, 20 mg and 40 mg of teneligliptin showed better antidiabetic effect, but 5 mg and 10 mg of teneligliptin showed poorer efficacy. Metformin (500–2,500 mg) is the first-line medication for treatment of T2DM (44). If the therapeutic effect of metformin is not ideal (HbA1c ≥ 7%), teneligliptin combined with metformin can achieve synergistic effect. As a note, the number of participants treated with metformin was limited (n = 35) in the included RCTs, which may lead to deviations for this study.
Bromocriptine was approved for the treatment of T2DM as an adjunct to diet and exercise to improve glycemic control by FDA in 2009 (45). It does not belong to popular antidiabetic drugs. Compared to four doses of teneligliptin, bromocriptine was poorer in most efficacy outcomes and better in reducing BMI, and showed a lower risk of hypoglycemia and a higher risk of GIAEs. As a note, limitation of the number of participants treated with bromocriptine (n = 25) may lead to deviation of the results.
We conducted a literature search and review about the meta-analyses containing teneligliptin, as summarized in Table S3. In previous studies (46, 47), efficacy and safety of teneligliptin were evaluated by very limited systematic reviews and traditional meta-analyses, which only included 10 (47) or 13 (46) RCTs. In the present study, 18 RCTs enrolling 3,290 patients with different interventions were included. There were also two network meta-analyses (48, 49) involving teneligliptin. However, teneligliptin was not the main evaluation object in the two network meta-analyses, and the outcome measure of evaluation of teneligliptin was very limited, which only included HbA1c (48) or incidence of GIAEs (49). Teneligliptin was the main evaluation object in the present Bayesian network meta-analysis, and eight outcome measures were included to systematically evaluate the efficacy and safety of teneligliptin.
In the present study, a serious search comprehensively covered the latest research findings, and independent study identification, selection, and data extraction were performed by two reviewers. Nevertheless, the heterogeneity and publication bias of included studies were not avoided. As a note, our inference was based on currently available data from a limited number of RCTs, and more large-scale, high-quality, and long-term clinical trials are needed to assess efficacy and safety of teneligliptin in the future.
Conclusion
In summary, efficacy and safety of teneligliptin in patients with T2DM were evaluated by Bayesian network meta-analysis of 18 RCTs in this study. Compared to sitagliptin, vildagliptin, metformin, bromocriptine, and placebo, teneligliptin displayed favorable efficacy and acceptable safety in the treatment of T2DM. Twenty milligrams or 40 mg per day could be chosen as the optimal dosage regimen for teneligliptin.
Data availability statement
The original contributions presented in the study are included in the article/. Further inquiries can be directed to the corresponding author. 1
Author contributions
MZ: Investigation, Methodology, Writing – original draft, Data curation, Validation. RG: Data curation, Investigation, Methodology, Validation, Writing – original draft. GM: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Project administration, Resources, Software, Supervision, Writing – review & editing, Validation.
Acknowledgments
We would like to thank the researchers and study participants for their contributions.
Funding Statement
The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This study was supported by the National Natural Science Foundation of China (Grants 82374133, 82074109, 81873078, and 81374051).
Abbreviations
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fendo.2023.1282584/full#supplementary-material↗
References
Associated Data
Supplementary Materials
Data Availability Statement
The original contributions presented in the study are included in the article/. Further inquiries can be directed to the corresponding author. 1