Seven studies reported a decrease in HbA1C. There were 1376 patients in the sitagliptin group and 1473 patients in the GLP-1 receptor agonist group. GLP-1 receptor agonists led to a greater reduction in HbA1C, and the mean difference was 0.42% (95% CI 0.27–0.56, P < .00001) for sitagliptin vs GLP-1 receptor agonists. There was, however, considerable heterogeneity observed across studies (I2 = 68%) (Table 2, Fig. 2 a). The percentage of patients achieving an HbA1C goal of <7.0% was lower in the sitagliptin group than the GLP-1 receptor agonist group, and the RR was 0.70 (95% CI 0.58 to 0.83, I2 = 80%, P < .00001) (Table 2, Fig. 2 b).
In subgroup analyses for HbA1C reduction, a significant difference was observed in all subgroups. In subgroup analyses for the percentage of patients achieving an HbA1C goal of <7%, a significant difference was observed in studies using exenatide, studies using long-acting GLP-1 receptor agonists, studies excluding the potential confounding factor and studies including the potential confounding factor. However, no significant difference was observed among subgroups using liraglutide and short-acting GLP-1 receptor agonists. The results are shown in Table 3.
All 8 studies reported a reduction in FPG. This result is shown in Table 2 and Fig. 2 c. We conducted a meta-analysis with 1418 participants in the sitagliptin group and 1514 participants in the GLP-1 receptor agonist group. The reduction in FPG was greater for patients in the GLP-1 receptor agonist group than for those in the sitagliptin group (MD = 0.78, 95% CI 0.36 to 1.19, I2 = 86%, P = .0003).
In subgroup analyses for FPG reduction, all subgroups showed a major difference except subgroups that used short-acting GLP-1 receptor agonists and excluded the potential confounding factor. The results are shown in Table 3.
Three trials reported a decrease in PPG. There were 238 patients in the sitagliptin group and 242 patients in the GLP-1 receptor agonist group. Both sitagliptin and GLP-1 receptor agonists significantly reduced PPG from baseline, but GLP-1 receptor agonists resulted in a greater reduction. The mean difference was 2.61 mmol/L (95% CI 1.35–3.87, I2 = 75%, P < .00001) for sitagliptin versus that of GLP-1 receptor agonists (Table 2, Fig. 2 d).
3.3.4 Weight loss
Six studies reported weight loss. There were 1115 participants in the sitagliptin group and 1226 participants in the GLP-1 receptor agonist group in our meta-analysis. GLP-1 receptor agonists were associated with a greater reduction in body weight than sitagliptin. The mean difference was 1.42 kg (95% CI 0.71–2.14, I2 = 85%, P < .00001) (Table 2, Fig. 2 e) for the sitagliptin group versus the GLP-1 receptor agonist group.
The results of subgroup analyses for weight loss are shown in Table 3. A significant difference was observed in all subgroups except the subgroup including the potential confounding factor.
3.3.5 Blood pressure
Five studies reported changes in blood pressure from baseline to the end of the study period. There were 954 participants in the sitagliptin group and 1073 participants in the GLP-1 receptor agonist group. Our results did not show a significant difference in lower blood pressure between sitagliptin and GLP-1 receptor agonists; the mean difference for SBP and DBP was 0.38 mm Hg (95% CI-1.14–1.89, I2 = 50%, P = .63) (Table 2, Fig. 2 f) and −0.30 mm Hg (95% CI -1.00–0.39, I2 = 5%, P = .4) (Table 2, Fig. 2 g).
3.4 Hypoglycemic risk
All 8 studies reported the proportion of patients experiencing hypoglycemia, and the definition of hypoglycemia of the included studies is shown in Table S6, http://links.lww.com/MD/D217. There were 1543 participants in the sitagliptin group and 1666 participants in the GLP-1 receptor agonist group. There was no difference in hypoglycemic risk between the GLP-1 receptor agonist group and the sitagliptin group, and the RR was 1.09 (95% CI 0.50 to 2.35, I2 = 77%, P = .84) (Table 2, Fig. 2 h).
The results of subgroup analyses for hypoglycemic risk are shown in Table 3. No significant difference was observed in any of the subgroups.
3.5 The quality assessment of the included studies
The participants of all 8 trials were randomly allocated, 5 studies adequately described the methods of randomization and others did not mention it.[6,7,20] There were no differences in the baseline characteristics between the sitagliptin group and the GLP-1 receptor agonist group. Studies by Charbonel et al and Gadde et al were not blinded to the participants.[24,25] All 8 studies clearly reported participants withdrawing from the trial and accounted for it. A summary for the risk of bias for the included studies is shown in Figure 3.
As mentioned above, obesity or being overweight contributes to insulin resistance, which makes it more difficult to control blood glucose and promotes complications of diabetes. Furthermore, some T2DM treatments can induce weight gain, which has a negative effect on the management of diabetes and worsens the weight issue already present. Moreover, the prevalence of abdominal overweight and obesity is directly related to increasing incidence of hypertension and dyslipidemia. As a result, weight control is a very important factor in diabetes treatment, and even modest weight loss has a favorable effect on preventing the progression of diabetes. A study evaluating the relationship between weight change and glycemic control indicated that weight loss of ≥3% was associated with improved glycemic control in patients newly treated for T2DM. As a result, a study on obese/overweight patients with T2DM is of great clinical value. Sitagliptin and GLP-1 receptor agonists can reduce glucose without risking hypoglycemia and weight gain compared with other antihyperglycemic agents. Our study found that for obese/overweight patients, sitagliptin exerts a less potent effect on the decrease in HbA1C, FPG, PPG, and body weight than GLP-1 receptor agonists, but there was no significant difference in terms of hypoglycemic risk. Our results were similar to the meta-analysis performed by Wang et al, in which the researchers found that DPP-4 inhibitors were less efficacious at reducing HbA1C, FPG, PPG, and body weight than GLP-1 receptor agonists. Therefore, according to these 2 studies (Wang and ours), whether patients are of normal weight or elevated weight, GLP-1 receptor agonists were more effective than DPP-4 inhibitors. In order to provide a more accurate analysis of these 2 classes of incretin therapy in obese/overweight, we are planning to conduct an RCT or a prospective study to compare other drugs of DPP-4 inhibitors with GLP-1 receptor agonists in the management of T2DM with obesity/overweight.
In subgroup analyses, we found that compared with long-acting GLP-1 receptor agonists, sitagliptin was less effective in FPG reduction and resulted in a lower proportion of participants achieving the HbA1C target (<7.0%), while compared with short-acting GLP-1 receptor agonists, there were no significant differences. A previous review examined the efficacy and safety of long-acting GLP-1 receptor agonists, and their results were similar to our findings. These researchers found that long-acting GLP-1 receptor agonists had a more sustained effect on FPG and greater HbA1C reduction but fewer gastrointestinal side effects than short-acting GLP-1 receptor agonists. At present, to the best of our knowledge, no studies have examined this phenomenon and compared sitagliptin and long- and short-acting GLP-1 receptor agonists. Nevertheless, long-acting GLP-1 receptor agonists might be more beneficial in the management of obese/overweight patients who fail with oral or insulin agents. Therefore, further research is needed to demonstrate the definitive superiority of long-acting GLP-1 receptor agonists over the other 2 classes.
Considerable heterogeneity was observed in our meta-analysis, and many factors, such as different control groups, different exposure durations, and the potential confounding factor, might lead to this heterogeneity. For example, in subgroup analyses by different control groups (exenatide, liraglutide), the values of I2 decreased in outcome of HbA1C reduction, while for FPG reduction, the values of I2 decreased to 0 when we conducted subgroup analysis based on studies using liraglutide. Notably, the exposure duration of these 2 studies was 26 weeks.[25,27] Accordingly, we believe that the exposure duration might also be an important factor for heterogeneity among studies. However, due to the limited data, subgroup analyses by exposure duration could not be conducted. Further analysis comparing different durations is needed.
There were several strengths of our meta-analysis. First, the quality of studies included in our meta-analysis was high. Second, a variety of outcomes were evaluated. Third, sensitivity analyses conducted by reanalyzing the data using a fixed effects model and omitting 1 study at a time demonstrated that our conclusion was robust (Table S7 and S8, http://links.lww.com/MD/D217). Furthermore, our subgroup analyses were sufficient.
Nevertheless, our study has several limitations. First, although we searched widely, there were only 8 studies included. Therefore, the sample size was small. As a result, we were unable to perform subgroup analysis for all outcome measures, and future study with large sample size was needed to confirm our conclusions. Second, different control groups, different exposure durations and the confounding factor of the included studies led to considerable heterogeneity in our meta-analysis. Moreover, because there was a shortage of individual data, obese and overweight patients could not be assessed separately. This was also one of reasons for considerable heterogeneity. Third, the standard deviation in our study was calculated from the sample size and the standard error, and therefore, the calculation error might not be avoided. Finally, the exposure durations of the included studies were 4–28 weeks, which was too short to evaluate endpoint events such as cardiovascular events, all-cause mortality and so on. One study had an exposure duration of only 4 weeks; therefore, our results concerning HbA1C might be somewhat biased.
In conclusion, for obese/overweight patients, sitagliptin might exert a less potent effect regarding HbA1C, FPG, PPG, and weight reduction than GLP-1 receptor agonists; however, there was no difference in hypoglycemic risk. Meanwhile, long-acting GLP-1 receptor agonists seemed more effective in reducing FPG. However, further research with more participants and longer treatment durations is needed to demonstrate the real superiority of GLP-1 receptor agonists, especially long-acting GLP-1 receptor agonists, over sitagliptin in terms of efficacy and safety, which could then help clinicians provide a more favorable therapeutic regimen for obese patients with T2DM in clinical practice.
We thank the authors of the included articles for their thoughtful work. We thank the study team for their cooperation.
Conceptualization: Haiying Jin, Wei Zhang.
Data curation: Danping Dai, Yiyang Mao.
Formal analysis: Danping Dai, Yiyang Mao, Haiying Jin.
Investigation: Danping Dai.
Methodology: Wei Zhang.
Writing – original draft: Danping Dai.
Writing – review & editing: Haiying Jin.
. Derosa G, Carbone A, Franzetti I, et al. Effects of a combination of sitagliptin
plus metformin vs metformin monotherapy on glycemic control, b-cell function and insulin resistance in type 2 diabetic patients. Diabetes Res Clin Pract 2012;98:51–60.
. Wu Y, Ding Y, Tanaka Y, et al. Risk factors contributing to type 2 diabetes and recent advances in the treatment and prevention. Int J Med Sci 2014;11:1185–200.
. Chen L, Magliano DJ, Zimmet PZ. The worldwide epidemiology of type 2 diabetes mellitus—present and future perspectives. Nat Rev Endocrinol 2011;8:228.
. Serván PR. Obesity and Diabetes. Nutr Hosp 2013;28(Supl. 5):138–43.
. Ji M, Xia LB, Cao JZ, et al. Sitagliptin
/metformin versus insulin glargine combined with metformin in obese
subjects with newly diagnosed type 2 diabetes. Medicine (Abingdon) 2016;95:1–7.
. Berg JK, Shenouda SK, Heilmann CR, et al. Effects of exenatide twice daily versus sitagliptin
on 24-h glucose, glucoregulatory and hormonal measures: A randomized, double-blind, crossover study. Diabetes Obes Metab 2011;13:982–9.
. Gaal LV, Souhami E, Zhou TY, et al. Efficacy
and safety of the glucagon-like peptide-1 receptor agonist lixisenatide versus the dipeptidyl peptidase-4 inhibitor sitagliptin
in young (<50 years) obese
patients with type 2 diabetes mellitus. J Clin Transl Endocrinol 2014;1:31–7.
. Umpierrez GE, Blevins T, Rosenstock J, et al. The effects of LY2189265, a long-acting glucagon-like peptide-1 analogue, in a randomized, placebo-controlled, double-blind study of overweight
patients with type 2 diabetes: the EGO study. Diabetes Obes Metab 2011;13:418–25.
. Wang T, Gou Z, Wang F, et al. Comparison of GLP-1 analogues versus sitagliptin
in the management of type 2 diabetes: systematic review and meta-analysis
of head-to-head studies. PLoS One 2014;9:e103798.
. Wirth A. Anti-diabetic drugs. Weight reduction as a favourable side effect. Internist 2011;52:451–61.
. Scheen AJ. Dipeptidylpeptidase-4 (DPP-4) inhibitors are favourable to glucagon-like peptide-1 (GLP-1) receptor agonists: yes. Eur J Intern Med 2012;23:126–31.
. Li S, Li H, Wang R, et al. The effect of sitagliptin
patients with insulin treatment-induced diabetes mellitus. Eur Rev Med Pharmacol Sci 2017;21:3490–5.
. Kodera R, Shikata K, Nakamura A, et al. The glucose-lowering efficacy
japanese patients with type 2 diabetes. Intern Med 2017;56:605–13.
. Deeks J, Higgins J, Altam D. Analysing data and undertaking meta-analyses. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. Available at: http://handbook-5-1.cochrane.org
. Accessed September 2, 2018.
. Sterne J, Egger M, Moher D. Addressing reporting biases. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. Available at: http://handbook-5-1.cochrane.org
. Accessed September 2, 2018.
. Higgins JPT, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. Educ Debate 2003;327:557–60.
. Evangelou E, Ioannidis J, Patsopoulos NA. Uncertainty in heterogeneity estimates in meta-analysis
. Br Med J 2007;335:914–6.
. Bergenstal RM, Wysham C, MacConell L, et al. Efficacy
and safety of exenatide once weekly versus sitagliptin
or pioglitazone as an adjunct to metformin for treatment of type 2 diabetes (DURATION-2): a randomised trial. The Lancet 2010;376:431–9.
. Skrivanek Z, Gaydos B, Chien J, et al. Dose-finding results in an adaptive, seamless, randomized trial of once-weekly dulaglutide combined with metformin in type 2 diabetes patients (AWARD-5). Diabetes Obes Metab 2014;16:748–56.
. Gerlanc NM, Cai J, Tkacz J, et al. The association of weight loss with patient experience and outcomes in a population of patients with type 2 diabetes mellitus prescribed canagliflozin. Diabetes Metab Syndr Obes 2017;10:89–99.
. McAdam-Marx C, Mukherjee J, Bellows BK, et al. Evaluation of the relationship between weight change and glycemic control after initiation of antidiabetic therapy in patients with type 2 diabetes using electronic medical record data. Diabetes Res Clin Pract 2014;402–11.
. Madsbad S, Kielgast U, Asmar M, et al. An overview of once-weekly glucagon-like peptide-1 receptor agonists-available efficacy
and safety data and perspectives for the future. Diabetes Obes Metab 2011;13:394–407.
. Gadde K, Vetter M, Lqbal N, et al. Efficacy
and safety of autoinjected exenatide once-weekly suspension versus sitagliptin
or placebo with metformin in patients with type 2 diabetes: The DURATION-NEO-2 randomized clinical study. Diabetes Obes Metab 2017;19:979–88.
. Charbonnel B, Steinberg H, Eymard E, et al. Efficacy
and safety over 26 weeks of an oral treatment strategy including sitagliptin
compared with an injectable treatment strategy with liraglutide in patients with type 2 diabetes mellitus inadequately controlled on metformin: a randomised clinical trial. Diabetologia 2013;56:1503–11.
. Russell-Jones D, Chan M, Cddihy R, et al. Efficacy
and safety of exenatide once weekly versus metformin,pioglitazone, and sitagliptin
used as monotherapy in drug-naïve patients with type2 diabetes (DURATION-4). Diabetes Care 2012;35:252–8.
. Pratley RE, Nauck M, Bailey T, et al. Liraglutide versus sitagliptin
for patients with type 2 diabetes who did not have adequate glycaemic control with metformin: a 26-week, randomised, parallelgroup, open-label trial. Lancet 2010;375:1447–56.
efficacy; GLP-1 receptor agonists; hypoglycemic risk; meta-analysis; obese; overweight; sitagliptin; T2DM
Supplemental Digital Content
Copyright © 2019 the Author(s). Published by Wolters Kluwer Health, Inc.