The authors have declared that no competing interests exist.
Conceived and designed the experiments: MCSW HHXW MWMK DDXZ. Performed the experiments: KQLL SWMC CKMF BCYF STSL. Analyzed the data: MCSW HHXW MWMK DDXZ SMG. Contributed reagents/materials/analysis tools: MCSW. Wrote the paper: MCSW HHXW MWMK DDXZ KQLL SWMC CKMF BCYF STSL SMG.
The prevalence of diabetes mellitus is rising globally, and it induces a substantial public health burden to the healthcare systems. Its optimal control is one of the most significant challenges faced by physicians and policy-makers. Whereas some of the established oral hypoglycaemic drug classes like biguanide, sulphonylureas, thiazolidinediones have been extensively used, the newer agents like dipeptidyl peptidase-4 (DPP-4) inhibitors and the human glucagon-like peptide-1 (GLP-1) analogues have recently emerged as suitable options due to their similar efficacy and favorable side effect profiles. These agents are widely recognized alternatives to the traditional oral hypoglycaemic agents or insulin, especially in conditions where they are contraindicated or unacceptable to patients. Many studies which evaluated their clinical effects, either alone or as add-on agents, were conducted in Western countries. There exist few reviews on their effectiveness in the Asia-Pacific region. The purpose of this systematic review is to address the comparative effectiveness of these new classes of medications as add-on therapies to sulphonylurea drugs among diabetic patients in the Asia-Pacific countries. We conducted a thorough literature search of the MEDLINE and EMBASE from the inception of these databases to August 2013, supplemented by an additional manual search using reference lists from research studies, meta-analyses and review articles as retrieved by the electronic databases. A total of nine randomized controlled trials were identified and described in this article. It was found that DPP-4 inhibitors and GLP-1 analogues were in general effective as add-on therapies to existing sulphonylurea therapies, achieving HbA1c reductions by a magnitude of 0.59–0.90% and 0.77–1.62%, respectively. Few adverse events including hypoglycaemic attacks were reported. Therefore, these two new drug classes represent novel therapies with great potential to be major therapeutic options. Future larger-scale research should be conducted among other Asia-Pacific region to evaluate their efficacy in other ethnic groups.
Diabetes mellitus is the most common non-communicable disease and a worldwide health crisis
Metformin has been used alone as the first line OHA for type 2 diabetes
Combination therapy with either SUs or thiazolidinediones and metformin is a common next step if the glycaemic target is not attained by monotherapy. This results in summation of the therapeutic effects, offering patients with extra benefits
When combination drug therapy fails to achieve therapeutic targets, insulin therapy will be initiated to provide sufficient amount of insulin for maintaining homeostasis of blood glucose
Despite the development of the above therapies for managing diabetes, large proportion of diabetic patients did not achieve their glycaemic targets. In a European cohort of 2,023 type 2 diabetic patients who were on metformin and either SUs or thiazolidinediones, only 25.5% of the patients had adequate glycaemic control. The average HbA1c level after a mean of 2.6 years of combination oral antihyperglycaemic agent therapy was 7.2%
Glucagon-like peptide-1 (GLP-1) analogues or agonists are a group of drugs which mimics the effect of the endogenous hormone GLP-1. They regulate blood glucose level by stimulating insulin secretion, suppressing glucagon secretion
DPP4-I are a class of OHAs that regulates blood glucose level by inhibiting the action of didpeptidyl peptidase 4 (DPP-4). Under normal physiological condition, incretin hormones GLP-1 and gastric inhibitory polypeptide (GIP) are released from the endocrine cells in the small intestine upon food consumption. They stimulate glucose-dependent release of insulin
DPP4-I are overall very well tolerated and have few side effects
In March 2013, the Food and Drug Administration (FDA) announced that the type 2 diabetic patients treated with the drug classes of incretin mimetics and DPP-4 inhibitors may have a higher risk of pancreatitis and pre-cancerous cellular changes
Whilst there is no consistent and concrete evidence against the safety profile of these two new agents, one of the important topics surrounding these novel agents includes their efficacy in achieving optimal glycaemic control among diabetic patients on OHAs, in particular SUs which are used more commonly among patients in the Asia-Pacific region. We therefore searched the existing databases to review the effectiveness of the two drug classes as add-on therapies as compared with SUs or placebo.
A comprehensive search was performed using a combination of medical subject headings (MeSH) and keywords in international databases including MEDLINE and EMBASE in August 2013. Original research studies were searched in the databases from their inception to August 2013 that reported the effectiveness of DPP-4 inhibitors and GLP-1 analogues as add-on therapies among diabetic patients. An additional manual search was conducted using reference lists from research studies, meta-analyses and review articles to identify other potential eligible studies.
(type 2 or type2 or type II or non-insulin dependent or adult onset).tw. (400869) | |
(diabet |
|
1 and 2 (199667) | |
Diabetes Mellitus, Type 2/(204183) | |
3 or 4 (262903) | |
(dpp-4 or dpp 4 or dpp4 or dpp-IV or dpp IV or dppIV or dipeptidyl-peptidase 4 or dipeptidyl peptidase 4).tw. (7463) | |
(incretin |
|
(sitagliptin |
|
6 or 7 or 8 (25825) | |
5 and 9 (11338) | |
(asia |
|
10 and 11 (522) |
We included studies which were designed as randomised controlled trials conducted in the Asia-Pacific region, including (1). participants having type 2 diabetes mellitus with inadequate glycaemic control at enrolment; (2). adults aged ≥18 years and were previously treated with SUs as background diabetic regimen; and that (3). the interventions aimed to assess the drug efficacy of either DPP-4 inhibitors or GLP-1 analogues as add-on anti-diabetic therapy. Only studies which were published as complete full papers in English were included. When the same patient population was reported in two or more journal articles, only the most recent and complete study was included.
All potentially relevant publications retrieved by electronic searching were reviewed independently by two investigators (HHW and MWK) based on examination of study titles and abstracts. Those studies which did not meet the inclusion criteria were excluded. The full-text of the article was obtained if either reviewer believed that the study might be eligible. Any disagreement was resolved by consensus after discussion or, when necessary, by appeal to the principal investigator (MCW).
A research nurse (BCF) and a research assistant (STL) independently extracted relevant information on study participants, design of interventions and controls, study duration and outcome measures from each included study by a standardised extraction form. For this systematic review, the primary outcomes were clinical values including HbA1c and fasting plasma glucose (FPG). The secondary outcomes were drug safety tolerability including the incidence and intensity of AEs, withdrawal due to accident emergency, symptomatic hypoglycaemic events, homeostasis model assessment for insulin resistance (HOMA-R), HOMA-beta cell function, etc. The details of all the included studies were shown in
Studies | Patients | Intervention | Control | Major Outcomes | Ethnicity | Findings |
1). Kaku et al, 2010 | 264 patients on a SU |
Liraglutide 0.6 mg daily or 0.9 mg daily | Placebo with no active ingredients | 1). Change of HbA1c at 24 wks | Japanese | −1.46% to −1.56% (liraglutide)vs. −0.40% (placebo) |
2). Proportion reaching HbA1c<7.0% | 46.5% to 71.3% (liraglutide) vs. 14.8% (placebo) | |||||
2). Seino et al, 2012b | 267 patients aged ≥20 years with suboptimal glycaemic control (HbA1c 7.4% to <10.4%) | Liraglutide 0.6 mg or 0.9 mg daily | Placebo with no active ingredients | Change of HbA1c at 24 wks | Japanese | Mean HbA1c was reduced by 1.00% to 1.27% points than placebo. |
3). Kadowaki et al, 2011 | 179 patients who had suboptimal glycaemic control despite use of SU |
Exenatide 5 µg BID or 10 µg BID offered subcutaneously | Placebo with no active ingredients | Change of HbA1c at 24 wks | Japanese | −1.34% to −1.62% (exenatide) vs. −0.28% (p<0.001) |
4). Inagaki et al, 2012 | 427 patients aged ≥20 years with insufficient glucose control | Exenatide QW 2 mg daily | Once daily insulin glarigine (starting dose 4U) | Change of HbA1c at 26 wks | Japanese | −1.11% (exenatide QW) vs. −0.68% (insulin Glargine), p<0.001 with the 95% C.I. upper limit <predefined non-inferiority margin (0.4%) |
5). Seino et al, 2012a | 311 patients aged 25–81 years with Hba1c between 7–10% and on basal insulin +/− SU |
Lixisenatide starting from 10 µg to 15 µg and 20 µg | Placebo with no active ingredients | Change of HbA1c at 24 wks | Japan, Republic of Korea, Taiwan, the Phillipines | −0.77% (lixisenatide) vs. 0.11% (placebo) (p<0.0001) |
6). Inagaki et al, 2013 | 618 patients aged ≥20 years on SU |
Linagliptin 5 mg daily | Metformin BD or TDS, up to 2,250 mg/day | 1). Change of HbA1c at 52 wks | Japanese | 1). −0.7% to −0.9% (linagliptin) vs. −0.8% to −1.0% (metformin), p = NS |
2).Hypoglycaemic attack rates | 2). 1.6% to 13.7% (linagliptin) vs. 3.2% to 15.9% (metformin); p = NS |
|||||
7). Zeng et al, 2013 | 192 patients on metformin and a SU |
Linagliptin 5 mg daily | Placebo with no active ingredients | 1). Change of HbA1c at 24 wks | Chinese | 1). −0.59% (linagliptin) vs. 0.08% (placebo), p<0.0001 |
2). Change in FPG | 2). −3.9 mg/dL (linagliptin) vs. 15.0% (placebo), p<0.001 | |||||
3). Adverse event rates | 3). 38.9% (linagliptin) vs. 43.8% (placebo), p = NS |
|||||
8). Takihata et al, 2013 | 115 patients inadequately controlled with metformin and/or sulphonylurea | Sitagliptin 50 mg daily | Pioglitazone 15 mg daily | Change of HbA1c at 24 wks | Japanese | −0.86 (SD 0.63%) (sitagliptin) vs. −0.58 (SD 0.68%) (pioglitazone), p = 0.024 |
9). Seino et al, 2012 | 312 Patients on glimepiride (1–4 mg/day) with poor glycaemic control | Alogliptin 12.5 mg or 25 mg | Placebo with no active ingredients | 1). Change of HbA1c at 12 wks | Japanese | 1). −0.59% to −0.65% (alogliptin) vs. 0.35% (placebo) |
2). Adverse events | 2). Comparable event rates with the majority being mild |
SU = sulphonylurea.
A-GI = α-glucosidase inhibitors.
NS = Non-significant.
A total of 522 articles were initially identified to be potentially relevant in MEDLINE and EMBASE. By reviewing the study titles and abstracts, 363 articles were excluded. Full-text articles were retrieved for the remaining 159 articles for formal inclusion and exclusion. Reasons for exclusion for the majority of trials included the duplicate publication of the same study, conference abstracts, and the absence of SUs treated prior to the medication of DPP-4 inhibitors and GLP-1 analogues. Nine studies fulfilled all the inclusion criteria and were finally included in this review. The details of the study selection and the reasons of exclusion were described in
In a double-blind, parallel-group trial by Kaku et al.
Similar findings about GLP-1 analogue liraglutide were reported in one of the trials described in a recent study by Seino et al.
Another GLP-1 agonist, exenatide, was studied and analyzed for its treatment effect and safety in a study involving 179 Japanese patients with type 2 diabetes. The mean changes in HbA1c from baseline to week 24 were −1.34±0.11%, −1.62±0.11%, and −0.28±0.15% for exenatide 5 µg, 10 µg, and placebo, respectively (both p<0.001, exenatide vs. placebo)
Another study by Inagaki et al. confirmed the treatment effect and safety of exenatide with 427 Japanese diabetic patients
A third type of GLP-1 agonist, lixisenatide, was examined in a randomized, double-blind, placebo-controlled trial by Seino et al.
A well-known DDP-4 inhibitor, linagliptin, has been supported to be as effective as metformin in providing add-on therapy for HbA1c reduction. In the 52-week, open-label, multicenter, parallel-group study conducted by Inagaki et al.
The efficacy of linagliptin was further supported by a randomized, placebo-controlled, 24-week trial in China
The COMPASS study by Takihata et al.
Furthermore, Seino et al.
Nine studies from Asia-Pacific countries, involving a total of 2,685 type 2 diabetic patients, were included in this review. Duration of the exposure to the drug treatments varied with studies, ranging from 12 weeks to 52 weeks. Though heterogeneities in treatment protocol and drug dosage existed, results showed that add-on therapy of GLP-1 analogues (liraglutide, exenatide, and lixisenatide) and DPP-4 inhibitors (linagliptin, sitagliptin, and aloglitpin) were effective in improving glycaemic control among type 2 diabetic patients. This conclusion was supported by the reduction in HbA1c levels at the study endpoint, and the greater proportion of patients achieved HbA1c targets after the treatment. These results were consistent with previous Cochrane reviews on GLP-1 analogues and DPP-4 inhibitors
The current study focused on diabetic patients from Asia-Pacific region. This was of particular interest as the risk factors for diabetes differed between ethnic groups. Existing evidence suggested that patients from Asia had a high risk of type 2 diabetes when compared to the Western counterparts at the same BMI. They were more prone to abdominal obesity and low muscle mass with increased insulin resistance
Despite current findings showed improvement in glycaemic control by this drug class, there was inconclusive information regarding its long term cardiovascular benefits. SAVOR
To summarize, the findings from these studies supported the efficacy and acceptable safety profiles of these two newer agents as add-on therapies to SUs and/or other medications among diabetic patients not having optimal glycaemic control. Nevertheless, some of the limitations of these studies should be mentioned. Firstly, almost all studies were conducted among Japanese patients and few representations from other ethnic groups were available in existing literature. This could be explained by the fact that Japanese type 2 diabetic patients are in general less obese with the ‘thrifty’ genotype, which causes more insulin deficiency and less insulin resistance than the Western population
(DOCX)