Beyond metformin: selecting a second therapy

Timothy M. E. DAVIS
University of Western Australia
School of Medicine and
Fremantle Hospital
Fremantle, AUSTRALIA

Beyond metformin:
selecting a second therapy


by T. M. E. Davis, Australia

Metformin is widely accepted first-line therapy for type 2 diabetes but glycemic progression often means that addition of a second drug becomes necessary to maintain adequate blood glucose control. The choice of second-line agents comprises oral therapies (sulfonylureas, thiazolidinediones, alpha-glucosidase inhibitors, dipeptidyl-peptidase 4 (DPP-4) inhibitors, or sodium glucose cotransporter 2 (SGLT2) inhibitors) or injectable therapies (glucagon-like peptide 1 (GLP-1) agonists or insulin). In systematic reviews and meta-analyses of available data, there is little glycemic difference between candidate oral therapies, which means that class- and/or drug-specific tolerability and potential other beneficial and adverse effects will be the main drivers of utilization in this situation. Injectable therapies appear more likely to achieve glycemic targets than oral medications, with GLP-1 agonists having advantages over insulin in producing less hypoglycemia and resulting in weight loss rather than weight gain. Other considerations in selecting a second-line agent are patient preference, cost (the requirement for phase 4 cardiovascular safety trials can add significantly to this in the case of new therapies), and adherence (which is likely to be better with once daily or once weekly rather than multiple times each day). More efficacy, tolerability, and adverse event data are being generated as the results of phase 4 cardiovascular safety trials for DPP-4 inhibitors, SGLT2 inhibitors, and GLP-1 agonists continue to be reported, and further comparative efficacy trials of candidate second-line therapies are in progress. Indeed, the recent positive cardiovascular outcomes in the first SGLT2 inhibitor phase 4 trial to have been completed may have an important effect on the choice of a suitable partner drug for metformin.

Medicographia. 2016;38:37-44 (see French abstract on page 44)

Metformin is conventionally recommended as first-line pharmacotherapy for type 2 diabetes.1 This is mainly because it is effective in improving glycemia without associated weight gain or hypoglycemia, properties that have been evident from a wide range of clinical studies and extensive use in usual care over many decades. There is some evidence, largely from the landmark UKPDS study (UK Prospective Diabetes Study),2 that metformin may be cardioprotective independent of its blood glucose–lowering effects, but this potential benefit has been debated in the light of more recent data.3 Apart from gastrointestinal symptoms, which can be reduced through use of extended-release formulations,4 it is well tolerated. Although not universally accepted,5 its most serious adverse effect is lactic acidosis, which is rare and which usually occurs in the presence of other risk factors such as renal failure. Indeed, since metformin is excreted unchanged by the kidneys, the dose should be reduced in patients with renal impairment, although there is no universally accepted threshold below which it should be avoided.6

Table I
Table. Characteristics of blood glucose–lowering therapies used second-line after metformin as assessed in the American Diabetes
Association and European Association for the Study of Diabetes position statement.

Abbreviations: DPP-4, dipeptidyl-peptidase 4; GLP-1, glucagon-like peptide 1; SGLT2, sodium glucose cotransporter.
After data from reference 1: Inzucchi et al. Diabetes Care. 2012;35(6):1364-1379 and reference 7: Inzucchi et al. Diabetes Care. 2015;38(1):140-149.

In patients who cannot tolerate metformin, in whom it is contraindicated or not recommended, or in those who do not achieve glycemic targets despite appropriate doses, a second oral agent is indicated. The choice of blood glucose– lowering therapy in this situation has broadened in recent years with the emergence of the incretin-based therapies (comprising inhibitors of dipeptidyl-peptidase 4 [DPP-4], and the glucagon-like peptide 1 [GLP-1] agonists) as well as inhibitors of the sodium glucose cotransporter 2 (SGLT2). These relatively new therapies can be considered as second-line agents alongside the more established drugs, which include sulfonylureas, thiazolidinediones, alpha-glucosidase inhibitors, and insulin. The choice of drug in this situation will depend mainly on relative efficacy, adverse effects, andcost (see Table).7 In addition, potential benefits beyond blood glucose–lowering should be considered. The merits of each of these second- line therapies will be considered in this review.


The drugs in this class (glibenclamide, gimepiride, glipizide, and gliclazide), as well as the shorter-acting meglitinides (repaglinide and nateglinide), bind to the sulfonylurea receptor on the membrane of the pancreatic β cells to stimulate endogenous insulin production, a mechanism which is distinct from— and thus complementary to—the blood glucose–lowering effects of metformin.

Glycemic efficacy
The sulfonylurea drugs have similar blood glucose–lowering effects to metformin and reduce glycated hemoglobin (HbA1c) from an average baseline of around 8.0% (64 mmol/mol) by a mean or median of 0.8%-1.0% (9-11 mmol/mol) when added to metformin in dual combination therapy.8,9 However, their glycemic efficacy depends on the time after initiation in monotherapy studies, with a relatively potent effect in the first 3 to 6 months which wanes progressively over the next 1 to 2 years.10 Given that type 2 diabetes is characterized by continuing loss of β-cell function, this time-dependent therapeutic failure may be more pronounced when sulfonylureas are combined with metformin since insulin secretory capacity is likely to be less in a patient who has progressed to two blood glucose–lowering therapies rather than one. This has yet to be demonstrated in comparative studies.

The limited durability of sulfonylureas has been attributed to β-cell toxicity and accelerated apoptosis in short-term animal and in vitro studies,11 but rapid weight reduction through bariatric surgery12 or very low energy diets13 in patients with type 2 diabetes on treatments including sulfonylurea-met- formin can resurrect β-cell function. This suggests that sulfonylureas are not directly toxic to βcells when given in conventional pharmacological doses. Sulfonylureas increase post-prandial glucagon secretion,14 and this may underlie or contribute to secondary failure as enhanced β-cell function starts to predominate over progressive β-cell loss.

Adverse effects
The sulfonylureas are generally well tolerated. However, consistent with their glucose-independent β-cell stimulatory effects, they are the most likely oral blood glucose–lowering agent to cause hypoglycemia. This includes at least a fourfold increased risk when combined with metformin.8,9 There are, however, within-class differences in hypoglycemia risk. The longer-acting sulfonylureas with active renally excreted metabolites such as glibenclamide and glimepiride are more likely to cause hypoglycemia, while gliclazide is the least likely.15 A pertinent example is provided by one study in which the glycemic efficacy of glimepiride and gliclazide were compared and in which the majority of patients were taking metformin monotherapy at baseline.16 The risk of hypoglycemia was significantly lower in gliclazide-treated patients to that seen in the glimepiride group for a similar reduction in HbA1c. Weight gain with insulin secretagogues is moderate and, in the case of the addition of a sulfonylurea to metformin therapy, typically 1 to 3 kg.8,9

Chronic complications
There are few micro- or macrovascular outcome data relating to the relative effect of combining metformin with a sulfonylurea. Perhaps one of the most controversial relevant studies was the UKPDS study, in which diabetes-related deaths were almost doubled in overweight patients taking metformin and sulfonylurea therapy compared with those on a sulfonylurea alone.2 This was attributed to the play of chance since the mortality in the sulfonylurea monotherapy group was unusually low. In addition, post-study monitoring showed that this finding became nonsignificant with accrual of a greater number of events in contrast to the continued significant benefits for intensively treated UKPDS patients as a whole.17 No increased cardiovascular disease (CVD) or mortality risk for sulfonylurea-metformin–treated patients has been found in other studies including observational data from representative community-based samples.18 In pooled analyses, sulfonylurea therapy with or without metformin does not appear to increase the risk of CVD events compared with a variety of comparator regimens, while gliclazide appears to have the safest CVD profile within the class.19


The thiazolidinediones or glitazones (pioglitazone and rosiglitazone) act to improve tissue insulin sensitivity through activation of the peroxisome proliferator–activated receptor gamma nuclear receptor and thus also have complementary blood glucose–lowering effects to those of metformin.

Glycemic efficacy
Compared with metformin monotherapy, the addition of a glitazone reduces the HbA1c from an average baseline of around 8.0% (64 mmol/mol) by a mean or median of 0.7%-0.8% (8-9 mmol/mol).8,9 The glycemic durability of the glitazones is significantly greater than both sulfonylurea and metformin in monotherapy studies.10 Although it is likely that a metforminglitazone combination will retain its efficacy for longer than metformin- sulfonylurea as a result, there are no prospective studies that have examined this hypothesis.20

Adverse effects
The addition of glitazone therapy to metformin does not increase the risk of hypoglycemia.8,9 This reflects the insulin sensitizing properties of this class. Weight gain resulting from the addition of a glitazone to background metformin is a mean or median of 2 to 4 kg in intervention studies,8,9 but real-life experience suggests that this figure may be greater.21 In contrast to sulfonylurea therapy, in which enhanced insulin secretion underlies weight gain, the glitazones stimulate adipogenesis but they also redistribute fat away from visceral depots.22 This could theoretically contribute to a reduction in CVD risk.

Chronic complications
As with sulfonylureas, there are few outcome data relating to the relative effect of combining metformin with a glitazone.8,9 The RECORD trial (Rosiglitazone Evaluated for Cardiovascular Outcomes in oRal agent combination therapy for type 2 Diabetes) compared rosiglitazone-metformin with sulfonylurea- metformin therapy and did not find that the two regimens differed in the number of major adverse cardiovascular events (MACE) outcomes (nonfatal myocardial infarction, nonfatal stroke, and CVD death),23 a reassuring result given the controversy surrounding rosiglitazone engendered by a meta-analysis of available CVD and mortality data in 2007.24 The PROactive trial (PROspective pioglitAzone Clinical Trial In macroVascular Events) showed that the addition of pioglitazone to usual therapy did not significantly reduce a primary end point that included MACE components,25 but a separate comparative analysis of the 10% of patients on metformin alone at study entry has not been reported.

There are no data assessing whether other recognized adverse effects of glitazones, including fluid retention and fractures,10 are increased or reduced in patients taking pioglitazone or rosiglitazone in combination with metformin. Concerns regarding an association between pioglitazone and bladder cancer initially raised by PROactive25 have been attenuated by recent population-level data.26

Alpha-glucosidase inhibitors

The drugs in this class (acarbose, miglitol, and voglibose) slow carbohydrate absorption through inhibition of alpha glucosidase, and they increase circulating GLP-1 concentrations.27 Their main effect is on postprandial glycemia.

Glycemic efficacy
Compared with metformin monotherapy, the addition of an alpha-glucosidase inhibitor reduces the HbA1c by a median of 0.7% (8 mmol/mol) from a meta-analysis of available data.9 A recent study has shown that acarbose significantly attenuates the bioavailability of metformin when they are coadministered,28 but the authors concluded that the accumulated evidence of the glycemic efficacy of this combination implied that this interaction was not clinically significant.

Adverse effects
The addition of an alpha-glucosidase inhibitor to metformin does not increase the risk of hypoglycemia or alter body weight.9 Although both drugs have the potential to cause gastrointestinal side effects such as nausea, vomiting, diarrhea, and abdominal pain, their combination does not appear to increase the incidence of these symptoms to a greater degree than monotherapy with either drug.28,29 Nevertheless, at least some of these studies were carried out in samples of Asian patients and there is evidence that Asian racial groups tolerate alpha-glucosidase inhibitors better than White Caucasians.30 In the UKPDS, for example, only 39% of patients remained on acarbose therapy after 3 years versus 58% on placebo.31

Chronic complications
There are no randomized trials that have examined whether the addition of an alpha-glucosidase inhibitor to metformin therapy influences the risk of chronic complications of diabetes.32 However, in the light of suggestions that acarbose is associated with CVD benefit in observational studies and in meta-analyses of data from intervention studies in which CVD events were not the primary outcome, a large-scale placebo controlled trial of acarbose in Asian patients with CVD and prediabetes has been started,33 but combination with metformin will only occur when the latter drug is prescribed in cases of progression to type 2 diabetes.

DPP-4 inhibitors

The drugs in this class (sitagliptin, vildagltipin, saxagliptin, linagliptin, and alogliptin) inhibit the DPP-4 enzyme that mediates physiological degradation of the incretin hormones GLP-1 and glucose-dependent insulinotropic polypeptide (GIP), thus promoting insulin secretion and inhibiting inappropriate glucagon secretion. There is evidence that the effect of a DPP-4 inhibitor given with metformin on circulating active GLP-1 concentrations is additive,34 supporting their therapeutic coadministration.

Glycemic efficacy
The DPP-4 inhibitors reduce HbA1c from an average baseline of around 8.0% (64 mmol/mol) by a mean or median of 0.7%-0.8% (8-9 mmol/mol) when added to metformin in dual combination therapy.8,9 An analysis of pooled data suggests that there are no differences in glycemic efficacy between drugs in this class when combined with metformin, apart from evidence that patients on alogliptin plus metformin achieve an HbA1c target <7.0% (53 mmol/mol) more frequently than those treated with saxagliptin plus metformin.35

Adverse effects
The DPP-4 inhibitors are well tolerated, including when given with metformin, with neutral effects on hypoglycemia risk and body weight.8,9 In the cardiovascular safety trials involving alogliptin,36 saxagliptin,37 and sitagliptin,38 in which at least two-thirds of patients were taking background metformin therapy, there were nonsignificant excesses of cases of acute pancreatitis in each active arm, which supports the warning of this potential adverse event that is already included in the product information of all the DPP-4 inhibitors. There was, however, no evidence of an increased risk of pancreatic cancer, 36-38 albeit from a follow-up period (up to 3 years) that is too short to rule out carcinogenicity.

Chronic complications
The primary end point in the three DPP-4 inhibitor cardiovascular safety trials that have been reported to date was MACE with or without unstable angina.36-38 The cumulative frequencies of these events in the active versus placebo arms in these trials were effectively co-linear, which—given how commonly metformin was also prescribed—suggests that DPP-4 inhibitor-metformin combination therapy has a neutral short-term effect on CVD. There are, however, apparent between- drug differences in other complications, with saxagliptin37 and perhaps alogliptin,39 but not sitagliptin,38 associated with an increased risk of heart failure. On the other hand, saxagliptin appears to be beneficial for albuminuria37 but there was no such effect in the case of sitagliptin.38

GLP-1 agonists

These drugs (including exenatide, lixisenatide, exenatide extended release, liraglutide, and semaglutide) bind to theGLP-1 receptor and, as with DPP-4 inhibitors, promote insulin secretion and inhibit inappropriate glucagon secretion. However, they also suppress appetite and, especially in the case of the shorter-acting drugs in the class (exenatide and lixisenatide), delay gastric emptying.

Glycemic efficacy
The GLP-1 agonists reduce HbA1c from an average baseline of around 8.0% (64 mmol/mol) by a median of 0.7%-0.8% (8- 9 mmol/mol) when added to metformin in dual combination therapy.9 However, there may be within-class differences. In a comparison of exenatide and liraglutide on a background of metformin with or without sulfonylurea therapy in most cases,40 patients treated with the longer-acting liraglutide had a 0.3% (3 mmol/mol) lower mean HbA1c. In addition, GLP-1 agonists such as exenatide and lixisenatide, which share only approximately 50% homology with human GLP-1, are potentially antigenic; this may attenuate their glycemic efficacy over time.41

Adverse effects
The GLP-1 agonists are relatively commonly associated with nausea and vomiting (up to 50% in some trials42), especially with the shorter-acting agents,43 but these symptoms can resolve with continued use. There are no published formal analyses of whether their use with metformin increases the risk of gastrointestinal side effects but, despite the fact that they are injected therapies, overall treatment satisfaction appears at least as good with this combination as with GLP-1 agonist-sulfonylurea treatment.42 This latter finding could reflect the beneficial psychological effects of weight loss (a median of around 2 kg in pooled analyses of GLP-1 agonist studies involving a majority of patients on background metformin), without an increase in hypoglycemia.9 As with DPP-4 inhibitors, there is evidence from case reports and intervention trials published to date of an excess of pancreatitis, which has been a low frequency adverse event.42,43 Whether their combination with metformin influences this risk is unknown.

Chronic complications
The results of only one CVD safety trial, that involving lixisenatide,44 have been reported and there was a neutral effect on MACE outcomes in patients who were mostly treated with background metformin alone or in combination with other oral agents (American Diabetes Association Scientific Sessions, Boston, June 2015). There are nonglycemic factors, which suggest that the GLP-1 agonists might prevent (through lower blood pressure and more advantageous serum lipid profiles) or contribute to (through increased resting pulse rate) CVD,42,43 but whether these are influenced by metformin coadministration is unknown.

SGLT2 inhibitors

The drugs in this class (dapagliflozin, canagiflozin, and empagliflozin) increase urinary glucose excretion through inhibiting its reabsorption from the proximal renal tubule, a mode of action that is independent of that of metformin.

Glycemic efficacy
The SGLT2 inhibitors reduce HbA1c from an average baseline of around 8.0% (64 mmol/mol) by a mean of 0.5%-0.9% (6-10 mmol/mol) when added to metformin in dual combination therapy.45 There is evidence from a study in healthy volunteers that canagliflozin may be the most active agent in the class because, in addition to comparable SGLT2 inhibitory activity, it has greater activity against SGLT1 in the gastrointestinal tract.46 Whether coadministered metformin influences this apparent difference in the context of type 2 diabetes is unknown.

Adverse effects
The SGLT2 inhibitors are associated with a higher incidence of genitourinary infections and more symptoms related to mild dehydration than comparator blood glucose–lowering therapies regardless of whether they are used alone or in combination with other blood glucose–lowering therapies.45 They do not increase the risk of hypoglycemia, and cause 2 to 4 kg of weight loss, mainly during the first 3 to 6 months of use, including when partnered with metformin.45 A recent development has been case reports of ketoacidosis, which could reflect the proactive reduction in insulin doses, increased glucagon secretion, and reduced renal clearance of ketone bodies associated with SGLT2 inhibitor treatment,47 but metformin does not appear to be influential.48

Chronic complications
There are randomized, placebo-controlled phase 4 CVD safety trials in progress in the case of each of the agents in this class and that involving empagliflozin has recently been completed.49 There were significant 14%, 38%, and 36% reductions in MACE, cardiovascular death, and hospitalization for heart failure, respectively, in patients allocated to empagliflozin. Almost three-quarters of patients were taking metformin at baseline but this was monotherapy in a relatively small proportion. Nevertheless, the overall results appear consistent with statistical modeling suggesting that the overall favorable effects of SGLT2 inhibitors on risk factors such as hypertension should have benefits for the chronic complications of diabetes managed conventionally with metformin as first-line oral therapy.50


Insulin therapy is acknowledged as a valid second-line therapy after metformin, especially in patients with relatively poor glycemic control.1

Glycemic efficacy
Basal and biphasic insulin therapy reduce HbA1c in metformin treated patients by medians of 0.8% (9 mmol/mol) and 1.0% (11 mmol/mol),9 but whether there is a clear difference in glycemic efficacy between these two insulin regimens is unclear.8

Adverse effects
As with insulin in any therapeutic context, the main limitation is hypoglycemia, which is increased by medians of 5.2 and 11.0 times that associated with metformin alone with basal and biphasic insulin, respectively,9 the greatest increases of any second-line therapy. Weight gain is also a concern, with median increases of 1.6 and 3.0 kg, respectively.9

Chronic complications
There are no published prospective outcome studies specifically assessing the effects of the addition of insulin to metformin on chronic complications of diabetes.


The choice of second-line therapy after metformin in type 2 diabetes should be informed by an objective evaluation of efficacy and adverse effects, attributes that will always need to be assessed together with patient preferences and the economic limitations of the health care system. Based on systematic reviews and meta-analyses,8,9 there is not much glycemic difference between candidate oral therapies, which means that tolerability, and potential other beneficial or adverse effects, will be the main drivers of utilization in this situation. Considering pooled analyses8,9 and structured comparative studies involving a range of expected comparator drugs (one example being liraglutide51), injectable therapies appear more likely to achieve glycemic targets than oral medications, with GLP-1 agonists having advantages over insulin in relation to hypoglycemia and body weight effects. Newer agents are expensive, however, especially since large-scale CVD safety trials are now done almost routinely as part of the development of blood glucose– lowering therapies. In addition, and in light of the rosiglitazone experience,24 sprescribers may adopt a conservative approach and wait for sufficient reassuring phase 4 adverse effect data to justify their use as part of usual care. The recent publication of positive cardiovascular findings for empagliflozin49 are likely to shift the choice of second-line therapy towards SGLT2 inhibitors. However, further trials of agents in this class and of the more potent GLP-1 agonists are in progress or completing, and their results may also be influential in modifying the therapeutic algorithm.

The main aspects of a position statement published by members of the American Diabetes Association (ADA) and European Association for the Study of Diabetes (EASD) that relate to second-line therapy are shown in the table.1,7 The conclusions that DPP-4 inhibitors, SGLT2 inhibitors, and alpha-glucosidase inhibitors have inferior relative glycemic efficacy when used in combination with metformin do not appear consistent with the available data.8,9,45 Nevertheless, the suggested framework for assessing treatment options is valid. Although beyond the scope of the present review, the cost of individual medications should be interpreted against benefits and risks. For example, a cost-effectiveness analysis of metformin suggested that, compared with adding rosiglitazone or a sulfonylurea, sitagliptin is likely to be either cost saving or cost effective.52 This was based on disutilities associated with side effects and the direct costs of diabetes-related complications as well as the costs of the medications themselves.

Adherence to treatment is also an issue that is not captured by intervention studies, which are designed to ensure that the medications are taken exactly as prescribed so that true differences in efficacy and side effects are captured. As pointed out in the ADA/EASD position statement, drugs such as the alpha-glucosidase inhibitors which have to be taken at least twice daily may not be associated with as good compliance as once daily (or even once weekly in the case of exenatide extended release and semaglutide) medications, with adverse effects on glycemic control.53

Although phase 4 CVD safety trial data are valuable for the range of information they provide, their pragmatic design (addition of active therapy or placebo to usual care) makes it difficult to dissect out the benefits and risks of particular treatment combinations, including with metformin alone as baseline therapy. Although the SGLT2 inhibitor class is not represented, the GRADE study (Glycemia Reduction Approaches in Diabetes: a comparative Effectiveness study)54 will hopefully provide much-needed comparative data on the long-term glycemic effectiveness of other second-line therapies to inform clinical use. GRADE may also add to the adverse effect databases of the individual therapies.

1. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes: a patient-centered approach: position statement of the American Diabetes Association (ADA) and the European Association for the Study of Diabetes (EASD). Diabetes Care. 2012;35(6):1364-1379.
2. UK Prospective Diabetes Study (UKPDS) Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998;352(9131):854-865.
3. Boussageon R, Supper I, Bejan-Angoulvant T, et al. Reappraisal of metformin efficacy in the treatment of type 2 diabetes: a meta-analysis of randomised controlled trials. PLoS Med. 2012;9(4):e1001204.
4. Schwartz SL, Wu JF, Berner B. Metformin extended release for the treatment of type 2 diabetes mellitus. Expert Opin Pharmacother. 2006;7(6):803-809.
5. Lu WR, Defilippi J, Braun A. Unleash metformin: reconsideration of the contraindication in patients with renal impairment. Ann Pharmacother. 2013;47(11):1488- 1497.
6. Kajbaf F, Arnouts P, de Broe M, Lalau JD. Metformin therapy and kidney disease: a review of guidelines and proposals for metformin withdrawal around the world. Pharmacoepidemiol Drug Saf. 2013;22(10):1027-1035.
7. Inzucchi SE, Bergenstal RM, Buse JB, et al. Management of hyperglycemia in type 2 diabetes, 2015: a patient-centered approach: update to a position statement of the American Diabetes Association and the European Association for the Study of Diabetes. Diabetes Care. 2015;38(1):140-149.
8. Bennett WL, Maruthur NM, Singh S, et al. Comparative effectiveness and safety of medications for type 2 diabetes: an update including new drugs and 2-drug combinations. Ann Intern Med. 2011;154(9):602-613.
9. McIntosh B, Cameron C, Singh SR, et al. Second-line therapy in patients with type 2 diabetes inadequately controlled with metformin monotherapy: a systematic review and mixed-treatment comparison meta-analysis. Open Med. 2011; 5(1):e35-e48.
10. Kahn SE, Haffner SM, Heise MA, et al. Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. N Engl J Med. 2006;355(23):2427-2443.
11. Maedler K, Carr RD, Bosco D, Zuellig RA, Berney T, Donath MY. Sulfonylurea induced beta-cell apoptosis in cultured human islets. J Clin Endocrinol Metab. 2005;90(1):501-506.
12. Brethauer SA, Aminian A, Romero-Talamas H, et al. Can diabetes be surgically cured? Long-term metabolic effects of bariatric surgery in obese patients with type 2 diabetes mellitus. Ann Surg. 2013;258(4):628-636;discussion 36-37.
13. Lim EL, Hollingsworth KG, Aribisala BS, Chen MJ, Mathers JC, Taylor R. Reversal of type 2 diabetes: normalisation of beta cell function in association with decreased pancreas and liver triacylglycerol. Diabetologia. 2011;54(10):2506-2514.
14. Ahren B, Foley JE, Ferrannini E, et al. Changes in prandial glucagon levels after a 2-year treatment with vildagliptin or glimepiride in patients with type 2 diabetes inadequately controlled with metformin monotherapy. Diabetes Care. 2010;33(4):730-732.
15. Schopman JE, Simon AC, Hoefnagel SJ, Hoekstra JB, Scholten RJ, Holleman F. The incidence of mild and severe hypoglycaemia in patients with type 2 diabetes mellitus treated with sulfonylureas: a systematic review and meta-analysis. Diabetes Metab Res Rev. 2014;30(1):11-22.
16. Schernthaner G, Grimaldi A, Di Mario U, et al. GUIDE study: double-blind comparison of once-daily gliclazide MR and glimepiride in type 2 diabetic patients. Eur J Clin Invest. 2004;34(8):535-542.
17. Holman RR, Paul SK, Bethel MA, Matthews DR, Neil HA. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359(15):1577- 1589.
18. Sillars B, Davis WA, Hirsch IB, Davis TM. Sulphonylurea-metformin combination therapy, cardiovascular disease and all-cause mortality: the Fremantle Diabetes Study. Diabetes Obes Metab. 2010;12(9):757-765.
19. Monami M, Genovese S, Mannucci E. Cardiovascular safety of sulfonylureas: a meta-analysis of randomized clinical trials. Diabetes Obes Metab. 2013;15(10): 938-953.
20. van Raalte DH, Diamant M. Glucolipotoxicity and beta cells in type 2 diabetes mellitus: target for durable therapy? Diabetes Res Clin Pract. 2011;93(suppl 1): S37-S46.
21. Hussein Z, Wentworth JM, Nankervis AJ, Proietto J, Colman PG. Effectiveness and side effects of thiazolidinediones for type 2 diabetes: real-life experience from a tertiary hospital. Med J Aust. 2004;181(10):536-539.
22. Tchernof A, Despres JP. Pathophysiology of human visceral obesity: an update. Physiol Rev. 2013;93(1):359-404.
23. Mahaffey KW, Hafley G, Dickerson S, et al. Results of a reevaluation of cardiovascular outcomes in the RECORD trial. Am Heart J. 2013;166(2):240-249.e1.
24. Nissen SE, Wolski K. Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med. 2007;356(24):2457-2471.
25. Dormandy JA, Charbonnel B, Eckland DJ, et al. Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet. 2005;366(9493):1279-1289.
26. Levin D, Bell S, Sund R, et al. Pioglitazone and bladder cancer risk: a multipopulation pooled, cumulative exposure analysis. Diabetologia. 2015;58(3):493-504.
27. Seifarth C, Bergmann J, Holst JJ, Ritzel R, Schmiegel W, Nauck MA. Prolonged and enhanced secretion of glucagon-like peptide 1 (7-36 amide) after oral sucrose due to alpha-glucosidase inhibition (acarbose) in Type 2 diabetic patients. Diabet Med. 1998;15(6):485-491.
28. Kim S, Jang IJ, Shin D, et al. Investigation of bioequivalence of a new fixed-dose combination of acarbose and metformin with the corresponding loose combination as well as the drug-drug interaction potential between both drugs in healthy adult male subjects. J Clin Pharm Ther. 2014;39(4):424-431.
29. Joshi SR, Ramachandran A, Chadha M, Chatterjee S, Rathod R, Kalra S. Acarbose plus metformin fixed-dose combination in the management of type 2 diabetes. Expert Opin Pharmacother. 2014;15(11):1611-1620.
30. Lin BJ, Wu HP, Huang HS, et al. Efficacy and tolerability of acarbose in Asian patients with type 2 diabetes inadequately controlled with diet and sulfonylureas. J Diabetes Complications. 2003;17(4):179-185.
31. Holman RR, Cull CA, Turner RC. A randomized double-blind trial of acarbose in type 2 diabetes shows improved glycemic control over 3 years (U.K. Prospective Diabetes Study 44). Diabetes Care. 1999;22(6): 960-964.
32. Home P. Cardiovascular disease and oral agent glucose-lowering therapies in the management of type 2 diabetes. Diabetes Technol Ther. 2012;14(suppl 1): S33-S42.
33. Holman RR, Bethel MA, Chan JC, et al. Rationale for and design of the Acarbose Cardiovascular Evaluation (ACE) trial. Am Heart J. 2014;168(1):23-29.e2.
34. Migoya EM, Bergeron R, Miller JL, et al. Dipeptidyl peptidase-4 inhibitors administered in combination with metformin result in an additive increase in the plasma concentration of active GLP-1. Clin Pharmacol Ther. 2010;88(6):801-808.
35. Craddy P, Palin HJ, Johnson KI. Comparative effectiveness of dipeptidylpeptidase- 4 inhibitors in type 2 diabetes: a systematic review and mixed treatment comparison. Diabetes Ther. 2014;5(1):1-41.
36. White WB, Cannon CP, Heller SR, et al. Alogliptin after acute coronary syndrome in patients with type 2 diabetes. N Engl J Med. 2013;369(14):1327-1335.
37. Scirica BM, Bhatt DL, Braunwald E, et al. Saxagliptin and cardiovascular outcomes in patients with type 2 diabetes mellitus. N Engl J Med. 2013;369(14): 1317-1326.
38. Green JB, Bethel MA, Armstrong PW, et al. Effect of Sitagliptin on Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med. 2015;373(6):586.
39. Standl E, Erbach M, Schnell O. Dipeptidyl-peptidase-4 Inhibitors and Heart Failure: Class Effect, Substance-Specific Effect, or Chance Effect? Curr Treat Options Cardiovasc Med. 2014;16(12):353.
40. Buse JB, Rosenstock J, Sesti G, et al. Liraglutide once a day versus exenatide twice a day for type 2 diabetes: a 26-week randomised, parallel-group, multinational, open-label trial (LEAD-6). Lancet. 2009;374(9683):39-47.
41. Buse JB, Garber A, Rosenstock J, et al. Liraglutide treatment is associated with a low frequency and magnitude of antibody formation with no apparent impact on glycemic response or increased frequency of adverse events: results from the Liraglutide Effect and Action in Diabetes (LEAD) trials. J Clin Endocrinol Metab. 2011;96(6):1695-1702.
42. Perry CM. Liraglutide: a review of its use in the management of type 2 diabetes mellitus. Drugs. 2011;71(17): 2347-2373.
43. Pinelli NR, Hurren KM. Efficacy and safety of long-acting glucagon-like peptide- 1 receptor agonists compared with exenatide twice daily and sitagliptin in type 2 diabetes mellitus: a systematic review and meta-analysis. Ann Pharmacother. 2011;45(7-8):850-860.
44. Bentley-Lewis R, Aguilar D, Riddle MC, et al. Rationale, design, and baseline characteristics in Evaluation of LIXisenatide in Acute Coronary Syndrome, a longterm cardiovascular end point trial of lixisenatide versus placebo. Am Heart J. 2015;169(5):631-638.e7.
45. Wilding JP. The role of the kidneys in glucose homeostasis in type 2 diabetes: clinical implications and therapeutic significance through sodium glucose cotransporter 2 inhibitors. Metabolism. 2014;63(10):1228-1237.
46. Sha S, Polidori D, Farrell K, et al. Pharmacodynamic differences between canagliflozin and dapagliflozin: results of a randomized, double-blind, crossover study. Diabetes Obes Metab. 2015;17(2):188-197.
47. Taylor SI, Blau JE, Rother KI. Perspective: SGLT2 inhibitors may predispose to ketoacidosis. J Clin Endocrinol Metab. 2015;100(8):2849-2852
48. Peters AL, Buschur EO, Buse JB, Cohan P, Diner JC, Hirsch IB. Euglycemic Diabetic Ketoacidosis: A Potential Complication of Treatment With Sodium- Glucose Cotransporter 2 Inhibition. Diabetes Care. 2015;38(9):1687-1693.
49. Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, cardiovascular outcomes, and mortality in type 2 diabetes. N Engl J Med. 2015; epub ahead of print.
50. Dziuba J, Alperin P, Racketa J, et al. Modeling effects of SGLT-2 inhibitor dapagliflozin treatment versus standard diabetes therapy on cardiovascular and microvascular outcomes. Diabetes Obes Metab. 2014;16(7):628-635.
51. Rigato M, Fadini GP. Comparative effectiveness of liraglutide in the treatment of type 2 diabetes. Diabetes Metab Syndr Obes. 2014;7:107-120.
52. Schwarz B, Gouveia M, Chen J, et al. Cost-effectiveness of sitagliptin-based treatment regimens in European patients with type 2 diabetes and haemoglobin A1c above target on metformin monotherapy. Diabetes Obes Metab. 2008; 10(suppl 1):43-55.
53. Rozenfeld Y, Hunt JS, Plauschinat C, Wong KS. Oral antidiabetic medication adherence and glycemic control in managed care. Am J Manag Care. 2008;14 (2):71-75.
54. Nathan DM, Buse JB, Kahn SE, et al. Rationale and design of the glycemia reduction approaches in diabetes: a comparative effectiveness study (GRADE). Diabetes Care. 2013;36(8):2254-2261.

Keywords: glycemic efficacy; metformin; second-line therapy; tolerability; type 2 diabetes