As Metformin Combinations Proliferate, Questions Arise About Value

Evidence-Based Diabetes Management, September 2016, Volume 22, Issue SP13

For some patients newly diagnosed with type 2 diabetes, skipping metformin monotherapy and going straight to fixed-dose combinations with a DPP-4 inhibitor or an SGLT2 inhibitor makes sense, evidence shows.

Earlier this year, when the FDA granted initial approval to Jentadueto XR and a first-line indication to Invokamet, it gave fresh momentum to an entire category of medications: single pills that combine metformin with some other treatment for type 2 diabetes (T2D).

Several recent studies suggest that physicians should employ combination therapy faster and more aggressively than they have in the past. Indeed, a growing body of evidence supports immediate combination therapy for many patients. There are questions, however, about the benefits of physically combining medications. Studies show that fixed-dose combinations do improve patient adherence to treatment regimens. There is less evidence, however, that patients fare better over time with combination pills rather than their component pills.

Although a number of payers declined requests from Evidence-Based Diabetes Management to discuss how much value they see in the physical combination of existing oral medications, their formularies suggest they see some. Most pharmacy benefits managers cover most combinations.

“There’s some value in 1 pill instead of 2. We just don’t know exactly how much,” John Buse, MD, PhD, a professor at the University of North Carolina School of Medicine and the director of the university’s diabetes care center, told Evidence-Based Diabetes Management in an interview.

No new trials were required for the approval of Jentadueto XR. The approval hinged on pre-existing validation of its method of delayed drug release and trial results from the immediate-release version of Jentadueto, which combines the dipeptidyl peptidase-4 (DPP-4) inhibitor linagliptin (Tradjenta) and metformin hydrochloride. Investigators in the immediate-release medication’s pivotal trial1 randomized 791 patients to twice-daily treatment with 6 regimens:

• 2.5-mg linagliptin + 1000 mg metformin

• 2.5-mg linagliptin + 500 mg metformin • 2.5-mg linagliptin monotherapy

• 500-mg metformin monotherapy • 1000-mg metformin monotherapy or placebo

Mean changes in glycated hemoglobin (A1C) from a baseline of 8.7% were:

• 1.7 percentage points (95% CI, —2.0 to –1.4 percentage points) for linagliptin + 1000-mg metformin

• —1.3 percentage points (95% CI, –1.6 to –1.1 percentage points) for linagliptin + 500-mg metformin

• —1.2 percentage points (95% CI, –1.5 to –0.9 percentage points) for 1000-mg metformin monotherapy

• —0.8 percentage points (95% CI, –1.0 to –0.5 percentage points) for 500-mg metformin monotherapy

• —0.6 percentage points (95% CI, –0.9 to –0.3 percentage points) for linagliptin monotherapy (all P<.0001).

Hypoglycemia occurred at a statistically similar rate with combination therapy (1.7%) and metformin monotherapy (2.4%). Adverse event rates were comparable across treatment arms, as was the absence of any significant change in body weight. Invokamet, which combines metformin with the sodium-glucose cotransporter 2 (SGLT2) inhibitor canagliflozin (Invokana), was originally approved as a second-line treatment in 2014. Earlier this year, after reviewing evidence from a new phase 3 trial,1 the FDA approved the drug as an initial treatment for patients with T2D who have A1C levels greater than 7.5%.

Investigators in that trial randomized 1186 untreated patients to 26 weeks of canagliflozin 100 mg + metformin, canagliflozin 300 mg + metformin, canagliflozin 100-mg monotherapy, canagliflozin 300-mg monotherapy, or metformin monotherapy. Mean A1C levels, which started at 8.8% for all groups, fell by 1.77 percentage points in the 100-mg combination group, 1.78 percentage points in the 300-mg combination group, 1.37 percentage points in the 100-mg monotherapy group, 1.42 percentage points in the 300-mg monotherapy group, and 1.3 percentage points in the metformin monotherapy group. Moreover, members of both combination therapy groups lost significantly more weight than metformin group members and were significantly more likely to attain A1C levels below 7%.

These results led the investigators to conclude that patients who resemble the trial population would fare better if they started treatment on Invokamet rather than metformin alone. “Initial therapy with canagliflozin plus metformin was more effective and generally well tolerated versus each monotherapy in drug-naïve type 2 diabetes,” they wrote in Diabetes Care.2

It was not the first time that study authors had suggested that some newly diagnosed patients would be better off skipping metformin monotherapy and moving straight to combination therapy. The trial that won Jentadueto’s approval in 2012 used it as a first-line treatment, and the investigators who ran the trial concluded that the combination performed better than the monotherapy. “Initial combination therapy with linagliptin plus metformin was superior to metformin monotherapy in improving glycemic control, with a similar safety and tolerability profile, no weight gain, and a low risk of hypoglycemia,” they wrote at the time.

Even before that, researchers who had tested several other metformin combinations had reached similar conclusions. A 2006 piece that appeared in Treatments in Endocrinology made a case for first-line combination therapy.3 “The glycosylated hemoglobin goal in patients with type 2 diabetes mellitus should be to achieve as low a value as can be obtained without causing significant or frequent hypoglycemia. This is best achieved by utilizing agents that lower glucose levels without causing hypoglycemia (thiazolidinediones and metformin),” wrote David S.H. Bell, MB, FACP, FACE, the piece’s author. “Utilizing small doses of two drugs will also result in a decreased incidence of adverse effects compared with a large dose of a single drug.”

A number of recent large studies have shown significant advantages to starting newly diagnosed patients on various types of combination therapies. “We decided to test the [American Diabetes Association] algorithm, which was, ‘We put you on metformin. When you fail, and you will, I’ll add [another oral medication]. When you fail, and you will, I’ll add basal insulin.’ We chose [to put it up against] a triple-therapy algorithm, which was a GLP-1 receptor agonist, low-dose pioglitazone,…and metformin,” said Ralph DeFronzo, MD, professor of medicine and chief of the Diabetes Division at the University of Texas Health Science Center in San Antonio, at an MD Magazine Peer Exchange, while describing a trial he’d worked on.“They all stared with A1C of about 8.8% and A1C came down in both groups to about 6.3%. Now, with the ADA algorithm, at the end of 3 years, they’re up to 6.7%. With the triple-therapy algorithm, they’re at 5.9%. The incidence of hypoglycemia was 7.5-fold greater with the ADA algorithm, but our A1C is 5.9% and theirs is 6.7%. We measured beta cell function and, at the end of 3 years, we had perfectly normal beta cell function with the triple-therapy algorithm. With the ADA algorithm, they’d lost about 90% of beta cell function. You also saw, with the triple-therapy algorithm, a 60% increase in insulin sensitivity and no change with the ADA algorithm. Weight gain with the ADA algorithm was 3.7 kg. Weight loss with the triple therapy algorithm, 3 kg. This is total destruction of the ADA algorithm.”

There is, however, far less evidence to show that patients who use early combination therapy to control A1C experience better outcomes than others, according to a 2016 review published in the Journal of Diabetes and Its Complications.4 “Support for improvement in microvascular and macrovascular outcomes with early, aggressive therapy is conflicting. [UK Prospective Diabetes Study] supports the use of intensive therapy to improve long-term microvascular and macrovascular outcomes, although significant improvements in [myocardial infarction] and death from any cause did not materialize until 10 years post trial,” wrote Steve Milligan, MD, the author of the review, citing Holman et al.5

On the other hand, “The ORIGIN trial was the first key evaluation of long-term [cardiovascular] outcomes in patients with T2D. Participants had [impaired fasting glucose], [impaired glucose tolerance], or newly diagnosed T2D, but insulin glargine therapy targeting near-normal [fasting plasma glucose] levels did not significantly reduce the risk of macrovascular outcomes,”6 Milligan added. Still, barring strong evidence to the contrary, professional societies assume that fast-acting, long-lasting blood sugar control will eventually improve outcomes. The newest guidelines from the American Association of Clinical Endocrinologists and the American College of Endocrinology call for metformin monotherapy in patients whose T2D is newly diagnosed and they have A1C less than 6.5%, dual therapy for new patients with A1C ranging between 7.5% and 9.0%, and triple therapy for the newest patients with A1C levels exceeding 9.0%.7 Even the ADA guidelines now say doctors should consider starting new patients with combination therapy, although only with 2 agents and only in patients with A1C greater than 9%. (For other patients, ADA guidelines still recommend at least 3 months of metformin monotherapy before the beginning of combination therapy.8) Guidelines from both organizations, moreover, urge the use of combination therapy whenever monotherapy fails to get patients below A1C targets, because there is no doubt that nearly any type of combination therapy reduces A1C more than monotherapy in most patients.

A 2014 meta-study9 looked at 15 randomized, controlled trials with a total of 6693 subjects whose characteristics varied considerably. Mean baseline age ranged from 48.4 to 62.7 years, mean baseline duration of disease ranged from 1.6 to 4.1 years, and mean baseline A1C ranged from 7.2% to 9.9%. The medication classes paired with metformin also varied considerably and included thiazolidinediones, insulin secretagogues, DPP-4 inhibitors, or SGLT2 inhibitors. The only thing that didn’t vary much was the outcome. Combination therapy reduced A1C more (weighted mean difference, —0.43 percentage points; 95% CI, –0.56 to –0.30 percentage points) and increased the chance that patients would get A1C under 7% (relative risk, 1.40; 95% CI, 1.33-1.48).

The evidence that combination therapy controls blood sugar better than monotherapy is overwhelming. The evidence that physically combining pills produces better outcomes than prescribing patients several pills is less dramatic. There are, however, many studies that find patient adherence increases as the effort required to adhere to treatment regimens decreases. For example, as noted in the cover story about injectable insulin combinations, a 2009 review of 20 studies found a significant negative correlation between doses per day and treatment-regimen adherence in all 20 studies,10 although the studies in question examined the impact of consolidating several oral doses given at various times rather than 2 injections taken right after each other. Efforts to measure the effect of combining multiple pills into single pills (rather than extending the release of a single medication) have also reported improved patient compliance. A 2008 study11 of patients who were prescribed metformin and a sulfonylurea before August 2000 and Glucovance (which combines the 2 medications) after that date used information from claims data to see whether the switch increased the number of days that patients at least had both medications on hand. The study authors found a 12.8% improvement in adherence.

Another pharmacy claims—based study of treatment adherence by patients with diabetes found an even greater effect.12 “Among the 1815 previously treated patients receiving glyburide or metformin monotherapy who required the addition of the alternative agent, resulting in combination therapy, adherence rates were significantly lower (54.0%; 95% CI, 0.52-0.55) than in the 105 patients receiving monotherapy who were switched to fixed-dose combination therapy (77.0%; 95% CI, 0.72-0.82). The 59 previously treated patients receiving combination therapy who were switched to fixed-dose combination therapy had a significant improvement in adherence after the switch (71.0% vs 87.0%; P <.001).”

(Another paper13 notes a potential pitfall with metformin combinations: many such combinations use immediate-release metformin, which requires twice-daily dosing. Replacing 2 pills that need to be taken once daily with 1 pill that needs to be taken twice daily does not reduce pill burden.) Much of the research into the effects of fixed-dose combinations on treatment adherence focuses on cardiovascular disease and hypertension, and it typically reaches similar conclusions. A 2010 meta-analysis of 6 studies of 30,295 patients combined the data to report that fixed-dose combinations were associated with a 29% increase in compliance compared with individual pills.14 A 2011 meta-analysis of 7 other studies that estimated compliance through medication possession found that a combination doubled the risk ratio of patient treatment persistence.15

In addition to the research showing that combination pills increase treatment adherence, there is much research demonstrating that treatment adherence improves long-term outcomes. There is not, however, much direct evidence that the use of combination pills improves long-term outcomes. Indeed, the few studies so far that have examined how the increased adherence associated with combination pill use affects known risk factors have found only small effects, although still significant ones. A 15-month trial that pitted a 4-pill combination against 4 separate pills in 2004 patients found only modest practical improvements.16 “Among patients with or at high risk of cardiovascular disease, use of a fixed-dose-combination strategy for blood pressure, cholesterol, and platelet control versus usual care resulted in significantly improved medication adherence at 15 months (86% vs 65%) and statistically significant, but small improvements, in systolic blood pressure (−2.6 mm Hg; 95% CI, −4.0 to −1.1 mm Hg; P <.001) and low-density lipoprotein cholesterol (−4.2 mg/dL; 95% CI, −6.6 to −1.9 mg/dL; P <.001).”

Without concrete evidence about the effects on long-term outcomes, it is impossible to calculate the cost-effectiveness of combination pills that cost more than the sum of their component parts. In many cases, however, combinations cost less than the sum of their component parts, so they are, by definition, cost-effective.


1. Haak T, Meinicke T, Jones R, Weber S, von Evnatten M, Woerle HJ. Initial combination of linagliptin and metformin improves glycaemic control in type 2 diabetes: a randomized, double- blind, placebo-controlled study. Diabetes Obes Metab. 2012;14(6):565-574. doi: 10.1111/j.1463-1326.2012.01590.

2. Rosenstock J, Chuck L, Gonzalez-Ortiz M, et al. Initial combination therapy with canagliflozin plus metformin versus each component as monotherapy for drug-naïve type 2 diabetes. Diabetes Care. 2016;39(3):353-362. doi: 10.2337/dc15-1736.

3. Bell DS. The case for combination therapy as first-line treatment for the type 2 diabetic patient. Treat Endocrinol. 2006;5(3):131-137.

4. Milligan S. Combination therapy for the improvement of long-term macrovascular and microvascular outcomes in type 2 diabetes: rationale and evidence for early initiation. J Diabetes Complications. 2016;30(6):1177-1185. doi: 10.1016/j.jdiacomp.2016.03.010.

5. Holman RR, Paul SK, Bethel A, Mathews DR, Neil AW. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med. 2008;359(15):1577-1589. doi: 10.1056/NEJMoa0806470.

6. Gerstein HC, Bosch J, Gagenais GR, et al; ORIGIN Trial Investigators. Basal insulin and cardiovascular and other outcomes in dysglycemia. N Engl J Med. 2012;367(4):319-328. doi: 10.1056/ NEJMoa1203858.

7. Garber AJ, Abrahamson MJ, Barzilay JI, et al; American Association of Clinical Endocrinologists (AACE); American College of Endocrinology (ACE). Consensus statement by the American Association of Clinical Endocrinologists and American College of Endocrinology on the Comprehensive type 2 diabetes management algorithm—2016 executive summary. Endocr Pract. 2016;22(1):84-113. doi: 10.4158/EP151126.CS.

8. American Diabetes Association. Approaches to glycemic treatment. Diabetes Care. 2016;39(suppl 1):S52-S59. doi: 10.2337/dc16-S004.

9. Phung OJ, Sobieraj DM, Engel SS, Rajpathak SN. Early combination therapy for the treatment of type 2 diabetes mellitus: systematic review and meta-analysis. Diabetes Obes Metab. 2014;16(5): 410-417. doi: 10.1111/dom.12233.

10. Saini SD, Schoenfield P, Kaulback K, Dubinsky MC. Effect of medication dosing frequency on adherence in chronic diseases. Am J Manag Care. 2009;15(6):e22-e33.

11. Pan G, Chernew ME, Fendrick AM. Impact of fixed-dose combination drugs on adherence to prescription medications. J Gen Intern Med. 2008;23(5):611-615. doi: 10.1007/s11606-008-0544-x.

12. Melikian C, White TJ, Vanderplas A, Dezii CM, Chang E. Adherence to oral antidiabetic therapy in a managed care organization: a comparison of monotherapy, combination therapy and fixed-dose combination therapy. Clin Ther. 2002;24(3):460-467.

13. Blonde L, San Juan ZT, Bolton P. Fixed-dose combination therapy in type 2 diabetes mellitus. Endocrin Pract. 2014;20(12):1322-1332. doi: 10.4158/EP14259.RA.

14. Gupta AK, Arshad S, Poulter NR. Compliance, safety, and effectiveness of fixed-dose combinations of antihypertensive agents: a meta-analysis. Hypertension. 2010;55(2):399-407. doi: 10.1161/HYPERTENSIONAHA.109.139816.

15. Sherrill B, Halpern M, Khan S. Zhang J, Panjabi S. Single-pill vs. free-equivalent combination therapies for hypertension: a meta-analysis of health care costs and adherence. J Clin Hyperten. 2011;13(12):898-909. doi: 10.1111/j.1751-7176.2011.

16. Thom S, Poulter N, Field J, et al; UMPIRE Collaborative Group. Effects of a fixed-dose combination strategy on adherence and risk factors in patients with or at high risk of CVD: the UMPIRE randomized clinical trial. JAMA. 2013;310(9):918-929. doi:10.1001/jama.2013.277064.