Lawrence A. Leiter, MD, FRCPC, FACP, FACE, FAHA
Division of Endocrinology & Metabolism,
St. Michael’s Hospital
Professor of Medicine and Nutritional Sciences,
University of Toronto
The Evolving Evidence Base for CV Safety of Antihyperglycemic Agents for Diabetes
Individuals with type 2 diabetes continue to have increased risk of cardiovascular disease (CVD) compared to those without diabetes.1 With extended follow-up, improved glycemic control is associated with reduction in cardiovascular (CV) risk.2 Nonetheless this does not necessarily imply that every antihyperglycemic agent is vasculoprotective.3 This issue was clearly highlighted with the thiazolidinedione agent rosiglitazone, which is an effective antihyperglycemic agent, but was associated with many questions about its effect on risk of myocardial infarction (MI) relative to other antihyperglycemic agents.4
The rosiglitazone controversy prompted the United States Food and Drug Administration (FDA) in 2008 to require that the manufacturers of any new antihyperglycemic agent demonstrate CV safety in the both the pre- and post-market approval settings.5
Several of these post-marketing studies had been completed over the past two years, including those with the dipeptidyl peptidase 4 (DPP-4) inhibitors alogliptin (),6 saxagliptin (),7 and sitagliptin (),8 as well as the glucagon-like peptide-1 (GLP-1) receptor agonist lixisenatide ().9
At the 2015 meeting of the European Association for the Study of Diabetes (EASD), the results of the first CV outcome study of an SGLT-2 inhibitor, using empagliflozin (),10 were reported. In addition, important new data were also presented for and .11-13 This latest information is summarized in this brief report.
EMPA-REG OUTCOME: CV Benefit of Empagliflozin
The Empagliflozin Cardiovascular Outcome Event Trial in Type 2 Diabetes Mellitus Patients (EMPA-REG OUTCOME) included 7,020 patients with type 2 diabetes and known CVD.10 Subjects were randomized to receive empagliflozin 10 mg or 25 mg daily or placebo, in addition to standard of care. As in all of these trials, investigators were encouraged to treat glycemia as per local guidelines with the aim of achieving appropriate A1C targets in all study participants.
[the EMPA-REG OUTCOME] population with established CVD in whom CV risk factors, including blood pressure and dyslipidemia, were well treated with the use of renin-angiotensin-aldosterone system inhibitors, statins, and acetylsalicylic acid. The reductions in the risk of CV death in the empagliflozin group were consistent across subgroups according to baseline characteristics.14
The patients were well balanced in terms of baseline characteristics across the study arms. Mean baseline A1C was approximately 8.1% in each arm and 56%-58% of patients had a disease duration of greater than
10 years. Approximately 74% were receiving metformin, 42%-44% were receiving a sulfonylurea, and 48%-49% were receiving insulin. Mean baseline body mass index (BMI) was 30.6-30.7 kg/m2 and mean blood pressure was 135-136/77 mmHg. A history of MI was present in 46%-47% of patients, while a history of heart failure was reported in 10%-11% of participants at baseline.10
For the primary outcome of the study, the first occurrence of nonfatal MI,
nonfatal stroke, or CV death, empagliflozin demonstrated a statistically significant superiority (HR 0.86, 95% CI 0.74-0.99, p = 0.0382; Figure 1) relative to placebo.10 A pre-specified secondary analysis of a four-point CV composite outcome (hospitalization of unstable angina added to the above components) demonstrated a similar, but not statistically significant, decrease in favour of the empagliflozin arm (HR 0.89, 95% CI 0.78-1.01, p = 0.08).
The reduction in the primary endpoint was driven by a dramatic and highly statistically significant reduction in CV mortality in favour of empagliflozin
(HR 0.62, 95% CI 0.49-0.77, p < 0.0001), with between-group differences in CV death emerging early in the course of the study. This was the only individual component of the three-point composite outcome that was significantly different between the empagliflozin- and placebo-treated participants. Nonfatal MI and nonfatal stroke were not significantly reduced with empagliflozin compared to placebo (Figure 2).10 Overall mortality was also significantly reduced (HR 0.68, 95% CI 0.57-0.82, p < 0.0001; Figure 2),10 with a pattern of early benefit of empagliflozin similar to that seen with CV death. There was no significant difference in non-CV death between the empagliflozin- and placebo-treated participants (HR 0.84, 95% CI 0.60-1.16, p = 0.2852; Figure 2).10 The incidence of hospitalization for heart failure was also found to be significantly lower for empagliflozin-treated patients compared to placebo (HR 0.65, 95% CI 0.50-0.85,
p = 0.0017). Interestingly, very similar benefits were seen in all the endpoints mentioned above with the 10 mg and 25 mg doses.
In terms of safety, as expected, there was a significant increase in risk for genital infections, although the absolute numbers were small. There were no significant differences between empagliflozin and placebo in terms of hypovolemic events, ketoacidosis, or bone fractures, nor were there any significant safety differences noted between the two empagliflozin doses.
[in EMPA-REG OUTCOME] may
have been due to a multiplicity of
factors. It may be a diuretic effect on
top of a small blood pressure reduction, glucose reduction, and weight loss,
or it may be due to an as-yet-unknown mechanism... we really don't know."
There was much discussion as to the possible mechanisms of the observed benefits. The benefits observed in the study are unlikely to be attributable to an impact on atherosclerosis since these occurred so early in the course of the study and were seen in the absence of significant reductions in MI or stroke. Benefits were also not likely mediated by differences in blood glucose levels; the between-group difference was small and no different than that of other trials, and such short-term benefits of glucose lowering have never been observed. It is also unlikely that the benefits were due to a difference in blood pressure control, again given the rapid benefit, the relatively small difference in blood pressure between the active treatments and placebo, and the lack of reduction in stroke observed in this study. The modest weight loss observed in the study in favour of empagliflozin was also most unlikely to be responsible for the treatment effect.
There has been speculation that some of the benefit may be due to a diuretic effect of the SGLT-2 inhibitor. However, since no diuretic has ever shown such a dramatic reduction in mortality and in such a short period of time, this is also unlikely to be the only reason for the dramatic decrease in CV and overall mortality.
Interestingly, analysis of the specific causes of death showed that, although there was some reduction in heart-failure-related mortality and sudden death, the greatest treatment effect was noted in "other" causes of CV mortality.
At this time, the mechanism of benefit remains unknown. The treatment effect may have been due to a multiplicity of factors, including all those mentioned above and/or an as-yet-unspecified, unknown mechanism, such as an anti-arrhythmic effect or a beneficial impact on myocardial fuel utilization and/or contractility.
TECOS: Confirmation of CV Safety of the DPP-4 Inhibitor Sitagliptin
of a signal for heart failure with sitagliptin.
The randomized, double-blind Trial Evaluating Cardiovascular Outcomes with Sitagliptin (TECOS) evaluated the CV safety of sitagliptin compared with placebo, each in addition to usual care.8 TECOS was the largest of the completed DPP-4 CV safety trials in terms of patient-years of follow-up, with 14,671 patients followed for a median of three years.
Subjects in TECOS were aged ≥ 50 years with type 2 diabetes, with an A1C at baseline ≥ 6.5% and ≤ 8.0%. All subjects had established CVD.8 In addition to
their existing antihyperglycemic therapy, subjects were randomized to receive either sitagliptin or placebo. All patients could be treated with open-label antihyperglycemic agents as required, to achieve individually appropriate A1C targets.
Baseline characteristics were well balanced between study arms. Mean duration of diabetes in both groups was 11.6 years, and mean baseline A1C was 7.2% in each arm. At baseline, approximately 80% of each group was taking metformin and approximately 45% were taking a sulfonylurea.
While once again there was a small difference in A1C between the two arms early in the study, with a slight increase in the placebo group and a slight decrease in the sitagliptin group, the overall least-squares mean difference for the entire follow-up was similar to the other studies at 0.29% (p < 0.0001).
For the primary composite endpoint (first confirmed event of CV death, nonfatal MI, nonfatal stroke, or hospitalization for unstable angina), sitagliptin therapy was statistically non-inferior to placebo (HR 0.98, 95% CI 0.88-1.09, p < 0.001).8
Pre-specified intent-to-treat analysis for superiority showed that the difference between the arms was not significant (p = 0.645). Similar observations were made for all CV outcomes studied in TECOS, including the secondary composite endpoint (first confirmed event of CV death, nonfatal MI, or nonfatal stroke; HR for noninferiority 0.99, 95% CI 0.89-1.11, p < 0.001).
Heart failure. Unexpectedly, the Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus—Thrombolysis in Myocardial Infarction 53 (SAVOR-TIMI 53) study demonstrated that saxagliptin was associated with a small but statistically significant increase in risk for hospitalization for heart failure (HR 1.27, 95% CI 1.07-1.51).7 Additionally, the smaller Examination of Cardiovascular Outcomes with Alogliptin vs. Standard of Care (EXAMINE) study showed a non-significant increased risk of hospitalization for heart failure with alogliptin versus placebo (HR 1.19, 95% CI 0.89-1.58).6 In TECOS, however, there was absolutely no difference between sitagliptin and placebo with respect to cumulative rate of hospitalization for heart failure (HR for superiority 1.00, 95% CI 0.83-1.20, p = 0.983; Figure 3).8
Important new subgroup analyses from TECOS presented at EASD 2015 did not show any significant differences between treatment arms, regardless of whether the population was subdivided on the basis of age, sex, BMI, diabetes duration, A1C level, use of insulin, renal function, history of coronary artery disease, or history of heart failure.9
Even among those participants at greatest risk of heart failure, those with a prior history of heart failure, there was absolutely no evidence of an increased
risk for hospitalization for heart failure or other adverse CV outcomes (HR 1.00, 95% CI 0.83-1.20, p = 0.98) (Figure 4).11
TECOS safety update. At EASD 2015, TECOS study investigators also presented an update of selected relevant safety outcomes from the study.12 For acute pancreatitis, as previously reported, there was a small, non-significant imbalance in risk for pancreatitis with more patients receiving sitagliptin developing pancreatitis (23/7,332 patients, 0.3%) compared to the placebo arm (12/ 7,339 patients, 0.2%) (HR 1.93, 95% CI 0.96-3.88, p = 0.065). New analyses of the occurrence of pancreatitis over time did not demonstrate a potential signal of harm, with no increased risk either early or later in therapy.12
The analysis of pancreatic cancer included numerically fewer events in the sitagliptin arm (9/7,332 patients, 0.1%) than in the placebo arm (14/7,339 patients, 0.2%). There was no significant difference between the two arms in this analysis (HR 0.66, 95% CI 0.28-1.51, p = 0.32).
The trends in TECOS for pancreatitis and pancreatic cancer were also observed in SAVOR-TIMI 537 and EXAMINE.6
ELIXA: Update on CV Safety of Lixisenatide
The results of the first completed trial with a GLP-1 receptor agonist, Lixisenatide in Acute Coronary Syndrome (ELIXA) trial, first presented at the American Diabetes Association (ADA) meeting in June 2015,9 had not yet been published at the time of the EASD.
ELIXA included 6,608 patients with type 2 diabetes and a history of acute coronary syndrome (MI or unstable angina). The mean baseline A1C was 7.7%. Patients were randomized to the GLP-1 receptor agonist lixisenatide or placebo added to usual care, with all patients treated at the investigators' discretion to achieve glycemic control.9
For the primary outcome of time to first confirmed event of CV death, MI, stroke,
or hospitalization for unstable angina, lixisenatide was statistically non-inferior to placebo (HR 1.02, 95% CI 0.89-1.17, p < 0.001).9,13 There was also no difference between groups in overall incidence of hospitalization for heart failure (HR 0.96, 95% CI 0.75-1.23) nor, as shown in Figure 5,9,13 for those with or without prior heart failure history.
CV safety trials with other GLP-1 receptor agonists will likely be reporting within the next year. These include the Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results - A Long Term Evaluation (LEADER) study15 with liraglutide and the Trial to Evaluate Cardiovascular and Other Long-term Outcomes With Semaglutide in Subjects With Type 2 Diabetes (SUSTAIN 6) with semaglutide.16
Interpretation and Integration Into Clinical Practice
Given that clinical practice guidelines recommend individualizing antihyperglycemic therapy to match the right agent to the right patient,1 CV safety information for each available agent is critical to help inform the decision-making process. The studies summarized in this report are important pieces of evidence to be added to what we already know about the efficacy and safety of these antihyperglycemic agents.
EMPA-REG OUTCOME presented
at EASD 2015, particularly with regard
to the dramatic reductions in CV
and all-cause mortality, suggest that
an SGLT-2 inhibitor, such as empagliflozin, should be included as
a vasculoprotective agent in antihyperglycemic regimens for patients similar to those included in the study, namely those with type 2 diabetes and established CV disease.
It has now been 20 years since the publication of the hypothesis regarding the potential utility of DPP-4 inhibition in the treatment of type 2 diabetes;17 DPP-4 inhibitors have become a popular choice for oral therapy in Canada since the first approval for clinical use in December of 2007. These agents lower glucose effectively without hypoglycemia or weight gain. They can be combined with virtually all other classes of agents, apart from GLP-1 receptor agonists; moreover, they are administered once daily (at any time of day with or without food) and are generally without any significant drug-drug interactions. The results of the TECOS study with sitagliptin, added to the results of the two previously published large CV safety trials with alogliptin and saxagliptin, confirm the overall CV safety of this class of medication. The new TECOS heart failure data presented at EASD 2015 provide additional reassurance on the lack of a signal for heart failure with sitagliptin.
The positive findings of EMPA-REG OUTCOME presented at EASD 2015, particularly with regard to the dramatic reductions in CV and all-cause mortality, suggest that an SGLT-2 inhibitor, such as empagliflozin, should be included as a vasculoprotective agent in antihyperglycemic regimens for patients similar to those included in the study, namely those with type 2 diabetes and established CV disease.
Why did the SGLT-2 inhibitor empagliflozin demonstrate a vasculoprotective benefit, while CV safety trials completed to date with incretin-based therapies (DPP-4 inhibitors and GLP-1 receptor agonists) show CV safety but no superiority?
one must not neglect the important
role of optimal glycemic control, as appropriate, to reduce the microvascular (retinopathy, nephropathy) complications of diabetes.
Incretin-based therapies have multiple theoretical mechanisms by which they may reduce CV risk.17 While a high-risk population of patients with known CVD may be the right population to assess CV safety, it may be the wrong population to assess CV benefit with these agents, as the benefit of incretin therapies may be greater among patients with shorter duration of diabetes and without known CVD. In contrast, the SGLT-2 inhibitors may have a mechanism or mechanisms that makes their use particularly advantageous in patients with known CVD. This, of course, is totally speculative.
For individuals with diabetes without known CVD, it is premature to extrapolate the EMPA-REG OUTCOME results. We must await the findings of other ongoing studies with SGLT-2 inhibitors that include patients without CVD; these studies are likely several years away from completion. The choice of antihyperglycemic agent(s) should continue to be individualized from amongst the many different antihyperglycemic options and one must not neglect the important role of optimal glycemic control, as appropriate, to reduce the microvascular (retinopathy, nephropathy) complications of diabetes.
1. Canadian Diabetes Association Clinical Practice Guidelines Expert Committee. Canadian Diabetes Association 2013 Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada. Can J Diabetes 2013; 37(suppl 1):S1-S212
2.Holman RR, Paul SK, Bethel MA, et al. 10-year follow-up of intensive glucose control in type 2 diabetes. N Engl J Med 2008; 359(15):1577-89.
3. Hayward RA, Reaven PD, Wiitala WL, et al. Follow-up of glycemic control and cardiovascular outcomes in type 2 diabetes. N Engl J Med 2015; 372(23):2197-206.
4. 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):2457-71.
5. United States Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research (CDER). Guidance for Industry: Diabetes Mellitus — Evaluating Cardiovascular Risk in New Antidiabetic Therapies to Treat Type 2 Diabetes. December 2008. Available at: www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm071627.pdf. Accessed 25 September 2015.
6. 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):1327-35.
7. 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):1317-26.
8. Green JB, Bethel MA, Armstrong PW, et al. Effect of Sitagliptin on Cardiovascular Outcomes in Type 2 Diabetes. N Engl J Med 2015; 373(3):232-42.
9. Pfeffer MA, et al. The Evaluation of Lixisenatide in Acute Coronary Syndrome — The Results of ELIXA. Presented at the American Diabetes Association 75th Scientific Sessions, June 5-9, 2015, Boston, MA. Session 3-CT-SY28.
10. Inzucchi S. EMPA-REG OUTCOME: Cardiovascular Outcomes. Presented at the European Association for the Study of Diabetes 2015 Annual Meeting, September 14-18, 2015, Stockholm, Sweden.
11. Holman RR. TECOS Results: Effect of Sitagliptin on Cardiovascular Outcomes in Type 2 Diabetes. Presented at the European Association for the Study of Diabetes 2015 Annual Meeting, September 14-18, 2015, Stockholm, Sweden.
12. Josse RG. Update on TECOS Safety Data: Hypoglycemia, Pancreatitis & Pancreatic Cancer. Presented at the European Association for the Study of Diabetes 2015 Annual Meeting, September 14-18, 2015, Stockholm, Sweden.
13. Dickstein K. ELIXA: Cardiovascular Perspective. Presented at the European Association for the Study of Diabetes 2015 Annual Meeting, September 14-18, 2015, Stockholm, Sweden.
14. Zinman B, Wanner C, Lachin JM, et al. Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes. N Engl J Med 2015 September 17. [Epub ahead of print].
15. Marso SP, Poulter NR, Nissen SE, et al. Design of the liraglutide effect and action in diabetes: evaluation of cardiovascular outcome results (LEADER) trial. Am Heart J 2013; 166(5):823-30.
16. Trial to Evaluate Cardiovascular and Other Long-term Outcomes With Semaglutide in Subjects With Type 2 Diabetes (SUSTAIN™ 6). ClinicalTrials.gov NCT01720446. Available at: www.clinicaltrials.gov/ct2/show/NCT01720446. Accessed 25 September 2015.
17. Deacon CF, Johnsen AH, Holst JJ. Degradation of glucagon-like peptide-1 by human plasma in vitro yields an N- terminally truncated peptide that is a major endogenous metabolite in vivo. J Clin Endocrinol Metab 1995; 80(3):952-7.
18. Ussher JR, Drucker DJ. Cardiovascular biology of the incretin system. Endocr Rev 2012; 33(2):187-215.
Copyright 2015 STA HealthCare Communications Inc. All rights reserved. This report was made possible through the support of Merck Canada Inc. The author had complete editorial independence in the development of this article and is responsible for its accuracy. The opinions and information contained herein are those of the author and do not necessarily reflect the views or opinions of Merck Canada Inc. or STA HealthCare Communications Inc. Any products mentioned herein should be used in accordance with the prescribing information contained in their respective product monograph.