In this pooled analysis of 14,611 patients from 25 clinical studies, treatment with sitagliptin was not associated with an increased risk of cardiovascular events relative to the control treatments used in the studies (both active and placebo). The present results extend and confirm the previously reported cardiovascular safety findings with sitagliptin. To control for the confounding effects of pooling active comparators, the present analysis also examined the effects of sitagliptin versus placebo on cardiovascular safety and found no increased risk of cardiovascular events with sitagliptin. Furthermore, cardiovascular-related analyses with other DPP-4 inhibitors support the present findings of no evidence of cardiovascular harm[17–20]. In these other analyses, there also tended to be a numerical reduction in risk of cardiovascular events with DPP-4 inhibitors relative to non-exposure to DPP-4 inhibitors. Recently, when Monami et al. combined the results of randomised clinical trials for DPP-4 inhibitors in a meta-analysis, a significant reduction in risk for MACE-related events was found with DPP-4 inhibitors. These findings need to be confirmed in prospective studies appropriately designed to assess cardiovascular outcomes. For sitagliptin, a placebo-controlled study assessing cardiovascular outcomes is currently underway in patients at increased risk for cardiovascular events (Clinicaltrials.gov: NCT00790205;).
In a subanalysis of three studies comparing treatment with sitagliptin to sulphonylureas in patients with type 2 diabetes mellitus, a lower incidence and risk of MACE-related events was observed with sitagliptin compared to sulphonylurea. In the sensitivity analysis including all blinded exposure to sitagliptin or sulphonylurea and more reported events overall, a similar trend was observed. A lower incidence of cardiovascular events was also reported with linagliptin compared to glimepiride in a two-year trial. The comparison between linagliptin and glimepiride is being further assessed in a long-term cardiovascular outcomes trial (Clinicaltrial.gov: NCT 01243424). These observations could reflect a deleterious effect of sulphonylureas, a beneficial effect of DPP-4 inhibitors, or a combination of both. While DPP-4 inhibitors and sulphonylureas both enhance insulin secretion, there are differences in their mechanisms of action, which have the potential for impacting cardiovascular safety.
Sulphonylureas bind to the SUR subunit (subtype SUR1) of the potassium ATP (KATP) channel in the beta cell membrane and by inhibiting KATP channel activity, increase insulin release irrespective of ambient glucose concentrations. Incretin-mediated enhancement of insulin release, in contrast, is glucose-dependent. As a result of the mechanisms of action, hypoglycaemia is a common side effect with sulphonylurea therapy, whereas rates of hypoglycaemia with incretin-based therapies are not increased relative to placebo. For example, in the three studies used for the present pooled analysis, there was a 3- to 6-fold increase in the incidence of symptomatic hypoglycaemia with sulphonylurea compared with sitagliptin[27–30]. Hypoglycaemia is associated with ischaemic complications in diabetic subjects. Additionally, sulphonylureas bind to the SUR receptor (subtype SUR2) on cardiac myocytes and on endothelial cells, and thus may have direct effects on cardiovascular function. Sulphonylureas have also been shown to abolish the beneficial effects of ischaemic preconditioning, a protective mechanism whereby a brief period of ischaemia is cardioprotective for subsequent periods of prolonged ischaemia[41, 42].
Sulphonylureas are widely used as pharmacologic therapy in patients with type 2 diabetes mellitus. A concern regarding the potential for sulphonylurea-induced cardiovascular toxicity was first raised as a result of the University Group Diabetes Program (UGDP) study, in which an increase in mortality was observed with tolbutamide compared to both insulin and to placebo. While controversy ensued regarding the validity of these findings[44, 45] and their applicability to other drugs in this class, a warning regarding the increased risk of cardiovascular mortality was introduced and remains in the product information for all sulphonylureas marketed in the United States. Long-term prospective clinical studies have not reported an increased risk of cardiovascular events or mortality with sulphonylurea relative to other treatments in patients with recently- or newly-diagnosed type 2 diabetes mellitus[46, 47]. However, many, but not all, observational studies have found an association between sulphonylurea use and an increase in cardiovascular events or mortality[21–25]. Of note, metformin was a main comparator in these observational studies and any differences could be attributed to the cardiovascular benefit of metformin rather than a risk related to sulphonylurea. The combination of metformin and sulphonylurea has been associated with an increased risk of death in a sub-study of UKPDS 34 or the composite endpoint of hospitalization for cardiovascular disease or mortality in a meta-analysis of observational studies. These results have not been confirmed in a randomized, prospective clinical study.
In contrast to the data suggesting a detrimental effect of sulphonylureas on cardiovascular outcomes, preclinical and clinical mechanistic studies have suggested potential benefits of incretins and incretin-based therapies, such as DPP-4 inhibitors or GLP-1 agonists, on cardiovascular function and outcomes[50, 51]. In animals, GLP-1 infusion protected the rat myocardium against ischaemia reperfusion injury[52, 53]. Genetically-modified mice that are lacking the DPP-4 enzyme had improved survival post-myocardial infarction; similar protection was observed in normal mice treated with sitagliptin[54, 55]. In humans, GLP-1 infusion enhanced endothelial function, as measured by forearm blood flow in response to acetylcholine; interestingly, coadministration of the sulphonylurea glyburide, but not glimepiride, abolished GLP-1 induced augmentation of forearm blood flow. GLP-1 infusion also improved left ventricular ejection and contractile function in patients with acute myocardial infarction. In a study of patients with coronary artery disease awaiting revascularisation, sitagliptin treatment was associated with improvement in left ventricular performance in response to dobutamine-induced stress and with mitigation of post-ischaemic stunning.
While the potential benefits of DPP-4 inhibition have been primarily attributed to the enhancement of GLP-1 activity, other endogenous substrates of DPP-4 could be relevant to cardiovascular function and outcomes. Treatment with sitagliptin increased plasma levels of stromal-derived factor-1α (SDF-1α) and circulating endothelial progenitor cells, while also reducing plasma monocyte chemoattractant protein-1, a proinflammatory chemokine, in patients with type 2 diabetes mellitus. SDF-1α, a chemokine, attracts stem cells to ischaemic sites and enhances post-ischaemia angiogenesis. Zaruba et al. found that, in mice in which myocardial infarction was induced via surgical ligation of the left anterior descending artery, coadministration of granulocyte-colony-stimulating factor (GCSF; to stimulate stem cell mobilisation) with a DPP-4 inhibitor resulted in increased myocardial homing of circulating CXCR-4+ stem cells, reduced cardiac remodeling, and improved heart function and survival. A clinical trial assessing the safety and efficacy of sitagliptin in combination with GCSF in patients with acute myocardial infarction is currently underway.
The present results should be interpreted with caution because of the post hoc nature of this analysis and the clinical studies were not specifically designed to assess cardiovascular outcomes. Additionally, the case definition used in the analysis was based on reports of adverse events that matched MACE-related terms using the Medical Dictionary for Regulatory Activities and on cardiovascular-related deaths, rather than on the results of a formal process of adjudication of reported events. Any potential impact of rosiglitazone on the present findings was likely minimal, as rosiglitazone therapy was included in only two of the twenty-five studies. In one study in which rosiglitazone was a comparator agent, there were no MACE-related events reported. In the other study that had three MACE-related events, rosiglitazone and metformin were background therapy in both the sitagliptin and non-exposed groups.
The strengths of this analysis include the pooling of data from randomized controlled trials, the use of patient-level data, and a large sample size. The potential influence of baseline characteristics on outcomes was controlled by using studies in which patients were randomly assigned to treatment groups. Lastly, numerous supporting analyses confirmed the primary findings.
In summary, these analyses suggest that treatment with sitagliptin does not increase cardiovascular risk in patients with type 2 diabetes mellitus. In a subanalysis, a higher rate of cardiovascular morbidity and mortality was associated with sulphonylurea therapy relative to sitagliptin. Whether this observation is related to a deleterious effect of sulphonylurea therapy, a protective effect of sitagliptin, or a combination of the two is unknown at this time and will require future research.