br Effect of DPP Inhibitors on the Clinical Course
Effect of DPP-4 Inhibitors on the Clinical Course of Heart Failure Four large-scale cardiovascular outcomes trials with DPP-4 inhibitors have been completed, and their findings have been published (Table 1) 31, 32, 33, 34. In 2 studies (with saxagliptin and alogliptin), DPP-4 inhibition was accompanied by an increased risk of hyPerFUsion™ high-fidelity DNA polymerase failure (Table 1), which prompted the U.S. Food and Drug Administration to mandate warnings about this risk in the labeling for both drugs (43). In a trial with saxagliptin (SAVOR-TIMI 53 [Saxagliptin Assessment of Vascular Outcomes Recorded in Patients With Diabetes Mellitus–Thrombolysis In Myocardial Infarction 53]) (44), patients treated with the drug experienced a significant increase in the risk of hospitalization for heart failure (hazard ratio: 1.27; 95% confidence interval: 1.07 to 1.51; p = 0.007), which was seen early following initiation of treatment and was observed primarily in patients with biomarker evidence for elevated cardiac filling pressures at study entry but without symptoms of heart failure. In a trial with alogliptin (EXAMINE [Examination of Cardiovascular Outcomes With Alogliptin Versus Standard of Care in Patients With Type 2 Diabetes Mellitus and Acute Coronary Syndrome]) (45), patients who were randomized to active therapy were hospitalized for heart failure more frequently than those assigned to placebo; this difference was nominally significant in patients without a history of heart failure (hazard ratio: 1.76; 95% confidence interval: 1.07 to 2.90; p = 0.026), who constituted the majority of the patients in the study. In contrast, no increase in the risk of hospitalization for heart failure was reported in trials with sitagliptin (TECOS [Trial Evaluating Cardiovascular Outcomes With Sitagliptin]) and omarigliptin (Table 1) 34, 46. However, the omarigliptin trial was terminated early, and the thoroughness of follow-up is uncertain; furthermore, the number of heart failure events in the trial was too small for reliable interpretation, and the hazard ratio may have been distorted by a higher risk of death in the omarigliptin group (34). What can explain the lack of an increased risk of heart failure hospitalization in the TECOS trial with sitagliptin? Compared with other large-scale trials, the patients in TECOS were more likely to be treated with metformin and less likely to be treated with insulin and thiazolidinediones at the time of randomization; furthermore, during follow-up, treatment with insulin was initiated 30% less frequently in the sitagliptin than placebo group (31). These characteristics differ meaningfully from those in the SAVOR-TIMI 53 trial with saxagliptin, which had a lower prevalence of metformin use and a higher prevalence of the use of insulin and thiazolidinediones at the start of the study; furthermore, the prevalence of insulin use was only modestly different between the 2 groups after randomization, particularly during the first year of the trial (32). In the SAVOR-TIMI 53 trial, the risk of heart failure was directly associated with the use of insulin and inversely associated with the use of metformin (44). These observations are noteworthy, because in large-scale studies of antidiabetic medications other than DPP-4 inhibitors, metformin has been associated with favorable effects on the clinical course of heart failure (47) (potentially due to benefits on autophagy-mediated cardiac remodeling ), whereas both insulin and thiazolidinediones have increased the risk of heart failure (in part related to their antinatriuretic actions) 6, 7, 49, 50. Importantly, concurrent treatment with insulin and thiazolidinediones mutually reinforces the risk of each agent; combined therapy markedly accentuates the likelihood of worsening heart failure in clinical practice 51, 52, possibly because the sodium-retentive actions of thiazolidinediones within the renal tubules are insulin dependent (53). It is therefore likely that concomitant treatment with antidiabetic medications can modulate the pathophysiological mechanisms of DPP-4 inhibitors that lead to heart failure. In the TECOS trial, the high prevalence of metformin use and the low and declining use of insulin in sitagliptin-treated patients may have spared them from experiencing adverse clinical consequences from the pathophysiological actions of DPP-4 inhibitors that can predispose to heart failure.