Cardiovascular risk with DPP-4 inhibitors: latest evidence and clinical implications

Early systematic reviews or pooled analyses of randomized control trials suggested a neutral effect of sitagliptin [Engel et al. 2013], vildagliptin [Schweizer et al. 2010] and alogliptin [White et al. 2013b] on cardiovascular outcomes. Some meta-analyses even suggested a probable decrease in cardiovascular adverse events associated with saxagliptin [Frederich et al. 2010] and linagliptin [Johansen et al. 2012], or dipeptidyl peptidase-4 (DPP-4) inhibitors as a drug class [Monami et al. 2013]. However, these analyses were based mainly on studies of short duration and did not incorporate data from recently published dedicated cardiovascular safety trials [Karagiannis et al. 2014]. We provide an update on the effect of DPP-4 inhibitors on major adverse cardiovascular events (MACE) and the incidence of heart failure, based on recently published results from long-term clinical trials of sitagliptin [Green et al. 2015], saxagliptin [Scirica et al. 2013] and alogliptin [White et al. 2013a], and relevant updated meta-analyses in light of recent findings supporting a cardiovascular benefit for novel glucose lowering drugs in patients at high cardiovascular risk [Zinman et al. 2015].

Following concerns about the uncertainty regarding the cardiovascular profile of glucose lowering agents [Matthews and Tsapas, 2008], regulatory agencies in both Europe and the United States issued guidance requesting the assessment of cardiovascular safety of new antidiabetic medications during the early stages of their marketing authorization [FDA, 2008; CHMP, 2012]. The SAVOR-TIMI-53 [Scirica et al. 2013] and EXAMINE [White et al. 2013a] trials assessed the cardiovascular safety of saxagliptin and alogliptin, respectively. Both studies suggested a neutral effect on a 3-point MACE composite outcome of cardiovascular death, myocardial infarction or ischemic stroke. Similarly, the TECOS trial assessed the effect of sitagliptin on cardiovascular outcomes in 14,671 patients with type 2 diabetes and established cardiovascular disease [Green et al. 2015]. Sitagliptin was noninferior to placebo both for a 3-point [hazard ratio (HR) 0.99; 95% confidence interval (CI) 0.89–1.10) and a 4-point (HR 0.98; 95% CI 0.89–1.08) MACE composite outcome (including cardiovascular death, nonfatal myocardial infarction, nonfatal stroke and hospitalization for unstable angina). A conclusive answer about the cardiovascular safety of linagliptin is still pending the completion of two relevant long-term trials [ClinicalTrials.gov identifier: NCT01897532] [Marx et al. 2015].

Similarly to individual trial findings, a meta-analysis that estimated the effect of DPP-4 inhibitors on cardiovascular outcomes based on data from 50 randomized controlled trials with a minimum follow up of 24 weeks concluded that cardiovascular mortality [relative risk (RR) 0.97; 95% CI 0.85–1.11) and incidence of acute coronary syndrome (RR 0.97; 95% CI 0.87–1.08) or stroke (RR 0.98; 95% CI 0.81–1.18) did not differ between DPP-4 inhibitors and comparators [Wu et al. 2014]. Likewise, in a meta-analysis of 94 trials, treatment with DPP-4 inhibitors did not affect cardiovascular mortality and incidence of stroke [Savarese et al. 2015]. Interestingly though, the risk of myocardial infarction was reduced with DPP-4 inhibitors in the short, but not in the long term (duration of treatment ⩾29 weeks) [Savarese et al. 2015]. Finally, updated meta-analyses of individual DPP-4 inhibitors also support that neither vildagliptin (RR 0.82; 95% CI 0.61–1.11) [McInnes et al. 2015] nor linagliptin (HR 0.78; 95% CI 0.55–1.12) [Rosenstock et al. 2015] are associated with an increased risk for a composite outcome of adjudicated MACE relative to comparators.

Despite the reassuring evidence from randomized controlled trials and meta-analyses about the overall cardiovascular profile of DPP-4 inhibitors, safety concerns have been raised regarding a potential risk for hospitalization for heart failure, as demonstrated for saxagliptin in the SAVOR-TIMI-53 study [Scirica et al. 2013]. Following this unexpected finding of a 27% increased RR of hospitalization for heart failure in patients assigned to saxagliptin, the US Food and Drug Administration (FDA) requested additional data to investigate a possible association between use of saxagliptin and heart failure [FDA, 2014]. Upon further examination of clinical trial data, the manufacturer concluded that the observed risk was more prominent in patients with chronic kidney disease, previous heart failure and elevated levels of natriuretic peptides [Scirica et al. 2014]. Subsequently, in April 2015, the FDA Advisory Committee almost unanimously concluded that saxagliptin has an acceptable cardiovascular risk profile, but recommended the addition of new safety information to the product’s labeling [AstraZeneca, 2015]. Unlike saxagliptin, rates of hospital admission for heart failure for sitagliptin (HR 1.00; 95% 0.83–1.20) [Green et al. 2015] and alogliptin (HR 1.07; 95% 0.79–1.46) [Zannad et al. 2015] were similar to placebo. In addition, treatment with alogliptin had a neutral effect on a composite outcome of hospitalization due to heart failure and cardiovascular death (HR 1.00; 95% 0.82–1.21) [Zannad et al. 2015].

Furthermore, recent meta-analyses of individual DPP-4 inhibitors do not associate either vildagliptin or linagliptin with increased risk for heart failure. In a meta-analysis of 40 trials with vildagliptin, incidence of adjudicated events of heart failure requiring hospitalization or new onset of heart failure did not differ between vildagliptin and comparator (RR 1.08; 95% 0.60–1.70) [McInnes et al. 2015]. Similarly, in a meta-analysis of 19, mainly placebo-controlled, studies, the risk for hospitalization for congestive heart failure was similar for linagliptin and the control group (HR 1.04; 95% CI 0.43–2.47) [Rosenstock et al. 2015]. Conversely, meta-analyses assessing all DPP-4 inhibitors as a drug class suggest that they are indeed associated with an increased risk for acute heart failure [Monami et al. 2014; Udell et al. 2015], heart failure hospitalization [Wu et al. 2014] or new heart failure onset [Savarese et al. 2015]. However, it should be noted that these findings are mainly driven by the results of the SAVOR-TIMI-53 and EXAMINE trials. Indeed, in the meta-analysis by Savarese and colleagues, a sensitivity analysis according to the type of DPP-4 inhibitor demonstrated an elevated risk of new heart failure onset only with saxagliptin (RR 1.20; 95% 1.01–1.41 ) and not with sitagliptin, vildagliptin, linagliptin or alogliptin [Savarese et al. 2015].

In conclusion, DPP-4 inhibitors do not seem to have any effect on major adverse cardiovascular outcomes and risk for heart failure, apart from saxagliptin which was associated with an increased risk for hospitalization for heart failure, predominantly in patients with chronic kidney disease, pre-existing heart failure or elevated baseline levels of natriuretic peptides. This neutral effect has been considered sufficient for the marketing approval of glucose lowering therapies by the FDA and the European Medicines Agency (EMA). However, publication of the EMPA-REG OUTCOME trial results for empagliflozin, a novel sodium glucose cotransporter 2 inhibitor, which demonstrated a highly beneficial effect on overall mortality and cardiovascular outcomes [Zinman et al. 2015], further undermines the clinical importance of data for surrogate outcomes and supports the need for use of hard, patient-important cardiovascular outcomes to guide therapeutic decision-making in diabetes.