Limited Extent of Pleiotropic Effects Mediated by Dipeptidyl Peptidase-4 Inhibitors in Patients With Diabetes Mellitus
Article ID: CJ-18-0694
Details
Article ID: CJ-18-0694
Diabetes mellitus (DM) is a strong risk factor for cardiovascular (CV) diseases, including diabetic macroangiopathy and microangiopathy, both of which lead to severe reductions in lifespan, activities of daily living, and quality of life. Definitely, the number of diabetic patients is increasing worldwide. Thus, the management of diabetes is critical and most countries have their own guidelines for diabetes management. Among the oral antidiabetics agents, such as biguanide, sulfonylurea, glinide, thiazolidinedione, α-glucosidase inhibitors (αGI), dipeptidyl peptidase-4 inhibitors (DPP4i), and sodium-glucose transporter 2 inhibitors (SGLT2i), etc., biguanide, αGI, DPP4i, and SGLT2i, which are unlikely to evoke hypoglycemia, are the most favorable choices because hypoglycemia could rather increase the number of CV events. Recently, DPP4i were shown to be protective against CV injury beyond the glucose-lowering effect, which is known as an agent-mediated pleiotropic effect.1,2 Nevertheless, it remains unclear whether DPP4i have pleiotropic effects clinically.
Article p ????
Statins are noted agents with pleiotropic effects and have the mechanisms to improve endothelial function via eNOS activation and inhibit cardiac hypertrophy via RhoA/ROCK inhibition. In clinical practice as well, statins have improved endothelial function in a meta-analysis3 and left ventricular (LV) ejection fraction in patients with heart failure.4 Recently, Matsubara et al demonstrated that 6-month treatment with sitagliptin (a DPP4i) improved endothelial function evaluated by reactive hyperemia peripheral artery tonometry (RH-PAT), a microvasculature assessment,5 in coronary artery disease (CAD) and uncontrolled type 2 DM (T2DM) patients compared with conventional therapy (baseline HbA1c: sitagliptin 7.3±0.8%, conventional 7.5±1.2%), despite similar glucose-lowering effects.1 Likewise, it has been shown that 6-month treatment with sitagliptin improved endothelial function evaluated by flow-mediated dilation (FMD) of the brachial artery, a conduit artery assessment,5 in T2DM patients compared with voglibose, an αGI agent (HbA1c: sitagliptin 7.04±0.56%, voglibose 6.94±0.45%).6 On the other hand, no improvement in endothelial function by RH-PAT with 12-week treatment with sitagliptin compared with placebo was reported in patients with acute coronary syndrome with newly detected impaired glucose tolerance or T2DM (median HbA1c: sitagliptin 5.7%, placebo 5.9%).7
In this issue of the Journal, Moriwaki et al demonstrate that after 24-week treatment with sitagliptin or voglibose neither improved endothelial function by RH-PAT in CAD and controlled T2DM patients (HbA1c: sitagliptin 6.62±0.48%, voglibose 6.64±0.51%).8 Importantly, the key difference between the effective and non-effective studies is higher baseline HbA1c (i.e., it may be postulated that DPP4i could improve endothelial function in uncontrolled T2DM patients, although this hypothesis cannot fully explain the discrepancy of the results).
Concerning coronary flow reserve (CFR) evaluated by cardiac magnetic resonance imaging (MRI), 12-week treatment with alogliptin was shown to improve CFR compared with glimepiride, a sulfonylurea agent, in T2DM and known or suspected CAD patients (HbA1c: alogliptin 7.2±0.6%, glimepiride 6.9±0.4%).9 Nevertheless, the present study showed no CFR improvement with sitagliptin compared with voglibose.8 Interestingly, again the key difference between the effective and non-effective studies was higher baseline HbA1c. Regarding the chronic effect of DPP4i on LV diastolic function as well, inconsistent data have been reported. The beneficial effect of DPP4i was demonstrated in a 24-week treatment with sitagliptin that improved LV diastolic function compared with NPH insulin in T2DM patients (HbA1c: sitagliptin 8.0±0.6%, NIH insulin 8.1±0.7%).10 Fujiwara et al also demonstrated that treatment with DPP4i improved LV diastolic function in patients with acute myocardial infarction (AMI), but not in those without DPP4i during a follow-up of 7.4±2.5 months (HbA1c: DPP4i 8.21±1.49%, non-DPP4i 8.10±1.63%).11 Meanwhile, the present study revealed a non-beneficial effect of DPP4i on LV diastolic function compared with voglibose.8 Concordantly, Oe et al revealed a non-beneficial effect of 24-week treatment with sitagliptin compared with voglibose on LV diastolic function in T2DM patients (HbA1c: sitagliptin 7.04±0.56%, voglibose 6.94±0.45%).12
DPP4i-mediated pleiotropic effects on endothelial function, CFR, and LV diastolic function have been shown to be controversial. Indeed, previous randomized, double-blind clinical trials showed no additional beneficial effect of DPP4i on CV outcomes.13–15 Alternatively, the Savor-TIMI 53 trial showed a significant increase in the hospitalization rate for heart failure in patients with saxagliptin use compared with placebo.14 Also, the baseline HbA1c in the randomized clinical trials of DPP4i has been high rather than low.13–15 Collectively, it is still premature at the present stage to determine that DPP4i have pleiotropic effects clinically. As a possible postulation, DPP4i may improve endothelial function in uncontrolled T2DM patients (Figure), whereas their potential to improve endothelial function generally may be quite smaller than that of statins. Indeed, in the studies showing improvement in endothelial function, the ratios of statins usage were 50% in the sitagliptin group and 45.2% in the voglibose group,6 and 95% in the group with both sitagliptin and conventional therapy.1 In studies reporting no improvement in endothelial function, the ratios of statins usage were 100% in the sitagliptin group and 94% with placebo,7 and 100% with both sitagliptin and voglibose.8 Therefore, the combination of DPP4i with statins may mask DPP4-mediated pleiotropic effects on endothelial function.
Possible pleiotropic effects mediated by dipeptidyl peptidase-4 inhibitors (DPPi) in type 2 diabetic patients. LV, left ventricular.
In conclusion, whether DPP4i have pleiotropic effects beyond a glucose-lowering effect remains unclear at present. If it exists, the extent of the DPP4i-mediated pleiotropic effect may be limited (e.g., in uncontrolled T2DM). In order to reduce both CV events and deaths, it is crucial to not only lower blood glucose but also to avoid the use of agents that could evoke hypoglycemia. Taken together, the data suggest that DPP4i should be used with an expectation of lowering blood glucose and avoiding hypoglycemia, but not used with an expectation of pleiotropic effects until further studies definitely reveal DPP4i-mediated pleiotropic effects.
There are no conflicts of interest to report for any of the authors.
