Summary | Practical management of cardiovascular risk: Lessons from latest diabetes trials
The two options to lower glucose based on incretins are DPP-4 inhibitors and GLP-1RAs. Effects of incretin-based glucose-lowering therapies are increased insulin secretion, decreased glucagon secretion, improved beta-cell mass in time and enhanced insulin sensitivity. Decreased gastric emptying and improved satiety are also effects of incretins, especially with GLP-1RAs (32). GLP-1RAs are administered by injections. These agents mimic the effect of natural GLP-1, which is released during eating. GLP-1RAs reduce HbA1c by ~1% or more and result in weight loss of 2-3 kg. There is a low risk of hypoglycemia when GLP-1RAs are used with metformin, and a reduced risk of hypoglycemia in combination with insulin.
The mechanisms of action of GLP-1RAs are broad. GLP-1RAs interfere with many organ systems and it is not known how exactly GLP-1RAs exert CV benefits. Many mechanistic studies are ongoing and to date, it has been suggested that the heart, kidney, vessels and intestines are all important targets of this drug class (33). There are two types of GLP-1RAs; short-acting and long-acting (34, 35). The short-acting is exendin-4 based and the long acting type exendin-4 based or a GLP-1 human analog. Lixisenatide and exenatide are based on exendin-4, which was derived from the saliva of the lizard. First, lixisenatide was tested in ~6000 T2DM patients with a recent acute coronary syndrome (ACS) (36). They received lixisenatide or placebo on top of standard care, follow-up was ~2 years and the impact on CV mortality and morbidity was studied. There was no difference in HbA1c and 4-point MACE, a composite of CV death, non-fatal MI, stroke, and hospitalization for ACS between the two groups at the end of the study. Overall, there was no benefit for CV outcomes with lixisenatide.
Exenatide was evaluated in a study with almost 1500 T2DM patients with or without CVD (37). Three quarters of patients had CVD, while the remaining group were at high CV risk. Patients received either exenatide or placebo on top of standard, additional care. Follow-up was 3.2 years and the primary endpoint was a composite of CV mortality, non-fatal MI and non-fatal stroke. At the end of the study, a non-significant change in HbA1c and CV outcomes was observed with exenatide. These studies were designed to show non-inferiority and lixisenatide and exenatide were demonstrated to be safe, but they did not result in CV benefit.
Liraglutide and semaglutide are human analogs of GLP-1. Liraglutide was also tested in the LEADER trial in a non-inferiority design, including patients with a history of CVD or high CV risk (28). On top of standard care treatment, they received liraglutide or placebo. Liraglutide resulted in a 13% reduction of the primary endpoint 3-point MACE. This meant that there was a drug now that was not only safe, but which also reduced the number of CV events. A similar result was observed with semaglutide with a reduction in CV outcomes for the semaglutide group compared to placebo and also a reduction in HbA1c compared to the placebo group (7.3% vs 8.3%) (38).
The patients in the LEADER trial were ~64 years old and the majority was male, which is not surprising as more men than women suffer from CVD. The patients had a diabetes duration of ~13 years, had somewhat elevated HbA1c levels, were overweight or obese based on BMI, a well-controlled BP and 18% had a history of HF. The majority had a prior CV event. Background therapy was a combination of metformin, sulfonylurea, insulin or other glucose-lowering drugs. Other drugs, such as BP-lowering diuretics and lipid-lowering drugs, were also commonly used. Thus, these were well-treated patients, as is evident by the treatment goals set by the investigators: HbA1c ≤7.0, BP target 130/80 mmHg, LDL-c <1.8 mmol/L and treated with antiplatelet therapy after CV event.
Reduction in MACE with liraglutide was observed in all subgroups, and it seems as though there was a slightly increased benefit in younger patients and in those with CVD, although these data have to be interpreted with caution as it was a post-hoc analysis. The important message here is that the beneficial CV impact was consistent in all subgroups. Looking at the individual endpoints, a reduction was seen in total mortality and CV death with liraglutide, but not in non-fatal MI, non-fatal stroke, and HF hospitalization. In the past, incretins have been associated with an increase in HF, but there was no sign of harm with liraglutide. Co-treatment with insulin did not affect the CV benefit observed with liraglutide. There was also a reduction of microvascular events, a composite of renal and retinal outcomes, which was primarily driven by an improvement of 22% for renal function over time. An immediate decrease in HbA1c is observed after initiation with liraglutide and a small significant difference was seen compared to the placebo group at the end of the study, despite the fact that more glucose-lowering agents were added to the treatment regimen over time in the placebo group. Other effects of liraglutide were weight loss (2.3 kg), SBP reduction (1.2 mmHg), a small increase in HDL-c, a small decrease in LDL-c, and a small increase of 3 beats per min in heart rate (HR). It is known that GLP-1RAs can increase HR somewhat, but this small elevation of HR does not counteract the beneficial effects of liraglutide.
Severe hypoglycemia was significantly less common in the liraglutide group compared to the placebo group and there was no difference in CV outcomes in patients with or without hypoglycemic episodes. Overall, the LEADER trial showed an absolute reduction of 1.9% in 3-point MACE, corresponding to a relative risk reduction of 13%, which was impressive considering that patients were already well-treated. CV death was reduced by 22% and all-cause mortality by 15%.
The patient group in the SUSTAIN-6 trial was similar to that in the LEADER, although 61% of patients in the SUSTAIN-6 trial had CVD compared to 81% in the LEADER trial. The SUSTAIN-6 trial was shorter in duration, but showed a clear significant impact on 3-point MACE, a composite of non-fatal MI, non-fatal stroke, and CV mortality. Considering that most benefit for individual endpoints was observed for MI, stroke and need for revascularization, it is thought that semaglutide may result in reduced progression of atherosclerosis and/or stabilization of plaques. Total mortality was not reduced by semaglutide, but CV mortality was. Glycemic control was more improved in the SUSTAIN-6 trial by semaglutide than by placebo in the LEADER trial and this was also observed for body weight, with a small dose effect.
A significant increase in retinopathy was observed in the semaglutide group in the SUSTAIN-6 trial and a non-significant increase with liraglutide in the LEADER trial. This has to be monitored in the future and more studies have to be done to examine the increased risk of retinopathy with GLP-1RAs. A possible explanation may be that a fast reduction in glucose may provoke already existing retinopathy.
Overall, treatment with semaglutide in the SUSTAIN-6 trail showed an absolute risk reduction in 3-point MACE of 2.3% and a relative risk reduction of 26%. This reduction was larger than that observed in the LEADER trial and achieved in a shorter period. CV mortality and all-cause mortality were not reduced, but perhaps this is due to the shorter duration of the trial. In time, less stroke and MI should lead to a reduction in mortality. A recent meta-analysis of all studies on GLP-1RAs showed there was no sign of hypoglycemia, pancreatitis and pancreatic cancer with the use of these agents (39).
There are several possible explanations for the differences in CV outcomes with GLP-1RAs. One of them is that the study populations in the trials were different. The ELIXA trial enrolled patients with T2DM and a recent ACS; these patients were perhaps more unstable than in other trials. Prior CVD was less evident in the ongoing REWIND trial (40) and in the EXSCEL trial (37) compared to other trials. The impact on CV benefit seems to be greater in patients with established CVD. Therefore, the results of the REWIND study are highly anticipated. Differences in use of non-study medications and differences in glycemic control may also affect the observed CV benefits. And, exendin-4 based GLP-1RAs have not shown reduction in CV events, while long-acting human analogs that mimic natural GLP-1 did.
The results of the REWIND trial, enrolling less ill patients than in the other GLP-1RAs trials, will be presented in June 2019. Other results to come are those of the HARMONY trial and the PIONEER-6 trial. Also, already available drugs will be studied, for example an oral semaglutide is currently being evaluated. Furthermore, people with impaired glucose tolerance or obesity, not necessarily inflicted by diabetes, are of interest because GLP-1RAs can result in weight reduction and prevention of metabolic syndrome. The results of the trials have been translated into a substantial number of local and international guidelines. The American standardized medical care in diabetes recommends that patients with T2DM and established atherosclerotic CVD should receive an agent proven to reduce major adverse CV events and CV mortality (41).
This is a summary of the presentation given by prof. Rydén, during the PACE symposium entitled 'Preventing Cardiovascular Disease in Patients with T2DM – How to apply novel outcome data with GLP-1RA to clinical practice', held during EuroPrevent in Ljubljana, Slovenia, on May 3, 2018.