Follow the LEADER—Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results Trial

In LEADER, the primary outcome was a composite of three major adverse cardiovascular (CV) events (three-point MACE), defined as first occurrence of death from CV causes, non-fatal myocardial infarction (MI), or non-fatal stroke. Table 1 summarizes the findings of this study, which revealed a significant 13% reduction in the primary outcome, a significant 22% fall in CV death, and a numerical, but statistically non-significant lowering of risk of non-fatal MI and non-fatal stroke [9].

Death from CV causes was reduced in participants taking liraglutide in LEADER (hazard ratio (HR) 0.8). A similar lowering of all-cause mortality was noticed in this group (HR 0.8 5). As 219 out of 381 deaths in the liraglutide arm (57.5%) and 278 out of 447 deaths in the placebo arm (62.2%) occurred due to CV causes, the major driver of improvement in all-cause death seemed to be the CV benefit of liraglutide [9].

All types of vascular disease were attenuated with liraglutide, including MI, stroke and heart failure, even though statistical significance could not be demonstrated for individual endpoints. It is certain, however, that the benefits of the drug extended to all vascular beds and to the myocardium as well.

It is self-evident that various CVO trial results be compared with each other (Table 2). However, it must be noted that CVO trials are not head-to-head trials of two molecules [barring a few exceptions such as DEVOTE (NCT01959529), CAROLINA (NCT01243424) and TOSCA. IT (NCT00700856)] [10‐12]. The results of these trials are influenced by various factors in study design such as inclusion criteria, baseline medication use, quality of ‘standard of care’, duration of follow-up, and statistical plan. This implies that while LEADER and EMPA-REG OUTCOME (NCT01131676) can be discussed together, their results cannot be compared numerically. While EMPA-REG OUTCOME has been able to demonstrate CV benefits of empagliflozin, there are subtle differences in its results, as compared to those of liraglutide. Liraglutide has a gradually developing positive effect on all aspects of CV disease and mortality, while empagliflozin has a relatively faster effect on CV mortality and heart failure, but a numerically negative effect on stroke [7]. The number needed to treat (NNT) to prevent coronary events or deaths is much lower with empagliflozin than with liraglutide (Table 2). This raises questions about the economic viability of extrapolating the results of these trials to routine clinical practice. ELIXA (NCT01147250), which studied lixisenatide, a shorter acting glucagon-like peptide-1 receptor agonist (GLP-1 RA), has demonstrated CV safety of the molecule, but could not find significant CV benefit of its use [8].

The authors of LEADER suggest that liraglutide modifies progression of atherosclerotic disease [9]. We highlight another facet of this drug which can be used to explain its beneficial action. Calorie restriction has long been known to improve longevity in both animal and human species [13]. Calorie restriction mimicry, using drugs designed to act on the adenosine monophosphate kinase (AMPK) pathway, in a manner similar to that of calorie restriction, has also been tried to achieve similar benefits. Perhaps, the most well-known calorie restriction mimetic is metformin, which has also shown improved outcomes in the UKPDS (United Kingdom Prospective Diabetes Study) trial (ISRCTN 75451837) [14].

Liraglutide acts as a direct calorie restrictor by reducing appetite, and also as a calorie restriction mimetic by modifying AMPK action. Its actions on the hypothalamus, gastrointestinal tract and pancreas are designed to mimic a state of calorie restriction [13, 15, 16]. This facet of liraglutide’s mechanism of action needs detailed study.

Another facet of the LEADER study that deserves close attention is the incidence of cholelithiasis. LEADER reports exceptional safety and tolerance with liraglutide, finding no increase in the risk of pancreatitis or neoplasms. There is, however, a higher risk of cholelithiasis in persons treated with liraglutide [9]. Risk factors for cholelithiasis that operate in the general population also tend to occur in participants of CVO trials. These include older age, female gender, heavy body weight and hypertriglyceridemia [17]. A higher incidence of gallstones has also been reported with exenatide [18].

While GLP-1-specific mechanisms have been postulated for this, such as reduced cholecystokinin, slower biliary tract motility and increased stasis of bile, it must be noted that cholelithiasis is a risk inherent to every weight-lowering therapy, including very low-calorie diet [19]. Gallstones may develop as soon as 4 weeks after initiation of weight-reducing diet. Hypotheses that have been suggested include an increase in cholesterol output, due to increased mobilization of tissue cholesterol to bile; increased gall bladder secretion of mucin and calcium; increased presence of prostaglandin E2 and arachidonic acid in bile; super-saturation of cholesterol in bile, due to reduced bile salt secretions; gall bladder stasis due to reduced stimulation by low-fat diet [20].

Liraglutide provides comprehensive metabolic modulation, including weight, systolic blood pressure and lipid control. Similar results are seen with empagliflozin as well. Once again, subtle differences are visible to the discerning eye. Empagliflozin has a diuretic effect, and reduces both systolic and diastolic blood pressure, without causing reflex tachycardia. However, its effects on lipid profile are not significant. Direct hemodynamic effects on the heart and vasculature are also suggested, as it has a more marked benefit on heart failure than on coronary events [21]. Liraglutide has a greater effect on body weight, systolic blood pressure and lipids, and seems to modify the progression of the basic atherosclerotic process in diabetes.

It may be that liraglutide prevents the actual CV event (primary prevention) while empagliflozin reduces mortality after the event (secondary prevention) [22]. However, though the numerical data from LEADER are in favor of this statement, more research is needed to support this claim.

The ‘Robin Hood effect’ has been suggested as a moniker for the metabolic effects of sodium glucose cotransporter-2 (SGLT2) inhibitors, which shift energy production from carbohydrate metabolism to lipid metabolism [23]. In a similar (Anglo-centric) vein, the term Lion Heart effect may be used to describe liraglutide. Richard the Lionheart was an iconic twelfth century English king who was famous for his military prowess. Richard the Lionheart and Robin Hood are shown as contemporaries in various quasi-historical accounts. The results of LEADER suggest a Ricardian property of liraglutide, which provides vascular safety and benefit, and modifies the natural history of diabetes in a favorable manner.